BSS Network Doctor Formulas

BSS Network Doctor Formulas

DN98619493 © Nokia Corporation 1 (324)Issue 2-5 en Nokia Proprietary and Confidential


The information in this documentation is subject to change without notice and describes onlythe product defined in the introduction of this documentation. This documentation is intendedfor the use of Nokia's customers only for the purposes of the agreement under which thedocumentation is submitted, and no part of it may be reproduced or transmitted in any form ormeans without the prior written permission of Nokia. The documentation has been prepared tobe used by professional and properly trained personnel, and the customer assumes fullresponsibility when using it. Nokia welcomes customer comments as part of the process ofcontinuous development and improvement of the documentation.

The information or statements given in this documentation concerning the suitability, capacity,or performance of the mentioned hardware or software products cannot be considered bindingbut shall be defined in the agreement made between Nokia and the customer. However, Nokiahas made all reasonable efforts to ensure that the instructions contained in the documentationare adequate and free of material errors and omissions. Nokia will, if necessary, explain issueswhich may not be covered by the documentation.

Nokia's liability for any errors in the documentation is limited to the documentary correction oferrors. NOKIA WILL NOT BE RESPONSIBLE IN ANY EVENT FOR ERRORS IN THISDOCUMENTATION OR FOR ANY DAMAGES, INCIDENTAL OR CONSEQUENTIAL(INCLUDING MONETARY LOSSES), that might arise from the use of this documentation orthe information in it.

This documentation and the product it describes are considered protected by copyrightaccording to the applicable laws.

NOKIA logo is a registered trademark of Nokia Corporation.

Other product names mentioned in this documentation may be trademarks of their respectivecompanies, and they are mentioned for identification purposes only.

Copyright © Nokia Corporation 2004. All rights reserved.

2 (324) © Nokia Corporation DN98619493Nokia Proprietary and Confidential Issue 2-5en

Contents

Contents 3

List of tables 5

List of figures 6

1 About this manual 331.1 Summary of changes 331.2 What you need to know first 361.3 Where to find more 361.4 Typographic conventions 361.4.1 Text styles 371.5 Terms and concepts 371.5.1 Abbreviations 381.5.2 Terms 39

2 BSS counter formulas 412.1 Additional GPRS channels (ach) 412.2 Multislot (msl) 432.3 TBF (tbf) 472.4 LLC (llc) 582.5 RLC (rlc) 592.6 Frame relay (frl) 782.7 HSCSD (hsd) 822.8 Dynamic Abis Pool (dap) 832.9 Random access (rach) 842.10 SDCCH drop failures (sd) 862.10.1 SDCCH drop counters 872.10.2 Problems with the SDCCH drop counters 892.11 SDCCH drop ratio (sdr) 902.12 Setup success ratio (cssr) 922.13 TCH drop failures 922.13.1 TCH drop call counters 922.13.2 Drop call ratio 952.13.3 Drop-out ratio 952.13.4 Problems with the drop call counters 962.14 Drop call failures (dcf) 962.15 TCH drop call % (dcr) 972.16 Adaptive Multirate (amr) 1142.17 Position based services (pbs) 1152.18 Handover (ho) 1172.19 Handover failure % (hfr) 1302.20 Handover success % (hsr) 1572.21 Handover failures (hof) 1632.22 Interference (itf) 1672.23 Congestion (cngt) 1692.24 Queuing (que) 1702.25 Blocking (blck) 173



2.26 Traffic (trf) 1842.27 Traffic directions 2482.27.1 Mobile originated calls (moc) 2482.27.2 Mobile terminated calls (mtc) 2502.28 Paging (pgn) 2522.29 Short message service (sms) 2552.30 Directed retry (dr) 2572.31 Availability (ava) 2592.32 Unavailability (uav) 2772.33 Location updates (lu) 2822.34 LU success % (lsr) 2832.35 Emergency call (ec) 2832.36 Emergency call success % (ecs) 2842.37 Call re-establishment (re) 2842.38 Call re-establishment success % (res) 2842.39 Quality 2852.39.1 Downlink quality (dlq) 2852.39.2 Uplink quality (ulq) 2892.40 Downlink and uplink level 2942.40.1 Downlink level (dll) 2942.40.2 Uplink level (ull) 2942.41 Power (pwr) 2952.42 Level (lev) 2952.43 Distance (dis) 2972.44 Link balance, power, level (lb) 2982.45 Call success (csf) 3012.46 Configuration (cnf) 319

3 Missing Counters 3213.1 XX1 3213.2 XX2 3213.3 XX3 3213.4 XX4 322

Index 323


List of tables

Table 1. Text styles in this document 37

Table 2. Abbreviations 38

Table 3. Terms used in this document 39

Table 4. SDCCH Drop Counters 88

Table 5. TCH drop call counters 93



List of figures

Figure 1. Additional GPRS channel use, S9PS (ach_1) 42

Figure 2. Average additional GPRS channel hold time, S9PS (ach_2) 42

Figure 3. Additional GPRS channels seized, S9PS (ach_3) 42

Figure 4. Total additional GPRS channel hold time, S9PS (ach_4) 43

Figure 5. Distribution of UL multislot requests, S9PS (msl_1) 43

Figure 6. Distribution of DL multislot requests, S9PS (msl_2) 43

Figure 7. Distribution of UL multislot allocations, S9PS (msl_3) 44

Figure 8. Distribution of DL multislot allocations, S9PS (msl_4) 44

Figure 9. Ratio of unreserved GPRS UL TSL requests, S9PS (msl_5) 44

Figure 10. Ratio of unreserved GPRS DL TSL requests, S9PS (msl_6) 44

Figure 11. UL multislot allocations, S9PS (msl_9) 45

Figure 12. DL multislot allocations, S9PS (msl_10) 45

Figure 13. Average number of allocated timeslots, UL S9PS (msl_11) 45

Figure 14. Average number of allocated timeslots, DL S9PS (msl_13) 45

Figure 15. Average number of requested UL timeslots, S9PS (msl_13) 46

Figure 16. Average number of requested DL timeslots, S9PS (msl_14) 46

Figure 17. UL multislot allocation %, S9PS (msl_15a) 46

Figure 18. DL multislot allocation %, S9PS (msl_16a) 47

Figure 19. UL multislot requests, S9PS (msl_17) 47

Figure 20. DL multislot requests, S9PS (msl_18) 47

Figure 21. Average number of LLC blocks per UL TBF, S9PS (tbf_3) 48

Figure 22. Average number of LLC blocks per DL TBF, S9PS (tbf_4) 48

Figure 23. Average UL TBF duration, S9PS (tbf_5) 48

Figure 24. Average UL TBF duration, S9PS (tbf_5a) 49

Figure 25. Average DL TBF duration, S9PS (tbf_6a) 49

Figure 26. Average UL TBF duration, unack mode, S9PS (tbf_7) 49

Figure 27. Average DL TBF duration, unack mode, S9PS (tbf_8) 49

Figure 28. UL mlslot allocation blocking, S9PS (tbf_15) 50

Figure 29. DL mlslot allocation blocking, S9PS (tbf_16) 50


Figure 30. UL TBF releases due to CS traffic %, S9PS (tbf_19) 51

Figure 31. DL TBF releases due to CS traffic %, S9PS (tbf_20) 51

Figure 32. UL drops per 10 Kbytes, MS lost, S9PS (tbf_27a) 51

Figure 33. UL drops per 10 Kbytes, MS lost, S9PS (tbf_27b) 52

Figure 34. UL drops per 10 Kbytes, MS lost, S10.5PS (tbf_27c) 53

Figure 35. DL drops per 10 Kbytes, MS lost, S9PS (tbf_28a) 53

Figure 36. DL drops per 10 Kbytes, MS lost, S9PS (tbf_28b) 53

Figure 37. DL drops per 10 Kbytes, MS lost, S10.5PS (tbf_28c) 54

Figure 38. UL TBF reallocation failure ratio, S9PS (tbf_29) 54

Figure 39. DL TBF reallocation failure ratio, S9PS (tbf_30) 55

Figure 40. UL TBF reallocation attempts, S9PS (tbf_31) 55

Figure 41. DL TBF reallocation attempts, S9PS (tbf_32) 55

Figure 42. TBF success % S9PS (tbf_34) 56

Figure 43. UL TBF releases due to flush %, S9PS (tbf_35) 56

Figure 44. DL TBF releases due to flush %, S9PS (tbf_36) 56

Figure 45. Average UL TBF per timeslot, S9PS (tbf_37b) 56

Figure 46. Average UL TBFper timeslot, S9PS (tbf_37c) 57

Figure 47. Average DL TBF per timeslot, S9PS (tbf_38b) 57

Figure 48. Average DL TBFper timeslot, S9PS (tbf_38c) 57

Figure 49. UL GPRS TBF establishments, S10.5PS (tbf_41) 58

Figure 50. DL GPRS TBF establishments, S10.5PS (tbf_42) 58

Figure 51. Normal TBF release ratio DL, to UL, S10.5PS (tbf_44) 58

Figure 52. Expired LLC frames % DL, S9PS (llc_1) 59

Figure 53. Discarded UL LLC frames, NSE unavailability %, S9PS (llc_2) 59

Figure 54. Ack. CS1 RLC blocks UL, S9PS (rlc_1) 59

Figure 55. Ack. CS1 RLC blocks DL, S9PS (rlc_2) 59

Figure 56. Ack. CS1 RLC DL block error rate, S9PS (rlc_3a) 60

Figure 57. Unack. CS1 RLC UL block error rate, S9PS (rlc_4a) 60

Figure 58. Ack. CS1 RLC UL block error rate), S9PS (rlc_5a) 60

Figure 59. UL CS1 RLC data share, S9PS (rlc_6a) 61

Figure 60. UL CS1 ack RLC data share, S9PS (rlc_6b) 61



Figure 61. UL CS1 unack RLC data share, S9PS (rlc_6c) 61

Figure 62. UL CS2 RLC data share, S9PS (rlc_7a) 62

Figure 63. DL CS1 RLC data share, S9PS (rlc_8a) 62

Figure 64. DL CS1 ack RLC data share, S9PS (rlc_8b) 62

Figure 65. DL CS1 unack RLC data share, S9PS (rlc_8c) 63

Figure 66. DL CS2 RLC data share, S9PS (rlc_9a) 63

Figure 67. UL CS1 RLC block error rate, S9PS (rlc_10a) 64

Figure 68. UL CS1 RLC block error rate, S9PS (rlc_10b) 65

Figure 69. UL CS1 ACK RLC block error rate, S9PS (rlc_10c) 66

Figure 70. UL CS2 ARLC block error rate, S9PS (rlc_11a) 66

Figure 71. UL CS2 RLC block error rate, S9PS (rlc_11b) 67

Figure 72. UL CS2 ACK RLC block error rate, S9PS (rlc_11c) 67

Figure 73. UL CS2 ACK RLC block error rate, S10.5PS(rlc_11d) 68

Figure 74. DL CS1 RLC block error rate, S9PS (rlc_12) 68

Figure 75. DL CS1 ACK RLC block error rate, S9PS (rlc_12a) 68

Figure 76. DL CS2 RLC block error rate, S9PS (rlc_13) 68

Figure 77. UL RLC blocks, S9PS (rlc_14) 69

Figure 78. DL RLC blocks, S9PS (rlc_15) 69

Figure 79. UL ACK EGPRS block error ratio S10.5PS (rlc_18) 69

Figure 80. DL ACK EGPRS block error ratio S10.5PS (rlc_19) 70

Figure 81. UL ACK EGPRS block error ratio MCS-n, S10.5PS (rlc_20) 70

Figure 82. DL ACK EGPRS block error ratio MCS-n, S10.5PS (rlc_21) 70

Figure 83. UL ACK RLC data share MCS-n, S10.5PS (rlc_22) 71

Figure 84. UL UNACK RLC data share MCS-n, S10.5PS (rlc_23) 71

Figure 85. DL ACK RLC data share MCS-n, S10.5PS (rlc_24) 71

Figure 86. DL UNACK RLC data share MCS-n, S10.5PS (rlc_25) 72

Figure 87. GMSK RLC data block share, S10.5PS (rlc_39) 72

Figure 88. GMSK RLC data share, S10.5PS (rlc_41) 73

Figure 89. GPRS UL ACK RLC data share, S10.5PS (rlc_42) 73

Figure 90. GPRS UL UNACK RLC data share, S10.5PS (rlc_43) 74

Figure 91. GPRS DL ACK RLC data share, S10.5PS (rlc_44) 74


Figure 92. GPRS DL UNACK RLC data share, S10.5PS (rlc_45) 75

Figure 93. EGPRS UL ACK RLC data share, S10.5PS (rlc_46) 75

Figure 94. EGPRS UL UNACK RLC data share, S10.5PS (rlc_47) 76

Figure 95. EGPRS DL ACK RLC data share, S10.5PS (rlc_48) 77

Figure 96. EGPRS DL UNACK RLC data share, S10.5PS (rlc_49) 77

Figure 97. Kbytes in sent frames, S9PS (frl_1) 78

Figure 98. Kbytes in received frames, S9PS (frl_2) 78

Figure 99. ‘Wrong check seq.’ errors per Mbyte, S9PS (frl_3) 79

Figure 100. ‘Other’ errors per Mbyte, S9PS (frl_4) 79

Figure 101. Bytes in discarded sent frames, S9PS (frl_5) 79

Figure 102. Bytes in discarded received frames, S9PS (frl_6) 79

Figure 103. Maximum sent load %, S9PS (frl_7) 80

Figure 104. Maximum received load %, S9PS (frl_8) 80

Figure 105. Sent frames, S9PS (frl_9) 81

Figure 106. Received frames, S9PS (frl_10) 81

Figure 107. Discarded sent frames, S9PS (frl_11) 81

Figure 108. Discarded received frames, S9PS (frl_12) 81

Figure 109. Discarded bytes, UL NS-VC congestion S9PS (frl_13a) 82

Figure 110. Throughput ratio, S7HS (hsd_15) 82

Figure 111. Bps traffic share, S7HS (hsd_49) 82

Figure 112. Bps traffic share, S7HS (hsd_50) 83

Figure 113. Average usage of DL Dynamic Abis Pool, S10.5PS (dap_1) 83

Figure 114. Average usage of UL Dynamic Abis Pool, S10.5PS (dap_2) 83

Figure 115. Average Available PCM Sub-TSL, S10.5PS (dap_3) 84

Figure 116. Average RACH slot, S1 (rach_1) 84

Figure 117. Peak RACH load, average, S1 (rach_2) 84

Figure 118. Peak RACH load %, S1 (rach_3) 85

Figure 119. Average RACH load %, S1 (rach_4) 85

Figure 120. Average RACH busy, S1 (rach_5) 85

Figure 121. RACH rejected due to illegal establishment, S5 (rach_6) 86

Figure 122. Total RACH rejection ratio, S7 (rach_7) 86



Figure 123. Ghosts detected on SDCCH and other failures, S1 (sd_1) 87

Figure 124. Ghosts detected on SDCCH and other failures, S1 (sd_1a) 87

Figure 125. SDCCH drop %, S3 (sdr_1a) 91

Figure 126. SDCCH drop %, abis fail excluded, S3 (sdr_2) 91

Figure 127. Illegal establishment cause % (sdr_3b) 91

Figure 128. SDCCH drop ratio without timer T3101 expiry % (sdr_4) 92

Figure 129. SDCCH, TCH setup success %, S4 (cssr_2) 92

Figure 130. TCH drop calls in HO, S2 (dcf_2) 96

Figure 131. TCH drop calls in BSC outgoing HO, S3 (dcf_3) 96

Figure 132. TCH drop calls in intra-cell HO, S3 (dcf_4) 97

Figure 133. TCH drop calls in intra BSC HO, S3 (dcf_6) 97

Figure 134. Drop calls in BSC incoming HO, S3 (dcf_7) 97

Figure 135. TCH drop calls in HO, S7 (dcf_11) 97

Figure 136. TCH drop call %, area, S3 (dcr_3c) 98

Figure 137. TCH drop call %, area, real, after re-establishment S3 (dcr_3f) 100

Figure 138. TCH drop call %, area, real, before re-establishment, S3 (dcr_3g) 101

Figure 139. TCH drop call %, area, real, after re-establishment, S7 (dcr_3h) 101

Figure 140. TCH drop call %, area, real, before re-establishment, S3 (dcr_3i) 102

Figure 141. TCH drop call %, area, real, after re-establishment, S7 (dcr_3j) 104

Figure 142. TCH drop-out %, BTS level, before call re-establishment, S3(dcr_4c) 105

Figure 143. TCH drop-out %, BTS level, before call re-establishment, S3(dcr_4d) 105

Figure 144. TCH drop-out %, BTS level, before call re-establishment, S7(dcr_4e) 106

Figure 145. TCH drop-out %, BTS level, before call re-establishment, S7(dcr_4f) 107

Figure 146. TCH drop call (dropped conversation) %, BSC level, S4 (dcr_5) 107

Figure 147. . . TCH dropped conversation %, area, re-establishment considered, S7(dcr_5b) 108

Figure 148. . .TCH drop call %, after TCH assignment, without RE, area level, S10.5(dcr_8c) 109

Figure 149.TCH drop call %, after TCH assignment, with RE, area level, S10.5(d


cr_8e) 109

Figure 150. Drops per erlang, before re-establishment, S4 (dcr_10) 110

Figure 151. Drops per erlang, after re-establishment, S4 (dcr_10a) 110

Figure 152. Drops per erlang, after re-establishment, S7 (dcr_10b) 111

Figure 153. TCH Rf loss in HO - ratio, IUO (dcr_14) 111

Figure 154. Transcoder failure ratio, FR (dcr_16) 111

Figure 155. Transcoder failure ratio, EFR (dcr_17) 112

Figure 156. Transcoder failure ratio, HR (dcr_18) 112

Figure 157. Transcoder failure ratio, AMR FR (dcr_19) 112

Figure 158. Transcoder failure ratio, AMR HR (dcr_20) 112

Figure 159. Transcoder failure ratio (dcr_21) 113

Figure 160. Call failures share of transcoder failures (dcr_22) 113

Figure 161. HO target share of transcoder failures (dcr_23) 113

Figure 162. HO source share of transcoder failures (dcr_24) 113

Figure 163. Transcoder failures (dcr_25) 114

Figure 164. Codec set upgrade attempts, S10 (amr_1) 114

Figure 165. Codec set downgrade attempts, S10 (amr_2) 114

Figure 166. Codec set upgrade failure ratio, S10 (amr_3) 114

Figure 167. Codec set downgrade failure ratio, S10 (amr_4) 115

Figure 168. Failure ratio of location calculations for external LCS clients, S10(pbs_1a) 115

Figure 169. Failure ratio of location calculations for emergency calls, S10(pbs_2a) 115

Figure 170. Failure ratio of E-OTD location calculations, S10 (pbs_3) 115

Figure 171. Failure ratio of E-OTD location calculations, S10 (pbs_3a) 116

Figure 172. Failure ratio of location calculations for MS, S10 (pbs_4a) 116

Figure 173. Failure ratio of location calculations for operator, S10 (pbs_5a) 116

Figure 174. Failure ratio of location calculations using stand-alone GPS, S10(pbs_6) 116

Figure 175. Failure ratio of location calculations using stand-alone GPS, S10(pbs_6a) 117

Figure 176. Unspecified LCS requests, S10 (pbs_8) 117

Figure 177. Return from super TRXs to regular TRX, S4 (ho_1) 117



Figure 178. HO attempts from regular TRXs to super, S4 (ho_2) 117

Figure 179. HO attempts from super to regular, S4 (ho_3) 118

Figure 180. Share of HO attempts from super to regular due to DL quality, S4(ho_4) 118

Figure 181. Share of HO attempts from super to regular due to DL interference, S4(ho_5) 118

Figure 182. Share of HO attempts from super to regular due to UL interference, S4(ho_6) 118

Figure 183. Share of HO attempts from super to regular due to bad C/I, S4(ho_7) 119

Figure 184. MSC incoming HO attempts (ho_8) 119

Figure 185. MSC outgoing HO attempts (ho_9) 119

Figure 186. BSC incoming HO attempts (ho_10) 119

Figure 187. BSC outgoing HO attempts (ho_11) 119

Figure 188. Intra-cell HO attempts, S6 (ho_12a) 120

Figure 189. HO attempts, S3 (ho_13a) 120

Figure 190. HO attempts, outgoing and intra-cell, S5 (ho_13b) 120

Figure 191. HO attempts, outgoing and intra-cell, S3 (ho_13e) 121

Figure 192. HO attempts, outgoing and intra-cell, S9, (ho_13g) 121

Figure 193. TCH requests for HO (ho_14a) 121

Figure 194. TCH requests for HO (ho_14b) 122

Figure 195. TCH seizures for HO (ho_15) 122

Figure 196. TCH-TCH HO attempts (ho_16) 122

Figure 197. SDCCH-TCH HO attempts (ho_17) 122

Figure 198. SDCCH-SDCCH HO attempts (ho_18) 123

Figure 199. TCH-TCH HO successes (ho_19) 123

Figure 200. SDCCH-TCH HO successes (ho_20) 123

Figure 201. SDCCH-SDCCH HO successes (ho_21) 124

Figure 202. MSC controlled HO attempts (ho_22) 124

Figure 203. BSC controlled HO attempts (ho_23) 124

Figure 204. Intra-cell HO attempts (ho_24) 124

Figure 205. MSC controlled HO successes (ho_25) 125

Figure 206. BSC controlled HO successes (ho_26) 125


Figure 207. Intra-cell HO successes (ho_27) 125

Figure 208. MSC incoming HO successes (ho_28) 125

Figure 209. MSC outgoing HO successes (ho_29) 125

Figure 210. BSC incoming HO successes (ho_30) 126

Figure 211. BSC outgoing HO successes (ho_31) 126

Figure 212. Incoming HO success (ho_32) 126

Figure 213. Outgoing HO successes (ho_33) 126

Figure 214. Outgoing HO attempts (ho_34) 126

Figure 215. Incoming HO attempts (ho_35) 127

Figure 216. Outgoing SDCCH-SDCCH HO attempts (ho_36) 127

Figure 217. Incoming SDCCH-SDCCH HO attempts (ho_37) 127

Figure 218. Outgoing SDCCH-TCH HO attempts (ho_38) 127

Figure 219. Incoming SDCCH-TCH HO attempts (ho_39) 127

Figure 220. Outgoing TCH-TCH HO attempts (ho_40) 128

Figure 221. Incoming TCH-TCH HO attempts (ho_41) 128

Figure 222. Outgoing SDCCH-SDCCH HO success (ho_42) 128

Figure 223. Incoming SDCCH-SDCCH HO success (ho_43) 128

Figure 224. Outgoing SDCCH-TCH HO success (ho_44) 128

Figure 225. Incoming SDCCH-TCH HO success (ho_45) 129

Figure 226. Outgoing TCH-TCH HO success (ho_46) 129

Figure 227. Incoming TCH-TCH HO success (ho_47) 129

Figure 228. Intra-cell HO share, S1 (ho_48) 129

Figure 229. MSC controlled incoming HO attempts (ho_49) 130

Figure 230. Total HO failure %, S1 (hfr_1) 130

Figure 231. Total HO failure %, S1 (hfr_2) 131

Figure 232. Intra-cell HO failure share, S1 (hfr_3a) 132

Figure 233. Intra-cell HO failure share, S1 (hfr_3b) 132

Figure 234. Intra-cell HO failure share, S1 (hfr_3c) 133

Figure 235. Intra-cell HO failure share, S1 (hfr_3d) 133

Figure 236. Incoming MSC ctrl HO failure %, S1 (hfr_4) 133

Figure 237. Incoming MSC ctrl HO failure share, S1 (hfr_4a) 134



Figure 238. Incoming MSC ctrl HO failure share, S1 (hfr_4b) 134

Figure 239. Incoming MSC ctrl HO failure share, S1 (hfr_4c) 134

Figure 240. Incoming MSC ctrl HO failure share, S1 (hfr_4d) 135

Figure 241. Outgoing MSC ctrl HO failure ratio %, S1 (hfr_5) 135

Figure 242. Outgoing MSC ctrl HO failure share %, S1 (hfr_5a) 135

Figure 243. Outgoing MSC ctrl HO failure share %, S1 (hfr_5b) 136

Figure 244. Outgoing MSC ctrl HO failure share %, S1 (hfr_5c) 136

Figure 245. Outgoing MSC ctrl HO failure share %, S1 (hfr_5d) 137

Figure 246. Incoming BSC ctrl HO failure %, S1 (hfr_6) 137

Figure 247. Incoming BSC ctrl HO failure share %, S1 (hfr_6a) 137

Figure 248. Incoming BSC ctrl HO failure %, S1 (hfr_6b) 138

Figure 249. Incoming BSC ctrl HO failure share %, S1 (hfr_6c) 138

Figure 250. Incoming BSC ctrl HO failure %, S1 (hfr_6d) 138

Figure 251. Outgoing BSC ctrl HO failure share, S1 (hfr_7) 139

Figure 252. Outgoing BSC ctrl HO failure share, S1 (hfr_7a) 139

Figure 253. Outgoing BSC ctrl HO failure share, S1 (hfr_7b) 139

Figure 254. Outgoing BSC ctrl HO failure share, S1 (hfr_7c) 140

Figure 255. Outgoing BSC ctrl HO failure share, S1 (hfr_7d) 140

Figure 256. Internal inter HO failure %, S4 (hfr_8) 140

Figure 257. Internal intra HO failure %, S4 (hfr_9) 141

Figure 258. External source HO failure %, S4 (hfr_10) 141

Figure 259. HO failure % from super to regular, S4 (hfr_12) 141

Figure 260. HO failure % from regular to super, S4 (hfr_13) 141

Figure 261. Share of HO failures from regular to super due to return, S4 (hfr_14) 142

Figure 262. Share of HO failures from regular to super due to MS lost, S4(hfr_15) 142

Figure 263. Share of HO failures from regular to super due to another cause, S4(hfr_16) 142

Figure 264. Share of HO failures from super to regular due to return, S4 (hfr_17) 143

Figure 265. Share of HO failures from super to regular due to MS lost, S4(hfr_18) 143

Figure 266. Share of HO failures from super to regular due to another cause, S4(hfr_19) 143


Figure 267. SDCCH-SDCCH HO failure %, S2 (hfr_20) 144

Figure 268. SDCCH-TCH HO failure %, S2 (hfr_21) 144

Figure 269. TCH-TCH HO failure %, S2 (hfr_22) 144

Figure 270. SDCCH-SDCCH incoming HO failure %, S2 (hfr_23) 144

Figure 271. SDCCH-SDCCH outgoing HO failure ratio, S2 (hfr_24) 145

Figure 272. SDCCH-TCH incoming HO failure %, S2 (hfr_25) 145

Figure 273. SDCCH-TCH outgoing HO failure %, S2 (hfr_26) 145

Figure 274. TCH-TCH incoming HO failure %, S2 (hfr_27) 145

Figure 275. TCH-TCH outgoing HO failure %, S2 (hfr_28) 146

Figure 276. MSC ctrl HO failure %, blocking (hfr_29) 146

Figure 277. MSC ctrl HO failure %, not allowed (hfr_30) 146

Figure 278. MSC ctrl HO failure %, return to old (hfr_31) 146

Figure 279. MSC ctrl HO failure %, call clear (hfr_32) 147

Figure 280. MSC ctrl HO failure %, end HO (hfr_33) 147

Figure 281. MSC ctrl HO failure %, end HO BSS (hfr_34) 147

Figure 282. MSC ctrl HO failure %, wrong A interface (hfr_35) 147

Figure 283. MSC ctrl HO failure %, adjacent cell error (hfr_36) 148

Figure 284. BSC ctrl HO failure %, blocking (hfr_37) 148

Figure 285. BSC ctrl HO failure %, not allowed (hfr_38) 148

Figure 286. BSC ctrl HO failure %, return to old (hfr_39) 148

Figure 287. BSC ctrl HO failure %, call clear (hfr_40) 149

Figure 288. BSC ctrl HO failure %, end HO (hfr_41) 149

Figure 289. BSC ctrl HO failure %, end HO BSS (hfr_42) 149

Figure 290. BSC ctrl HO failure %, wrong A interface (hfr_43) 149

Figure 291. BSC ctrl HO drop call % (hfr_44) 150

Figure 292. Intra-cell HO failure %, cell_fail_lack (hfr_45) 150

Figure 293. Intra-cell HO failure %, not allowed (hfr_46) 150

Figure 294. Intra-cell HO failure %, return to old (hfr_47) 150

Figure 295. Intra-cell HO failure %, call clear (hfr_48) 151

Figure 296. Intra-cell HO failure %, MS lost (hfr_49) 151

Figure 297. Intra-cell HO failure %, BSS problem (hfr_50) 151



Figure 298. Intra-cell HO failure %, drop call (hfr_51) 151

Figure 299. HO failure % to adjacent cell (hfr_52) 152

Figure 300. HO failure % from adjacent cell (hfr_53) 152

Figure 301. HO failure %, blocking excluded (hfr_54a) 152

Figure 302. HO failure % due to radio interface blocking (hfr_55) 153

Figure 303. Intra-cell HO failure %, wrong A interface (hfr_56) 153

Figure 304. Intra-cell HO failure % (hfr_57) 153

Figure 305. HO failures to target cell, S6 (hfr_58) 154

Figure 306. HO failures from target cell, S6 (hfr_59) 154

Figure 307. HO drop ratio (hfr_68) 154

Figure 308. HO failures to target WCDMA cell, S10.5 (hfr_69) 155

Figure 309. HO failures from target WCDMA cell, S10.5 (hfr_70) 155

Figure 310. Intra-Segment SDCCH-SDCCH HO failure ratio from BCCH to non-BCCH layer, BSC level, S10.5 (hfr_71) 155

Figure 311. IIntra-segment SDCCH-SDCCH HO failure ratio between BTS types,BSC level, S10.5 (hfr_72) 156

Figure 312. Intra-segment TCH-TCH HO failure ratio between bands (due to load),BSC level, S10.5 (hfr_73) 156

Figure 313. Intra-segment TCH-TCH HO failure ratio between bands (due to downlinksignal level), BSC level, S10.5 (hfr_74) 157

Figure 314. Intra-segment TCH-TCH HO failure ratio between BTS types (due toload), BSC level, S10.5 (hfr_75) 157

Figure 315. MSC controlled outgoing SDCCH-SDCCH HO success %, S1(hsr_1) 157

Figure 316. MSC controlled outgoing SDCCH-TCH HO success %, S1 (hsr_2) 158

Figure 317. MSC controlled outgoing TCH-TCH HO success %, S1 (hsr_3) 158

Figure 318. BSC controlled outgoing SDCCH-SDCCH HO success %, S1(hsr_4) 158

Figure 319. BSC controlled outgoing SDCCH-TCH HO success %, S1 (hsr_5) 158

Figure 320. BSC controlled outgoing TCH-TCH HO success %, S1 (hsr_6) 159

Figure 321. Intra-cell SDCCH-SDCCH HO success %, S1 (hsr_7) 159

Figure 322. Intra-cell SDCCH-TCH HO success %, S1 (hsr_8) 159

Figure 323. Intra-cell TCH-TCH HO success %, S1 (hsr_9) 159

Figure 324. MSC controlled incoming SDCCH-SDCCH HO success %, S1


(hsr_10) 160

Figure 325. MSC controlled incoming SDCCH-TCH HO success %, S1 (hsr_11) 160

Figure 326. MSC controlled incoming TCH-TCH HO success %, S1 (hsr_12) 160

Figure 327. BSC controlled incoming SDCCH-SDCCH HO success %, S1(hsr_13) 160

Figure 328. BSC controlled incoming SDCCH-TCH HO success %, S1 (hsr_14) 161

Figure 329. BSC controlled incoming TCH-TCH HO success %, S1 (hsr_15) 161

Figure 330. BSC controlled incoming HO success %, S1 (hsr_16) 161

Figure 331. MSC controlled incoming HO success %, S1 (hsr_17) 161

Figure 332. Incoming HO success %, S1 (hsr_18) 161

Figure 333. Outgoing HO success %, S1 (hsr_19) 162

Figure 334. Intra-cell SDCCH-SDCCH HO success %, S1 (hsr_20) 162

Figure 335. Intra-cell SDCCH-TCH HO success %, S1 (hsr_21) 162

Figure 336. Intra-cell TCH-TCH HO success %, S1 (hsr_22) 162

Figure 337. Outgoing HO failures due to lack of resources (hof_1) 163

Figure 338. Incoming HO failures due to lack of resources (hof_2) 163

Figure 339. TCH HO failures when return to old channel was successful (hof_3) 163

Figure 340. SDCCH HO failures when return to old channel was successful(hof_4) 163

Figure 341. MSC incoming HO failures (hof_5) 164

Figure 342. MSC outgoing HO failures (hof_6) 164

Figure 343. MSC outgoing HO failures (hof_6a) 164

Figure 344. BSC incoming HO failures (hof_7) 164

Figure 345. BSC incoming HO failures (hof_7a) 165

Figure 346. BSC outgoing HO failures (hof_8) 165

Figure 347. BSC outgoing HO failures (hof_8a) 165

Figure 348. Intra-cell HO failures (hof_9) 165

Figure 349. Intra-cell HO failures (hof_9a) 165

Figure 350. Failed outgoing HO, return to old (hof_10) 166

Figure 351. Outgoing HO failures (hof_12) 166

Figure 352. Intra-cell HO failure, return to old channel (hof_13) 166

Figure 353. Intra-cell HO failure, drop call (hof_14) 166



Figure 354. Incoming HO failures (hof_15) 167

Figure 355. UL interference, BTS level, S1 (itf_1) 167

Figure 356. Idle TSL percentage of time in band X, TRX level, IUO, S4 (itf_2) 168

Figure 357. UL interference from IUO, TRX level, S4 (itf_3) 168

Figure 358. UL interference from Power Control, TRX level, S6 (itf_4) 168

Figure 359. TCH congestion time, S1 (cngt_1) 169

Figure 360. SDCCH congestion time, S1 (cngt_2) 169

Figure 361. FTCH time congestion % (cngt_3) 169

Figure 362. FTCH time congestion % (cngt_3a) 169

Figure 363. HTCH time congestion % (cngt_4) 170

Figure 364. HTCH time congestion % (cngt_4a) 170

Figure 365. Queued, served TCH call requests % (que_1a) 170

Figure 366. Queued, served TCH HO requests % (que_2) 171

Figure 367. Queued, served TCH HO requests % (que_2a) 171

Figure 368. Successful queued TCH requests (que_3) 171

Figure 369. Successful non-queued TCH requests (que_4) 171

Figure 370. Successful queued TCH HO requests (que_5) 172

Figure 371. Successful non-queued TCH HO requests (que_6) 172

Figure 372. Non-queued, served TCH call requests % (que_7) 172

Figure 373. Non-queued, served TCH HO requests % (que_8) 172

Figure 374. Non-queued, served TCH HO requests % (que_8a) 173

Figure 375. TCH raw blocking, S1 (blck_1) 173

Figure 376. SDCCH blocking %, S1 (blck_5) 173

Figure 377. SDCCH real blocking %, S1 (blck_5a) 174

Figure 378. TCH raw blocking % on super TRXs, S4 (blck_6) 174

Figure 379. TCH raw blocking % on regular TRXs, S4 (blck_7) 174

Figure 380. TCH call blocking, before DR, S2 (blck_8) 174

Figure 381. TCH call blocking %, DR compensated, S2 (blck_8b) 175

Figure 382. TCH call blocking %, DR and DAC compensated, EFR excluded, S5(blck_8d) 176

Figure 383. Blocked calls, S5 (blck_9b) 177


Figure 384. Blocked calls, S5 (blck_9c) 177

Figure 385. Blocked TCH HOs, S2 (blck_10a) 178

Figure 386. Blocked TCH HOs, S5 (blck_10b) 178

Figure 387. TCH HO blocking, S2 (blck_11a) 178

Figure 388. TCH HO blocking without Q, S2 (blck_11b) 179

Figure 389. TCH HO blocking, S5 (blck_11c) 179

Figure 390. Blocked incoming and internal HO, S2 (blck_12) 179

Figure 391. Blocked incoming and internal HO, S2 (blck_12a) 180

Figure 392. AG blocking, S1 (blck_13) 180

Figure 393. FCS blocking, S5 (blck_14) 180

Figure 394. Blocked SDCCH seizure attempts, S5 (blck_15) 180

Figure 395. HO blocking % (blck_16a) 181

Figure 396. Handover blocking % (blck_16b) 181

Figure 397. Blocked FACCH call setup TCH requests (blck_18) 181

Figure 398. Handover blocking to target cell (blck_19) 182

Figure 399. Handover blocking from target cell (blck_20) 182

Figure 400. NACK ratio of p-immediate assignment, S9PS (blck_21) 182

Figure 401. Territory upgrade rejection %, S9PS (blck_22) 183

Figure 402. Handover blocking to target WCDMA cell, S10.5 (blck_27) 183

Figure 403. Handover blocking from target WCDMA cell, S10.5 (blck_28) 183

Figure 404. TCH denied for Call request, Ratio, S10 (blck_29) 184

Figure 405. TCH traffic sum, S1 (trf_1) 184

Figure 406. TCH traffic sum, S1 (trf_1a) 185

Figure 407. TCH traffic sum of normal TRXs, S1 (trf_1b) 185

Figure 408. TCH traffic sum of extended TRXs, S1 (trf_1c) 185

Figure 409. Average call length, S1 (trf_2b) 186

Figure 410. Average call length, S1 (trf_2d) 186

Figure 411. CS territory usage, S1 (trf_3) 187

Figure 412. FTCH usage, S5 (trf_3b) 187

Figure 413. Average SDCCH holding time, S1 (trf_4) 187

Figure 414. Average FTCH holding time, S1 (trf_5) 188



Figure 415. TCH seizures for new call (call bids), S1 (trf_6) 188

Figure 416. SDCCH usage %, S1 (trf_7b) 188

Figure 417. SDCCH usage %, S1 (trf_7c) 189

Figure 418. TCH traffic absorption on super, S4 (trf_8) 189

Figure 419. TCH traffic absorption on super, S4 (trf_8a) 189

Figure 420. Average cell TCH traffic from IUO, S4 (trf_9) 190

Figure 421. Cell TCH traffic from IUO, S4 (trf_9a) 190

Figure 422. Super TRX TCH traffic, S4 (trf_10) 190

Figure 423. Sum of super TRX TCH traffic, S4 (trf_10a) 191

Figure 424. Average SDCCH traffic, erlang, S2 (trf_11) 191

Figure 425. Average SDCCH traffic, erlang, S2 (trf_11b) 191

Figure 426. Average TCH traffic, erlang, S2 (trf_12) 191

Figure 427. Average TCH traffic, erlang, S2 (trf_12a) 192

Figure 428. Average CS traffic, erlang, S2 (trf_12b) 192

Figure 429. Handover/call % (trf_13b) 193

Figure 430. Intra-cell handover/call % (trf_13c) 193

Figure 431. HO / call % (trf_13d) 193

Figure 432. Handover/call % (trf_13e) 194

Figure 433. IUO, average TCH seizure length on super TRXs, S4 (trf_14b) 194

Figure 434. IUO, average TCH seizure length on regular TRXs, S4 (trf_15b) 195

Figure 435. Average TRX traffic, IUO, S4 (trf_16) 195

Figure 436. Average TRX TCH seizure length, IUO, S4 (trf_17) 195

Figure 437. Average TRX TCH seizure length, IUO, S4 (trf_17a) 195

Figure 438. Average TRX TCH seizure length, IUO, S4 (trf_17b) 195

Figure 439. TCH requests for a new call, S3 (trf_18) 196

Figure 440. TCH requests for a new call, S3 (trf_18a) 196

Figure 441. Peak busy TCH (trf_19) 196

Figure 442. Average unit load (trf_20) 196

Figure 443. Call time difference between TRXs, S4 (trf_21) 197

Figure 444. Call time difference between TRXs, S4 (trf_21a) 198

Figure 445. Number of mobiles located in a cell, BSC (trf_23a) 198


Figure 446. Total TCH seizure time (call time in seconds) (trf_24b) 199

Figure 447. Total TCH seizure time (call time in hours) (trf_24c) 199

Figure 448. SDCCH true seizures (trf_25) 200

Figure 449. SDCCH true seizures, S7 (trf_25a) 200

Figure 450. SDCCH true seizures for call and SS (trf_26) 200

Figure 451. SDCCH true seizures for call, SMS, SS (trf_27) 200

Figure 452. Peak busy SDCCH seizures (trf_28) 201

Figure 453. IUO layer usage % (trf_29) 201

Figure 454. SDCCH seizures (trf_30) 201

Figure 455. Call time (minutes) from p_nbsc_res_avail (trf_32) 201

Figure 456. Non-AMR call time from p_nbsc_rx_qual (trf_32a) 202

Figure 457. Call time from p_nbsc_rx_statistics (trf_32b) 202

Figure 458. SDCCH HO seizure % out of SDCCH seizure attempts (trf_33) 203

Figure 459. SDCCH assignment % out of SDCCH seizure attempts (trf_34) 203

Figure 460. TCH HO seizure % out of TCH HO seizure request (trf_35) 203

Figure 461. TCH norm seizure % out of TCH call request (trf_36) 203

Figure 462. TCH normal seizure % out of TCH call requests (trf_36a) 204

Figure 463. TCH FCS seizure % out of TCH FCS attempts (trf_37) 204

Figure 464. TTCH FCS (due to SDCCH congestion) seizure % out of SDCCH seizureattempts (trf_38) 204

Figure 465. TCH seizures for new calls (trf_39) 204

Figure 466. TCH seizures for new calls (trf_39a) 205

Figure 467. HTCH usage, S5 (trf_40) 205

Figure 468. MOC rate, S6 (trf_41) 205

Figure 469. MTC rate, S6 (trf_42) 206

Figure 470. TCH single band subscriber holding time, S6 (trf_43) 206

Figure 471. TCH dual band subscriber holding time, S6 (trf_44) 206

Figure 472. Average SDCCH traffic, S1 (trf_45) 207

Figure 473. Share of single band traffic (trf_47) 207

Figure 474. Share of dual band traffic (trf_48) 207

Figure 475. Call retries due to A interface pool mismatch (trf_49) 207



Figure 476. HO retries due to A interface pool mismatch (trf_50) 208

Figure 477. TCH single band subscribers’ share of holding time, S6 (trf_51) 208

Figure 478. TCH dual band subscribers’ share of holding time, S6 (trf_52) 208

Figure 479. Calls started as FACCH call setup (trf_53) 208

Figure 480. SDCCH seizures (trf_54) 209

Figure 481. TCH normal seizures (trf_55) 209

Figure 482. Total FTCH seizure time (trf_56) 209

Figure 483. Total HTCH seizure time (trf_57) 209

Figure 484. Average TCH hold time for HSCSD, S7 (trf_58) 210

Figure 485. Average number of HSCSD users, S7HS (trf_60) 210

Figure 486. . . . . . . . . . . Average HSCSD main channel traffic, S7HS (trf_60a) 210

Figure 487. 210

Figure 488. . . . . . . . . . . . . .Average upgrade pending time for HSCSD (trf_62) 211

Figure 489. . . . . . . . Average upgrade pending time due to congestion (trf_63) 211

Figure 490. . . . . . . . . . . . . Total reporting time of ph1 and ph2 mobiles (trf_64) 211

Figure 491. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total TCH seizures (trf_65) 211

Figure 492. . . . . . . . . . . . . . . . .Net UL data traffic per timeslot, S9PS (trf_69a) 212

Figure 493. . . . . . . . . . . . . . . . .Net DL data traffic per timeslot, S9PS (trf_70a) 212

Figure 494. Average UL throughput per allocated timeslot, S9PS (trf_72b) 213

Figure 495. Average effective UL throughput per used tsl, S9PS (trf_72d) 214

Figure 496. Average effective UL throughput per used TSL, S10PS (trf_72f) 214

Figure 497. Average effective UL throughput per used TSL, S10PS (trf_72h) 215

Figure 498. . . . . Average DL throughput per allocated timeslot, S9PS (trf_73b) 216

Figure 499. Average effective DL throughput per used timeslot, S9PS (trf_73d) 217

Figure 500.Average effective DL throughput per used timeslot, S10PS (trf_73f)218

Figure 501. Average effective DL throughput per used TSL, S10PS (trf_73g) 219

Figure 502. Total RLC data, S9PS (trf_74) 220

Figure 503. . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total RLC data, S9PS (trf_74a) 220

Figure 504. Total GPRS RLC data, S9PS (trf_74b) 220

Figure 505. . . . . . . . . . . . . . . . . . .GPRS territory UL utilisation, S9PS (trf_76b) 221


Figure 506. . . . . . . . . . . . . . . . . . GPRS territory DL utilisation, S9PS (trf_77a) 222

Figure 507. UL GPRS traffic, S9PS (trf_78a) 223

Figure 508. DL GPRS traffic, S9PS (trf_79a) 223

Figure 509. TCH free margin, S9PS (trf_81) 224

Figure 510. Normal TCH usage % for CS (trf_83a) 224

Figure 511. TCH usage % for PS, S9PS (trf_84a) 224

Figure 512. Normal TCH usage % for PS, S9PS (trf_84b) 224

Figure 513. Total TCH usage % for CS, S9PS (trf_85) 225

Figure 514. Total TCH usage % for CS and PS, S9PS (trf_85b) 225

Figure 515. Free TCH %, S9PS (trf_86a) 225

Figure 516. Free TCH %, S9PS (trf_86b) 226

Figure 517. Free TCH %, S10.5PS (trf_86c) 226

Figure 518. Total TCH % for PS (trf_87b) 227

Figure 519. Total TCH % for dedicated PS, S9PS (trf_88b) 227

Figure 520. Average total UL throughput per used timeslot, S9PS (trf_89) 228

Figure 521. Average total UL throughput per used TSL, S10PS (trf_89a) 229

Figure 522. Average total DL throughput per used timeslot, S9PS (trf_90) 230

Figure 523. Average total DL throughput per used timeslot, S10PS (trf_90a) 230

Figure 524. SDCCH true seizures for call (trf_91) 230

Figure 525. Average HSCSD subchannel traffic, S7HS (trf_92) 231

Figure 526. Average HSCSD subchannel traffic, S7HS (trf_92a) 231

Figure 527. Voice calls on SDCCH, S1 (trf_93) 231

Figure 528. TCH traffic, S1 (trf_94) 232

Figure 529. GPRS traffic sum, S9PS (trf_95a) 232

Figure 530. GPRS territory utilisation, S9PS (trf_96a) 233

Figure 531. PS territory utilisation, S10.5PS (trf_96b) 235

Figure 532. Average CS traffic, normal TRXs, erlang, S2 (trf_97) 235

Figure 533. Average CS traffic, extended TRXs S2 (trf_98) 236

Figure 534. Average HSCSD traffic, normal TRXs, S7HS (trf_99) 236

Figure 535. Average HSCSD traffic, extended TRXs, S7HS (trf_100) 236

Figure 536. Average HTCH traffic, S7HS (trf_101) 236



Figure 537. Average HTCH traffic, normal TRXs, S7HS (trf_102) 237

Figure 538. Average HTCH traffic, extended TRXs, S7HS (trf_103) 237

Figure 539. Average HSCSD main channel traffic, normal TRXs, S7HS (trf_104) 237

Figure 540. Average HSCSD main channel traffic, extended TRXs, S7HS(trf_105) 237

Figure 541. Average FTCH single traffic, S7HS (trf_106) 238

Figure 542. Average FTCH single traffic, normal TRXs, S7HS (trf_107) 238

Figure 543. Average FTCH single traffic, extended TRXs, S7HS (trf_108) 238

Figure 544. Peak busy TCH on normal TRXs (trf_109) 238

Figure 545. Peak busy TCH on normal TRXs (trf_110) 239

Figure 546. Normal TCH usage % for CS (trf_111) 239

Figure 547. Normal TCH usage % for CS (trf_112) 239

Figure 548. CS call samples, non-AMR call (trf_113) 240

Figure 549. CS call samples, AMR call (trf_114) 240

Figure 550. TCH traffic time, non-AMR calls (trf_115) 240

Figure 551. TCH traffic time, AMR calls (trf_116) 241

Figure 552. TCH traffic time, FR AMR calls (trf_117) 241

Figure 553. TCH traffic time, HR AMR calls (trf_118) 241

Figure 554. TCH traffic time, all calls (trf_119) 242

Figure 555. TCH traffic share of non-AMR calls (trf_120) 242

Figure 556. TCH traffic share of FR AMR calls (trf_121) 242

Figure 557. TCH traffic share of HR AMR calls (trf_122) 242

Figure 558. Average effective UL timeslot throughput per TBF, S10PS (trf_123) 243

Figure 559. Average effective DL timeslot throughput per TBF, S10PS (trf_124) 243

Figure 560. MS specific flowrate (trf_125) 244

Figure 561. Total RLC payload data (Kbytes), MCS-n, S10.5PS (trf_131) 244

Figure 562. UL RLC data MCS-n, S10.5PS (trf_140) 245

Figure 563. DL RLC data MCS-n, S10.5PS (trf_141) 245

Figure 564. Normal TCH usage % for EGPRS, S10.5PS (trf_160) 246

Figure 565. . . . . . . . . . . . . . . . . . . . . . . . .UL EGPRS traffic, S10.5PS (trf_161) 246

Figure 566. DL EGPRS traffic, S10.5PS (trf_162) 247


Figure 567. Total EGPRS RLC data, S9PS (trf_167) 248

Figure 568. Average SDCCH traffic, Area, S1 (trf_168) 248

Figure 569. SDCCH seizures for MO calls, S2 (moc_1) 248

Figure 570. Successful MO speech calls, S3 (moc_2) 249

Figure 571. Successful MO data calls, S3 (moc_3) 249

Figure 572. MO call success ratio, S6 (moc_4) 249

Figure 573. MO speech call attempts, S3 (moc_5) 250

Figure 574. MO call bids, S2 (moc_6) 250

Figure 575. SDCCH seizures for MT calls, S2 (mtc_1) 250

Figure 576. Successful MT speech calls (mtc_2) 250

Figure 577. Successful MT data calls, S3 (mtc_3) 251

Figure 578. MT call success ratio, S6 (mtc_4) 251

Figure 579. MT speech call attempts (mtc_5) 251

Figure 580. MT call attempts, S2 (mtc_6) 251

Figure 581. Number of paging messages sent, S2 (pgn_1) 252

Figure 582. Paging buffer size average, S1 (pgn_2) 252

Figure 583. Average paging buffer space, S1 (pgn_3) 253

Figure 584. Average free space of paging GSM buffer area, S1 (pgn_3a) 253

Figure 585. Paging success ratio, S1 (pgn_4) 253

Figure 586. . . . . . . Average paging buffer air interface occupancy, S7 (pgn_5) 253

Figure 587. . . . . . . . . . . .Average paging buffer Gb occupancy, S7PS (pgn_6) 254

Figure 588. Average air interface DRX buffer load, due to paging, S7 (pgn_7) 254

Figure 589. Average air interface DRX buffer load, due to DRX AG, S7 (pgn_8) 254

Figure 590. Average air interface non-DRX buffer load due to AG, S7 (pgn_9) 255

Figure 591. Average free space of paging GPRS buffer area, S9 (pgn_10) 255

Figure 592. SMS establishment failure % (sms_1) 255

Figure 593. SMS TCH establishment failure % (sms_2) 255

Figure 594. SMS SDCCH establishment failure % (sms_3) 256

Figure 595. SMS establishment attempts (sms_4) 256

Figure 596. SMS SDCCH establishment attempts (sms_5) 256

Figure 597. SMS TCH establishment attempts (sms_6) 257



Figure 598. DR, outgoing attempts, S3 (dr_1) 257

Figure 599. DR attempts, S3 (dr_1a) 257

Figure 600. DR, incoming attempts, S3 (dr_2) 257

Figure 601. DR, outgoing success to another cell, S3 (dr_3) 257

Figure 602. DR, incoming success from another cell, S3 (dr_4) 258

Figure 603. DR, intra-cell successful HO, S3 (dr_5) 258

Figure 604. % of new calls successfully handed over to another cell by DR, S3(dr_6) 258

Figure 605. DR, outgoing to another cell, failed, S3 (dr_7) 258

Figure 606. DR, intra-cell failed, S3 (dr_8) 259

Figure 607. TCH availability %, S4 (ava_1a) 259

Figure 608. TCH availability %, S9 (ava_1c) 260

Figure 609. TCH availability %, S9 (ava_1d) 260

Figure 610. Average available TCH, S1 (ava_2) 260

Figure 611. Average available SDCCH, S1 (ava_3) 261

Figure 612. SDCCH availability %, S4 (ava_4) 261

Figure 613. Average available FTCH in area, S1 (ava_5) 261

Figure 614. DMR availability %, S6 (ava_6) 262

Figure 615. DN2 availability %, S6 (ava_7) 262

Figure 616. TRU availability %, S6 (ava_8) 262

Figure 617. Average defined HTCH, S1 (ava_9) 262

Figure 618. SC ET availability %, S7 (ava_10) 263

Figure 619. BSC ET availability %, S7 (ava_11) 263

Figure 620. SC TCSM availability %, S7 (ava_12) 263

Figure 621. BSC TCSM availability %, S7 (ava_13) 263

Figure 622. TRE availability %, S6 (ava_14) 264

Figure 623. Average CS territory, S9 (ava_15) 264

Figure 624. . . . . . . . . . . . . . Average available PDTCH in BTS, S9PS (ava_16) 265

Figure 625. Average PS territory, S9PS (ava_16a) 265

Figure 626. . . . . Average available dedicated GPRS channels, S9PS (ava_17) 266

Figure 627. Average available dedicated GPRS channels, S9PS (ava_17a) 266


Figure 628. TRE-SEL availability %, S6 (ava_20) 266

Figure 629. Number of timeslots available for CS traffic, S9 (ava_21) 267

Figure 630. Number of timeslots available for CS traffic on normal TRXs, S9(ava_21a) 267

Figure 631. Number of HR timeslots available, S9 (ava_22) 267

Figure 632. Number of HR timeslots available, S9 (ava_22a) 268

Figure 633. Number of FR timeslots available, S9 (ava_23) 268

Figure 634. Number of FR timeslots available, S9 (ava_23a) 268

Figure 635. Number of dual timeslots available, S9 (ava_24) 268

Figure 636. Number of dual timeslots available, S9 (ava_24a) 269

Figure 637. Average number of available TCH timeslots, S9 (ava_25a) 269

Figure 638. Number of available TCH timeslots, PS and CS common, S9(ava_26) 269

Figure 639. Number of available TCH timeslots, PS and CS common, S9(ava_26a) 270

Figure 640. Average CS TCH in normal TRXs, S9 (ava_28) 270

Figure 641. Average available CS TCH in extended TRXs, S9 (ava_29) 271

Figure 642. Number of HR tsls available, normal TRXs, S9 (ava_30) 271

Figure 643. Number of HR tsls available, extended TRXs S9 (ava_31) 271

Figure 644. Number of FR timeslots available, normal TRXs, S9 (ava_32) 272

Figure 645. Number of FR timeslots available, extended TRXs, S9 (ava_33) 272

Figure 646. Number of dual timeslots available, normal TRXs, S9 (ava_34) 272

Figure 647. Number of dual timeslots available, extended TRXs, S9 (ava_35) 273

Figure 648. GPRS enable time %, S10 (ava_36) 273

Figure 649. Number of HR TSLs available, extended TRXs, Area, S9 (ava_37) 273

Figure 650. Number of FR TSLs available, extended TRXs, Area, S9 (ava_38) 274

Figure 651. Average number of available TCH TSLs, Area, S9 (ava_39) 274

Figure 652. Number of dual TSLs available, extended TRXs, Area, S9 (ava_40) 275

Figure 653. Number of dual TSLs available, normal TRXs, Area, S9 (ava_41) 275

Figure 654. Number of HR TSLs available, normal TRXs, Area, S9 (ava_42) 276

Figure 655. Number of FR TSLs available, normal TRXs, Area, S9 (ava_43) 276

Figure 656. Average PS territory, Area, S9PS (ava_44) 277



Figure 657. Average available SDCCH, Area, S1 (ava_45) 277

Figure 658. Average unavailable TSL per BTS, S1 (uav_1) 277

Figure 659. Average unavailable TSL per BTS, S1 (uav_1a) 277

Figure 660. Average unavailable TSL per BTS, S1 (uav_1b) 278

Figure 661. Total outage time (uav_2) 278

Figure 662. Number of outages (uav_3) 278

Figure 663. Share of unavailability due to user (uav_4) 279

Figure 664. Share of unavailability due to internal reasons (uav_5) 279

Figure 665. Share of unavailability due to external reasons (uav_6) 279

Figure 666. TRX unavailability time due to user (uav_7) 279

Figure 667. TRX unavailability time due to internal reasons (uav_8) 280

Figure 668. TRX unavailability time due to external reasons (uav_9) 280

Figure 669. Average unavailable SDCCH, S5 (uav_10) 280

Figure 670. Average unavailable TCH, S5 (uav_11a) 280

Figure 671. Average bearer unavailability, S9PS (uav_12) 281

Figure 672. Average unavailable TCH on normal TRXs, S5 (uav_13) 281

Figure 673. Average unavailable TCH on extended TRXs, S5 (uav_14) 281

Figure 674. Average unavailable TCH on normal TRXs, Area, S5 (uav_15) 281

Figure 675. Average unavailable TCH on extended TRXs, Area, S5 (uav_16) 282

Figure 676. Average unavailable TCH, Area, S5 (uav_17) 282

Figure 677. Number of LU attempts, S1 (lu_1) 282

Figure 678. Average of LU attempts per hour, S1 (lu_2) 282

Figure 679. Number of LU attempts, S1 (lu_3) 283

Figure 680. LU success %, S6 (lsr_2) 283

Figure 681. Emergency calls, S6 (ec_1) 283

Figure 682. Emergency call success %, S6 (ecs_1) 284

Figure 683. Call re-establishment attempts, S6 (re_1) 284

Figure 684. Call re-establishments, S6 (re_2) 284

Figure 685. Call re-establishment success %, S6 (res_1) 285

Figure 686. DL BER, S1 (dlq_1) 285

Figure 687. DL cumulative quality % in class X, S1 (dlq_2) 285


Figure 688. DL cumulative quality % in class X, S1 (dlq_2a) 286

Figure 689. DL quality %, FER based, S10 (dlq_3) 286

Figure 690. DL cumulative quality % in class X, HR AMR, S10 (dlq_4) 287

Figure 691. DL cumulative quality % in class X, FR AMR, S10 (dlq_5) 287

Figure 692. DL cumulative quality % in class X,S10 (dlq_6) 288

Figure 693. DL quality 0-5 %, HR, FER based, S10 (dlq_7) 288

Figure 694. DL quality 0-5 %, FR, FER based, S10 (dlq_8) 288

Figure 695. DL quality 0-5 % EFR, FER based, S10 (dlq_9) 289

Figure 696. DL quality 0-5 % AMR HR, FER based, S10 (dlq_10) 289

Figure 697. DL quality 0-5 % AMR FR, FER based, S10 (dlq_11) 289

Figure 698. UL BER, S1 (ulq_1) 289

Figure 699. UL cumulative quality % in class X, S1 (ulq_2) 290

Figure 700. UL cumulative quality % in class X, S1 (ulq_2a) 290

Figure 701. UL quality %, FER based, S10 (ulq_3) 291

Figure 702. UL cumulative quality % in class X, HR AMR, S10 (ulq_4) 291

Figure 703. UL cumulative quality % in class X, FR AMR, S10 (ulq_5) 291

Figure 704. UL cumulative quality % in class X, non-AMR S10 (ulq_6) 292

Figure 705. UL quality 0-5 %, HR, FER based, S10 (ulq_7) 292

Figure 706. UL quality 0-5 %, FR, FER based, S10 (ulq_8) 293

Figure 707. UL quality 0-5 % EFR, FER based, S10 (ulq_9) 293

Figure 708. UL quality 0-5 % AMR HR, FER based, S10 (ulq_10) 293

Figure 709. UL quality 0-5 % AMR FR, FER based, S10 (ulq_11) 294

Figure 710. Share % per range, S4 (dll_1) 294

Figure 711. Sorting factor for undefined adjacent cell, S4 (dll_2) 294

Figure 712. Share % per range, S4 (ull_1) 295

Figure 713. Average MS power, S2 (pwr_1) 295

Figure 714. Average MS power, S2 (pwr_1b) 295

Figure 715. Average BS power, S2 (pwr_2) 295

Figure 716. Average DL signal strength, S2 (lev_1) 296

Figure 717. Average DL signal strength, S2 (lev_1a) 296

Figure 718. Average UL signal strength, S2 (lev_2) 296



Figure 719. Average UL signal strength, S2 (lev_2a) 296

Figure 720. Average MS-BS distance (dis_1) 297

Figure 721. Average MS-BS distance (dis_1a) 297

Figure 722. MS-BS distance class upper range (dis_3a) 298

Figure 723. Link balance, S1 (lb_1) 298

Figure 724. Share in acceptance range, S4 (lb_2) 298

Figure 725. Share in normal, S4 (lb_3) 298

Figure 726. Share in MS limited, S4 (lb_4) 299

Figure 727. Share in BS limited, S4 (lb_5) 299

Figure 728. Share in maximum power, S4 (lb_6) 299

Figure 729. Average MS power attenuation, S2 (lb_7) 299

Figure 730. Average MS power, S2 (lb_7b) 300

Figure 731. Average UL signal strength, S2 (lb_9) 300

Figure 732. Average DL signal strength, S2 (lb_10) 300

Figure 733. Average MS power attenuation, S2 (lb_11) 300

Figure 734. Average BS power attenuation, S2 (lb_12) 301

Figure 735. Average link imbalance, S2 (lb_13) 301

Figure 736. SDCCH access probability, before FCS (csf_1) 301

Figure 737. SDCCH access probability (csf_1a) 302

Figure 738. SDCCH success ratio (csf_2a) 303

Figure 739. SDCCH success ratio, area (csf_2e) 303

Figure 740. SDCCH success ratio, BTS, S6 (csf_2g) 304

Figure 741. SDCCH success ratio, BTS (csf_2i) 305

Figure 742. SDCCH success ratio, area (csf_2m) 306

Figure 743. SDCCH success ratio, BTS (csf_2n) 307

Figure 744. SDCCH success ratio, area, S10.5 (csf_2o) 308

Figure 745. TCH access probability without DR (csf_3a) 309

Figure 746. TCH access probability without DR and Q (csf_3b) 309

Figure 747. TCH access probability without Q (csf_3c) 309

Figure 748. TCH access probability, real (csf_3d) 310

Figure 749. TCH access probability without DR (csf_3i) 310


Figure 750. TCH access probability without DR and Q (csf_3j) 311

Figure 751. TCH access probability, real (csf_3k) 311

Figure 752. TCH access probability, real (csf_3l) 311

Figure 753. TCH access probability without DR and Q (csf_3m) 312

Figure 754. TCH success ratio, area, before call re-establisment (csf_4o) 313

Figure 755. TCH success ratio, area, after call re-establishment, S6 (csf_4p) 313

Figure 756. TCH success ratio, BTS, before call re-establisment (csf_4q) 314

Figure 757. TCH success ratio, BTS, after call re-establishment (csf_4r) 315

Figure 758. TCH success ratio, BTS, after call re-establishment (csf_4t) 315

Figure 759. TCH success ratio, area, before call re-establishment, S7(csf_4u) 316

Figure 760. TCH success ratio, area, after call re-establishment, S7 (csf_4v) 317

Figure 761. TCH success ratio, BTS, after call re-establishment (csf_4x) 317

Figure 762. TCH success ratio, BTS, before call re-establishment (csf_4y) 318

Figure 763. Activation related SDCCH access probability, S7, (csf_12) 318

Figure 764. SDCCH call success probability, S10.5 (csf_13a) 319

Figure 765. Reuse pattern (cnf_1) 319

Figure 766. Reuse pattern, S1 (cnf_2) 319




About this manual

Note

1 About this manualThis document defines the formulas used to calculating the Key PerformanceIndicators based on the Nokia NetAct PM database.

This document contains also formulas which are not used in any actual reports ofthe Nokia NetAct post-processing tools. You can see the formulas used in post-processing from the actual reports of the tools.

This document serves as a reference to the available formulas and does notinclude information on whether the formula is in use or not.

The information contained in this document relates to BSS Network Doctorsoftware version 3.1.7, Nokia NetAct release OSS3.1 and to release S10.5 ED ofthe Nokia BSC software. This document should not be used with any otherversions of the Nokia NetAct or Nokia BSC software.

This document is intended for the network operators of the Nokia NetAct.

This chapter covers the following topics:

• Summary of changes

• What you need to know first

• Where to find more

• Typographic conventions

• Concepts and terminology

1.1 Summary of changes

In this Change Delivery (OSS CD 0355)

As a result of the changes made to BSS Network Doctor software from version to3.1.6. to 3.1.7, the following changes have been made into this document:



New formulas

• Dynamic Abis Pool

- dap_3

• TBF

- tbf_37b replaced with tbf_37c- tbf_38b replaced with tbf_38c

• Call success

- csf_2o

• Traffic

- trf_4 replaced with trf_169- trf_45 replaced with trf_168- trf_72f replaced with trf_72h- trf_73f replaced with trf_73g

Availability

- ava_1d replaced with ava_1e- ava_3 replaced with ava_45- ava_33 replaced with ava_38- ava_30 replaced with ava_42- ava_31replaced with ava_37- ava_32 replaced with ava_43- ava_35 replaced with ava_40- ava_16 replaced with ava_16b- ava_4 replaced with ava_4a- ava_1d replaced with ava_1e

Unavailability

- uav_14 replaced with uav_17- uav_13 replaced with uav_15- uav_14 replaced with uav_16- uav_11 replaced with uav_11a- uav_11a replaced with uav_17

In the previous Change Delivery (OSS CD 0189)

As a result of the changes made to BSS Network Doctor software from version to3.1.5. to 3.1.6, the following changes have been made into this document:

New formulas


About this manual

• Blocking

- blck_29

• HFR

- hfr_71- hfr_72- hfr_73- hfr_74- hfr_75

• Availability

- ava_36

• RLC

- rlc_10c- rlc_11d

• TBF

- tbf_44

• SDR

- sdr_4

• DCR

- dcr_8c- dcr_8e

• TRF

- trf_94

Removed formulas

• TRF

- trf_123- trf_124

• DCR

- dcr_8- dcr_8b



1.2 What you need to know first

This document assumes that you are familiar with the following areas:

• The concepts of the Nokia NetAct and GSM networks in general

• A text processing utility, such as vi or dtpad. These text processors areused for editing certain configuration files.

1.3 Where to find more

When you perform the user’s tasks described in this document, you may need torefer to other documentation for further information. Below is a list of manualsthat you will find useful, as well as a brief description of the manual’s contents.

Other BSS Network Doctor documents

• Administering BSS Network Doctor, DN98619369, for systemadministrator’s tasks related to running BSS Network Doctor.

• BSS Network Doctor Reports, DN98614186, for a detailed description onutilising the Network Doctor reports.

OSS NED library documents

• Database Description for BSC Measurements, DN98619454, for adescription of the structure of performance management (PM) tables in theNokia NetAct PM database and the records, including counters, in eachtable.

Other Nokia documents

• Call Related DX Causes in BSC, Functional Description, DN9814277, foran explanation of phases and for a list of causes in TCH and SDCCHobservations to find details for dropping calls.#1

• DX Cause Coding Mapping, DN9813878, for an explanation to therelationship between DX cause codes and PM counters and for the analysisof TCH and SDCCH observations.#2

1.4 Typographic conventions

The following tables present the typographic conventions which have been usedin this manual to describe different actions and restrictions.


About this manual

1.4.1 Text styles

The following table presents the typefaces and fonts and their indications.

1.5 Terms and concepts

The lists below presents the terms and abbreviations used in this document.

Table 1. Text styles in this document

Style Explanation

Initial Upper-caseLettering

Application names

Italiced text Emphasis

State, status or mode

Courier File and directory names

Names of database tables

Parameters

User names

System output

User input

UPPER-CASELETTERING

Keys on the keyboard (ALT, TAB, CTRL etc.)

Bold text User interface components

Initial Upper-caseLettering in Italics

Referenced documents

Referenced sections and chapters within a document

<bracketed text> Variable user input



1.5.1 Abbreviations

Table 2. Abbreviations

Abbreviation Explanation

AG Access Grant

AMR Adaptive Multirate

BCCH Broadcast Control Channel

BCF Base Control Function

BER Bit Error Ratio

BSC Base Station Controller

BSS Base Station Subsystem

BTS Base Transceiver Station

CI Cell Identity

DL Downlink

DR Directed Retry

FCS Frame Check Sequence

GPRS General Packet Radio Service

HO Handover

HSCSD High Speed Circuit Switched Data

IUO Intelligent Underlay Overlay

KPI Key Performance Indicator

LU Location Update

MML Man-machine Language

MOC Mobile Originated Call

MR Maintenance Region

MS Mobile Station

MSC Mobile Services Switching Centre

OMC Operation and Maintenance Centre

PBS Position Based Services

PI Performance Indicator

PLMN Public Land Mobile Network

PM Performance Management


About this manual

1.5.2 Terms

The lists below presents the abbreviations and terms used in this document.

RACH Random Access Control Channel

SDCCH Stand Alone Dedicated Control Channel

SMS Short Message Service

SQL Standard Query Language

SS Supplementary Service

TCH Traffic Channel

TR Trunk Reservation

TRX Transceiver

TSL Timeslot

UL Uplink

Table 2. Abbreviations (Continued)

Abbreviation Explanation

Table 3. Terms used in this document

Term Explanation

AGCH A downlink control channel that is used to carry aresponse to a mobile channel allocation request.The AGCH assigns the mobile to operate on aspecific TDMA timeslot.

Bit Error Ratio The ratio of the number of the bit errors to the totalnumber of bits transmitted within a given time period.

Broadcast Control Channel (BCCH) A channel from a base station to a mobile station(MS) used for transmission of messages to allmobile stations located in the cell coverage area.

Cell Identity (CI) A number which identifies a cell to the networkswithin a location area (LA).

Clear Code Code that describes why the call set-up or the callitself has been disconnected.



For any other terms, refer to Glossary, DN9763965.

Day The counting of data per day is based on theperiod_start_time field in the measurementtables. This field always tells the starting hour of themeasurement period. Under one day there are hoursfrom 00 to 23.

Directed Retry A procedure used in a call set-up phase forassigning a mobile station to a traffic channel of acell other than the serving cell when the traffic iscongested.

Frame Check Sequence Extra characters added to a frame for the purposesof error control. The FCS is used in HDCL, FrameRelay, and other data link layer protocols.

Key Performance Indicator The performance of the network is calculated fromthe Nokia NetAct based on the Network Elementcounter information. Sometimes the plain counter assuch describes an important performance aspect(number of calls, for example) of the network butsometimes a formula of counters is needed (e.g.drop call ratio).

Maintenance Region Each object in the NetAct database belongs to oneand only one Maintenance Region (MR).

Mobile Terminated Call A call in which the called subscriber used a mobiletelephone.

Nokia NetAct A product of Nokia Telecommunications for theoperation and maintenance of cellular networks.

SQL*Plus An interactive program for accessing the database.

Stand-alone Dedicated ControlChannel (SDCCH)

A control channel (CCH) used for roaming,authentication, encryption activation and call control.

Timeslot (TSL) A timeslot in the time division multiple access(TDMA) frame in the GSM radio interface.

Traffic Channel A logical radio channel assigned to a base stationand primarily intended for conversation.

Trunk Reservation A stochastic algorithm employed in a channelallocation from a cell.

Table 3. Terms used in this document (Continued)

Term Explanation


BSS counter formulas

Note

2 BSS counter formulasThis chapter lists all BSS Network Doctor formulas. The more commonly usedones are described in further detail concerning their use or known problems withthem, for example. In connection with the name of a formula there is also areference to the BSC release (S1 to S10.5) since when the counters of the formulahave been available.

The use of formulas backwards, S4 formulas with S3 for example, gives either noresults because the measurement is not available, or false results because somecounters, which are new in S4, will be filled with value -1 by the OMC for S3BSCs.

When running the reports with newer counters, be careful especially when youhave two BSC software releases running in the network simultaneously. Thesimplest way to avoid problems is to start to use new counters of a new BSCrelease only when the new software release is used in the entire network under theNokia NetAct framework release.

2.1 Additional GPRS channels (ach)

Additional GPRS channel use, S9PS (ach_1)

Use: BTS level, especially BH values can be used for adjusting theCDEF parameter.

Example: If the value equals to one, on average one additional timeslothas been used for GPRS. If the situation continues, it indicatesa need to extend the default or the dedicated territory.

Known problems: The numerator is incremented when the TBF is released.Therefore, if there is, for example, one long TBF but no othertraffic, the value of this KPI can be totally incorrect becausethe whole TBF duration is counted on one measurementperiod.

Experiences on use: ach_1 is included in ava_16.

total hold time of all additional GPRS ch. seizures (sec)



----------------------------------------------------------- =period duration

sum(AVE_ADD_GPRS_CH_HOLD_TIME_SUM)/100---------------------------------------sum(period_duration*60)

Counters from table(s):p_nbsc_res_avail

Unit: timeslot

Figure 1. Additional GPRS channel use, S9PS (ach_1)

Average additional GPRS channel hold time, S9PS (ach_2)

Use: If the value is high, more default area is needed.

total hold time of all additional GPRS ch. seizures (sec)--------------------------------------------------------------- =total nbr of all additional GPRS channel seizures

sum(AVE_ADD_GPRS_CH_HOLD_TIME_SUM)/100--------------------------------------------------------------------------sum(decode(AVE_ADD_GPRS_CH_HOLD_TIME_SUM ,0,0,AVE_ADD_GPRS_CH_HOLD_TIME_DEN)

Counters from table(s):p_nbsc_res_availUnit: sec

Figure 2. Average additional GPRS channel hold time, S9PS (ach_2)

Additional GPRS channels seized, S9PS (ach_3)

Use: How many times an additional channel has been released (acase of territory downgrade).

Known problems: Shows slightly incorrect values in the case of an extended cell.

sum(decode(AVE_ADD_GPRS_CH_HOLD_TIME_SUM,0,0,AVE_ADD_GPRS_CH_HOLD_TIME_DEN)


Figure 3. Additional GPRS channels seized, S9PS (ach_3)

Total additional GPRS channel hold time, S9PS (ach_4)

Use: How many times an additional channel has been released (acase of territory downgrade).

Known problems: Shows slightly incorrect values in the case of an extended cell.

sum(AVE_ADD_GPRS_CH_HOLD_TIME_SUM)/100

Counters from table(s):



p_nbsc_res_availUnit: sec

Figure 4. Total additional GPRS channel hold time, S9PS (ach_4)

2.2 Multislot (msl)

Distribution of UL multislot requests, S9PS (msl_1)

Use: Indicates the share of a multislot request type to all multislotrequests.

req_X_TSL_UL100 * -------------------------------------------------------------------------- %

sum(req_1_TSL_UL+req_2_TSL_UL+req_1_TSL_UL +req_4_TSL_UL+ req_5_8_TSL_UL)

req_X_TSL_UL = one of the componentsof the denominator.

Counters from table(s):p_nbsc_packet_control_unit

Figure 5. Distribution of UL multislot requests, S9PS (msl_1)

Distribution of DL multislot requests, S9PS (msl_2)


req_X_TSL_DL100 * -------------------------------------------------------------------------- %

sum(req_1_TSL_DL+req_2_TSL_DL+req_1_TSL_DL +req_4_TSL_DL+ req_5_8_TSL_DL)

req_X_TSL_UL = one of the components of the denominator.


Figure 6. Distribution of DL multislot requests, S9PS (msl_2)

Distribution of UL multislot allocations, S9PS (msl_3)


alloc X TSL UL100 * ---------------------------------------------------- %

sum(alloc_1_TSL_UL + alloc_2_TSL_UL + alloc_1_TSL_UL+alloc_4_TSL_UL + alloc_5_8_TSL_UL)

req_X_TSL_UL = one of the components of the denominator.




Figure 7. Distribution of UL multislot allocations, S9PS (msl_3)

Distribution of DL multislot allocations, S9PS (msl_4)


alloc_X_TSL_DL100 * ------------------------------------------------------ %

sum(alloc_1_TSL_DL+alloc_2_TSL_DL+alloc_1_TSL_DL+alloc_4_TSL_DL+ alloc_5_8_TSL_DL)

req_X_TSL_DL = one of the components of the denominator.


Figure 8. Distribution of DL multislot allocations, S9PS (msl_4)

Ratio of unreserved GPRS UL TSL requests, S9PS (msl_5)


sum(alloc_1_TSL_UL+2*alloc_2_TSL_UL+3*alloc_3_TSL_UL +4*alloc_4_TSL_UL)100 * ----------------------------------------------------------------------- %

sum(req_1_TSL_UL+2*req_2_TSL_UL+3*req_3_TSL_UL +4*req_4_TSL_UL)


Figure 9. Ratio of unreserved GPRS UL TSL requests, S9PS (msl_5)

Ratio of unreserved GPRS DL TSL requests, S9PS (msl_6)


sum(alloc_1_TSL_DL+2*alloc_2_TSL_DL+3*alloc_3_TSL_DL +4*alloc_4_TSL_DL)100 * ----------------------------------------------------------------------- %

sum(req_1_TSL_DL+2*req_2_TSL_DL+3*req_3_TSL_DL +4*req_4_TSL_DL)


Figure 10. Ratio of unreserved GPRS DL TSL requests, S9PS (msl_6)



UL multislot allocations, S9PS (msl_9)

Use: Total number of multislot allocations in UL.

sum(alloc_1_TSL_UL+ alloc_2_TSL_UL+ alloc_3_TSL_UL

+ alloc_4_TSL_UL+ alloc_5_8_TSL_UL)


Figure 11. UL multislot allocations, S9PS (msl_9)

DL multislot allocations, S9PS (msl_10)

Use: Total number of multislot allocations in DL.

sum(alloc_1_TSL_DL+ alloc_2_TSL_DL+ alloc_3_TSL_DL

+ alloc_4_TSL_DL+ alloc_5_8_TSL_DL)


Figure 12. DL multislot allocations, S9PS (msl_10)

Average number of allocated timeslots, UL S9PS (msl_11)

sum(alloc_1_TSL_UL + 2*alloc_2_TSL_UL + 3*alloc_3_TSL_UL + 4*alloc_4_TSL_UL)----------------------------------------------------------------------------sum(alloc_1_TSL_UL + alloc_2_TSL_UL + alloc_3_TSL_UL+alloc_4_TSL_UL)


Figure 13. Average number of allocated timeslots, UL S9PS (msl_11)

Average number of allocated timeslots, DL S9PS (msl_12)

sum(alloc_1_TSL_DL + 2*alloc_2_TSL_DL + 3*alloc_3_TSL_DL + 4*alloc_4_TSL_DL)----------------------------------------------------------------------------sum(alloc_1_TSL_DL + alloc_2_TSL_DL + alloc_3_TSL_DL + alloc_4_TSL_DL)


Figure 14. Average number of allocated timeslots, DL S9PS (msl_13)



Average number of requested UL timeslots, S9PS (msl_13)

Known problems: If one phase access is used (MS on CCCH), only the requestedone single timeslot can be requested. Otherwise the MS classdefines how many timeslots are requested. This makes themeaning of the KPI less accurate.

sum(req_1_TSL_UL + 2*req_2_TSL_UL + 3*req_3_TSL_UL + 4*req_4_TSL_UL)--------------------------------------------------------------------sum(req_1_TSL_UL + req_2_TSL_UL + req_3_TSL_UL + req_4_TSL_UL)


Figure 15. Average number of requested UL timeslots, S9PS (msl_13)

Average number of requested DL timeslots, S9PS (msl_14)

Known problems: If one phase access is used (MS on CCCH), only the requestedone single timeslot can be requested. Otherwise the MS classdefines how many timeslots are requested. This makes themeaning of the KPI less accurate.

sum(req_1_TSL_DL + 2*req_2_TSL_DL + 3*req_3_TSL_DL + 4*req_4_TSL_DL)--------------------------------------------------------------------sum(req_1_TSL_DL + req_2_TSL_DL + req_3_TSL_DL + req_4_TSL_DL)


Figure 16. Average number of requested DL timeslots, S9PS (msl_14)

UL multislot allocation %, S9PS (msl_15a)

Use: Indicates how well the requested multislots could beallocated.

Known problems: Works until there are MSs with multislot class greater than 4.

100* average allocated tsl / average requested tsl % =

sum(alloc_1_TSL_UL+2*alloc_2_TSL_UL+3*alloc_3_TSL_UL +4*alloc_4_TSL_UL)------------------------------------------------------------------

sum(alloc_1_TSL_UL+alloc_2_TSL_UL+alloc_3_TSL_UL+alloc_4_TSL_UL++NO_RADIO_RES_AVA_UL_TBF)

100* ------------------------------------------------------------------------- %sum(req_1_TSL_UL+2*req_2_TSL_UL+3*req_3_TSL_UL +4*req_4_TSL_UL)----------------------------------------------------------------sum(req_1_TSL_UL+req_2_TSL_UL+req_3_TSL_UL+req_4_TSL_UL)


Figure 17. UL multislot allocation %, S9PS (msl_15a)



DL multislot allocation %, S9PS (msl_16a)

Use: Indicates how well the requested multislots could beallocated.

Known problems: Works until there are MSs with multislot class greater than 4.

100* average allocated tsl / average requested tsl % =

sum(alloc_1_TSL_DL+2*alloc_2_TSL_DL+3*alloc_3_TSL_DL +4*alloc_4_TSL_DL)------------------------------------------------------------------

sum(alloc_1_TSL_DL+alloc_2_TSL_DL+alloc_3_TSL_DL+alloc_4_TSL_DL+ NO_RADIO_RES_AVA_DL_TBF)

100* ------------------------------------------------------------------------- %sum(req_1_TSL_DL+2*req_2_TSL_DL+3*req_3_TSL_DL +4*req_4_TSL_DL)----------------------------------------------------------------sum(req_1_TSL_DL+req_2_TSL_DL+req_3_TSL_DL+req_4_TSL_DL)


Figure 18. DL multislot allocation %, S9PS (msl_16a)

UL multislot requests, S9PS (msl_17)

Use: Total number of multislot requests in UL.



Figure 19. UL multislot requests, S9PS (msl_17)

DL multislot requests, S9PS (msl_18)

Use: Total number of multislot requests in DL.



Figure 20. DL multislot requests, S9PS (msl_18)

2.3 TBF (tbf)

Average number of LLC blocks per UL TBF, S9PS (tbf_3)

Use: Indicates the average number of LLC data blocks pernormally released TBF.



sum(Ave_UL_LLC_per_TBF_sum)----------------------------sum(Ave_UL_LLC_per_TBF_den)


Figure 21. Average number of LLC blocks per UL TBF, S9PS (tbf_3)

Average number of LLC blocks per DL TBF, S9PS (tbf_4)

Use: Indicates the average number of LLC data blocks pernormally released TBF.

sum(Ave_DL_LLC_per_TBF_sum)----------------------------sum(Ave_DL_LLC_per_TBF_den)


Unit: second

Figure 22. Average number of LLC blocks per DL TBF, S9PS (tbf_4)

Average UL TBF duration, S9PS (tbf_5)

Known problems: The unit has changed to 10 ms in BSC CD 1.2. After thattbf_5a is needed.

sum(ave_dur_ul_tbf_sum)------------------------sum(ave_dur_ul_tbf_den)


Unit: second

Figure 23. Average UL TBF duration, S9PS (tbf_5)

Average UL TBF duration, S9PS (tbf_5a)

Use: Counted from the normally released TBFs.Known problems: Contains part of TBF establishment delays.

sum(ave_dur_ul_tbf_sum)/100----------------------------sum(ave_dur_ul_tbf_den)




Unit: second

Figure 24. Average UL TBF duration, S9PS (tbf_5a)

Average DL TBF duration, S9PS (tbf_6a)

Use: Counted from the normally released TBFs.Known problems: Contains part of TBF establishment delays.

sum(ave_dur_dl_tbf_sum)/100---------------------------sum(ave_dur_dl_tbf_den)


Unit: second

Figure 25. Average DL TBF duration, S9PS (tbf_6a)

Average UL TBF duration, unack mode, S9PS (tbf_7)

sum(ave_dur_ul_tbf_unack_mode_sum/100)-------------------------------------sum(ave_dur_ul_tbf_unack_mode_den)


Unit: second

Figure 26. Average UL TBF duration, unack mode, S9PS (tbf_7)

Average DL TBF duration, unack mode, S9PS (tbf_8)

sum(ave_dur_dl_tbf_unack_mode_sum/100)--------------------------------------sum(ave_dur_dl_tbf_unack_mode_den)


Unit: second

Figure 27. Average DL TBF duration, unack mode, S9PS (tbf_8)



UL mlslot allocation blocking, S9PS (tbf_15)

Use: If the blocking is met regularly, there is a need either toexpand the territory (CS traffic low) or TCH capacity (CStraffic high).

Note: If the statistics show that there is blocking but no upgraderequests yet, the reason may be that the territory has beensmaller than the default setting defines (CS use). The PCUwill not make an upgrade request. This is because the CS sidewill return the default channels back to the PS territory assoon as the CS load allows that.

sum(NO_RADIO_RES_AVA_UL_TBF)100 * ------------------------------------------------------------------------- %



Figure 28. UL mlslot allocation blocking, S9PS (tbf_15)

DL mlslot allocation blocking, S9PS (tbf_16)

Use: If the blocking is met regularly, there is a need either toexpand the territory (CS traffic low) or TCH capacity (CStraffic high).

Note: If the statistics show that there is blocking but no upgraderequests yet, the reason may be that the territory has beensmaller than the default setting defines (CS use). The PCUwill not make an upgrade request. This is because the CS sidewill return the default channels back to the PS territory assoon as the CS load allows that.

sum(NO_RADIO_RES_AVA_DL_TBF)100 * ------------------------------------------------------------------------- %



Figure 29. DL mlslot allocation blocking, S9PS (tbf_16)

UL TBF releases due to CS traffic %, S9PS (tbf_19)

Use: In GPRS the CS takes priority over the radio interfaceresources outside the dedicated territory. This KPI indicatesthe impact of CS traffic on PS (TBF drops) when radiointerface capacity (TCH) is not sufficient and CS traffic takesthe capacity from PS traffic by force.

sum(UL_TBF_rel_due_CSW_traffic)100 * -------------------------------- %

sum(Nbr_of_UL_TBF)




Figure 30. UL TBF releases due to CS traffic %, S9PS (tbf_19)

DL TBF releases due to CS traffic %, S9PS (tbf_20)

Use: In GPRS the CS takes priority over the radio interfaceresources outside the dedicated territory. This KPI indicatesthe impact of CS traffic on PS (TBF drops) when radiointerface capacity (TCH) is not sufficient and CS traffic takesthe capacity from PS traffic by force.

sum(DL_TBF_rel_due_CSW_traffic)100 * ---------------------------------- %

sum(Nbr_of_DL_TBF)


Figure 31. DL TBF releases due to CS traffic %, S9PS (tbf_20)

UL drops per 10 Kbytes, MS lost, S9PS (tbf_27a)

Use: This KPI tries to build something similar that we are used toseeing in the CS side to indicate the bad radio conditions thataffect the connection i.e. drops in ratio to volume.

Known problems: - Slow cell reselections (flushes) can increment thenumerator.- Combining TBF related information (numerator) to RLCblock information (numerator) causes problems because thebehaviour of TBFs is quite independent of the RLC blocks(TBF length varies strongly).- 1000 used for 1 kbyte instead of 1024.

sum(UL_TBF_rel_due_no_resp_MS)------------------------------------------------------------------sum(rlc_data_blocks_ul_cs1*20 + rlc_data_blocks_ul_cs2*30 )/ 10000


Figure 32. UL drops per 10 Kbytes, MS lost, S9PS (tbf_27a)



UL drops per 10 Kbytes, MS lost, S9PS (tbf_27b)

Use: This KPI tries to build something similar that we are used toseeing in the CS side to indicate the bad radio conditions thataffect the connection i.e. drops in ratio to volume.

Known problems: - Slow cell reselections (flushes) can increment thenumerator.- Combining TBF related information (numerator) to RLCblock information (numerator) causes problems because thebehaviour of TBFs is quite independent of the RLC blocks(TBF length varies strongly).-In S10 the EGPRS modulation coding scheme counters haveto be added to the denominator.

sum(UL_TBF_rel_due_no_resp_MS)------------------------------------------------------------------sum(rlc_data_blocks_ul_cs1*20 + rlc_data_blocks_ul_cs2*30 )/ 10240


Unit: drops per 10Kbytes

Figure 33. UL drops per 10 Kbytes, MS lost, S9PS (tbf_27b)

UL drops per 10 Kbytes, MS lost, S10.5PS (tbf_27c)

Use: To indicate the bad radio conditions that affect the connectioni.e. drops in ratio to volume.

Known problems: - Slow cell reselections (flushes) can increment thenumerator.- Combining TBF related information (numerator) to RLCblock information (numerator) causes problems, because thebehaviour of TBFs is quite independent of the RLC blocks(TBF length varies strongly).

sum(UL_TBF_rel_due_no_resp_MS)------------------------------------------------------------(sum(rlc_data_blocks_ul_cs1*20 + rlc_data_blocks_ul_cs2*30 )

+ sum over MCS-1 (xx)* 22+ sum over MCS-2 (xx)* 28+ sum over MCS-3 (xx)* 37+ sum over MCS-4 (xx)* 44+ sum over MCS-5 (xx)* 56+ sum over MCS-6 (xx)* 74+ sum over MCS-7 (xx/2)*112+ sum over MCS-8 (xx/2)*136+ sum over MCS-9 (xx/2)*148))/ 10240

Where xx = (UL_RLC_BLOCKS_IN_ACK_MODE + UL_RLC_BLOCKS_IN_UNACK_MODE)

Counters from table(s):p_nbsc_packet_control_unitp_nbsc_coding_scheme




Figure 34. UL drops per 10 Kbytes, MS lost, S10.5PS (tbf_27c)

DL drops per 10 Kbytes, MS lost, S9PS (tbf_28a)

Use: See tbf_27bKnown problems: - Slow cell reselections (flushes) can increment the

numerator.- Combining TBF related information (numerator) to RLCblock info (numerator) causes problems because thebehaviour of TBFs is quite independent of the RLC blocks(TBF length varies strongly).- In S10 the EGPRS modulation coding scheme counters haveto be added to the denominator.- 1000 used for 1 kbyte instead of 1024.

sum(DL_TBF_rel_due_no_resp_MS)-------------------------------------------------------------------sum(rlc_data_blocks_dl_cs1*20 + rlc_data_blocks_dl_cs2*30) / 10000


Figure 35. DL drops per 10 Kbytes, MS lost, S9PS (tbf_28a)

DL drops per 10 Kbytes, MS lost, S9PS (tbf_28b)

Use: See tbf_27b.Known problems: - Slow cell reselections (flushes) can increment the

numerator.- Combining TBF related info (numerator) to RLC block info(numerator) causes problems because the behaviour of TBFsis quite independent of the RLC blocks (TBF length variesstrongly).- In S10 the EGPRS modulation coding scheme counters haveto be added to the denominator.

sum(DL_TBF_rel_due_no_resp_MS)------------------------------------------------------------------sum(rlc_data_blocks_dl_cs1*20 + rlc_data_blocks_dl_cs2*30 )/ 10240


Unit: Drops per 10Kbytes

Figure 36. DL drops per 10 Kbytes, MS lost, S9PS (tbf_28b)



DL drops per 10 Kbytes, MS lost, S10.5PS (tbf_28c)

Use: See tbf_27b.Known problems: - Slow cell reselections (flushes) can increment the

numerator.- Combining TBF related info (numerator) to RLC block info(numerator) causes problems because the behaviour of TBFsis quite independent of the RLC blocks (TBF length variesstrongly).

sum(DL_TBF_rel_due_no_resp_MS)----------------------------------------

(sum(rlc_data_blocks_dl_cs1*20+ rlc_data_blocks_dl_cs2*30+ sum over MCS-1 (yy)* 22+ sum over MCS-2 (yy)* 28+ sum over MCS-3 (yy)* 37+ sum over MCS-4 (yy)* 44

+ sum over MCS-5 (yy)* 56+ sum over MCS-6 (yy)* 74+ sum over MCS-7 (yy/2)*112+ sum over MCS-8 (yy/2)*136+ sum over MCS-9 (yy/2)*148)) / 10240

Where yy = (DL_RLC_BLOCKS_IN_ACK_MODE + DL_RLC_BLOCKS_IN_UNACK_MODE)

Counters from table(s):p_nbsc_packet_control_unitp_nbsc_coding_schemes


Figure 37. DL drops per 10 Kbytes, MS lost, S10.5PS (tbf_28c)

UL TBF reallocation failure ratio, S9PS (tbf_29)

sum(UL_TBF_REALLOC_FAILS)100 * ------------------------------------------------- %

sum(UL_TBF_RE_ALLOCATIONS + UL_TBF_REALLOC_FAILS)


Figure 38. UL TBF reallocation failure ratio, S9PS (tbf_29)

DL TBF reallocation failure ratio, S9PS (tbf_30)

sum(DL_TBF_REALLOC_FAILS)100 * ------------------------------------------------- %

sum(DL_TBF_RE_ALLOCATIONS + DL_TBF_REALLOC_FAILS)




Figure 39. DL TBF reallocation failure ratio, S9PS (tbf_30)

UL TBF reallocation attempts, S9PS (tbf_31)

sum(UL_TBF_RE_ALLOCATIONS + UL_TBF_REALLOC_FAILS)


Figure 40. UL TBF reallocation attempts, S9PS (tbf_31)

DL TBF reallocation attempts, S9PS (tbf_32)

sum(DL_TBF_RE_ALLOCATIONS + DL_TBF_REALLOC_FAILS)


Figure 41. DL TBF reallocation attempts, S9PS (tbf_32)

TBF success % S9PS (tbf_34)

Use: Also called ‘TBF retainability’.This KPI is used to measure the quality of the radio interfacefor TBF sessions. The KPI measures purely the retainabilityof TBFs and is not dependent on interfaces and NEs outsideBSS.

Known problems: Not usable on BTS level in S10 if common a BCCH is usedbecause TBF can start in one BTS of a segment and end inanother.

100 - TBF failure % =

TBF establishments -Normal TBF releases- releases due to flush releases due to suspend

100 - 100 * -------------------------------------------------------- % =TBF establishments

- releases due to flush releases due to suspend

sum(NBR_OF_UL_TBF + NBR_OF_DL_TBF ;TBF establishments- decode(AVE_DUR_UL_TBF_SUM,0,0,AVE_DUR_UL_TBF_DEN)- decode(AVE_DUR_DL_TBF_SUM,0,0,AVE_DUR_DL_TBF_DEN)- UL_TBF_REL_DUE_TO_FLUSH - DL_TBF_REL_DUE_TO_FLUSH- UL_TBF_REL_DUE_TO_SUSPEND - DL_TBF_REL_DUE_TO_SUSPEND)

100 - 100 * ----------------------------------------------------------- %sum(NBR_OF_UL_TBF + NBR_OF_DL_TBF

- UL_TBF_REL_DUE_TO_FLUSH - DL_TBF_REL_DUE_TO_FLUSH- UL_TBF_REL_DUE_TO_SUSPEND - DL_TBF_REL_DUE_TO_SUSPEND)




Figure 42. TBF success % S9PS (tbf_34)

UL TBF releases due to flush %, S9PS (tbf_35)

Use: This KPI indicates that there is mobility from this cell to othercells by cell reselection. Cell reselection affects thethroughput.

sum(UL_TBF_REL_DUE_TO_FLUSH)100 * -------------------------------- %

sum(Nbr_of_UL_TBF)


Figure 43. UL TBF releases due to flush %, S9PS (tbf_35)

DL TBF releases due to flush %, S9PS (tbf_36)

Use: This KPI indicates that there is mobility from this cell to othercells by cell reselection. Cell reselection affects thethroughput.

sum (DL_TBF_REL_DUE_TO_FLUSH)100 * ---------------------------------- %

sum(Nbr_of_DL_TBF)


Figure 44. DL TBF releases due to flush %, S9PS (tbf_36)

Average UL TBF per timeslot, S9PS (tbf_37b)

Use: Indicates how many UL TBFs on average there are pertimeslot.

sum(aver_tbfs_per_tsl_ul_sum)--------------------------------------sum(aver_tbfs_per_tsl_ul_den)


Figure 45. Average UL TBF per timeslot, S9PS (tbf_37b)



Average UL TBF per timeslot, S9PS (tbf_37c)

Use: Indicates how many UL TBFs there are on average pertimeslot.

Known problems: The implementation of the counters is wrong. Correctionpending.

sum(aver_tbfs_per_tsl_ul_sum)--------------------------------------sum(aver_tbfs_per_tsl_ul_den) * 100


Figure 46. Average UL TBFper timeslot, S9PS (tbf_37c)

Average DL TBF per timeslot, S9PS (tbf_38b)

Use: Indicates how many DL TBFs on average there are pertimeslot.

sum(aver_tbfs_per_tsl_dl_sum)-------------------------------------sum(aver_tbfs_per_tsl_dl_den)


Figure 47. Average DL TBF per timeslot, S9PS (tbf_38b)

Average DL TBF per timeslot, S9PS (tbf_38c)

Use: Indicates how many DL TBFs there are on average pertimeslot.

Known problems: The implementation of the counters is wrong. Correctionpending.

sum(aver_tbfs_per_tsl_dl_sum)----------------------------------------sum(aver_tbfs_per_tsl_dl_den) * 100


Figure 48. Average DL TBFper timeslot, S9PS (tbf_38c)

UL GPRS TBF establishments, S10.5PS (tbf_41)

Use: Indicates only GPRS TBF establishments. EGPRSestablishments are not counted.

Sum (NBR_OF_UL_TBF - EGPRS_TBFS_UL)




Unit: Numbers

Figure 49. UL GPRS TBF establishments, S10.5PS (tbf_41)

DL GPRS TBF establishments, S10.5PS (tbf_42)

Use: Indicates only GPRS TBF establishments. EGPRSestablishments are not counted.

Sum (NBR_OF_DL_TBF - EGPRS_TBFS_DL)


Unit: Numbers

Figure 50. DL GPRS TBF establishments, S10.5PS (tbf_42)

Normal TBF release ratio, DL to UL, S10.5PS (tbf_44)

Use: Ratio of normally released TBFs in DL to UL. To detect therelative activity in DL compared to UL, used in findingsleeping GPRS cells.

(100*(AVE_DUR_DL_TBF_DEN/AVE_DUR_UL_TBF_DEN))


Unit: %

Figure 51. Normal TBF release ratio DL, to UL, S10.5PS (tbf_44)

2.4 LLC (llc)

Expired LLC frames % DL, S9PS (llc_1)

Use: Ratio of expired LLC frames to all frames. Indicatesthroughput problems in SGSN or in the network under it. Thelifetime of the packets is set by SGSN and that may expirealready in SGSN. If a packet is sent to PCU, the remaininglifetime is the time when it should be sent further in PCU. Ifthe lifetime expires, the packet is discarded.

sum(disc_llc_blocks_due_to_exp)



100 * ------------------------------------------------------- %sum(ave_dl_llc_per_tbf_sum+ disc_llc_blocks_due_to_exp)


Figure 52. Expired LLC frames % DL, S9PS (llc_1)

Discarded UL LLC frames, NSE unavailability %, S9PS (llc_2)

Use: Ratio of discarded LLC bytes to UL RLC data bytes. Bytesdiscarded due to unavailable NSE which may mean problemsin the Gb interface.

sum(disc_UL_LLC_data)100 * --------------------------------------------------------- %

sum(RLC_data_blocks_UL_CS1*20 +RLC_data_blocks_UL_CS2*30)


Figure 53. Discarded UL LLC frames, NSE unavailability %, S9PS (llc_2)

2.5 RLC (rlc)

Ack. CS1 RLC blocks UL, S9PS (rlc_1)

Use: Number of UL blocks in RLC ack mode using CS1.Retransmission is not included.

sum(RLC_DATA_BLOCKS_UL_CS1 - RLC_DATA_BLOCKS_UL_UNACK)


Figure 54. Ack. CS1 RLC blocks UL, S9PS (rlc_1)

Ack. CS1 RLC blocks DL, S9PS (rlc_2)

Use: Number of DL blocks in RLC ack mode using CS1.Retransmission is not included.

sum(RLC_DATA_BLOCKS_DL_CS1 - RLC_DATA_BLOCKS_DL_UNACK)


Figure 55. Ack. CS1 RLC blocks DL, S9PS (rlc_2)



Ack. CS1 RLC DL block error rate, S9PS (rlc_3a)

Use: Number of DL blocks in RLC ack mode using CS1.

sum(BAD_FRAME_IND_UL_CS1 - BAD_FRAME_IND_UL_UNACK )100 * -------------------------------------------------- %

sum(RLC_DATA_BLOCKS_DL_CS1- RLC_DATA_BLOCKS_DL_UNACK+ BAD_FRAME_IND_UL_CS1- BAD_FRAME_IND_UL_UNACK)


Figure 56. Ack. CS1 RLC DL block error rate, S9PS (rlc_3a)

Unack. CS1 RLC UL block error rate, S9PS (rlc_4a)

sum(BAD_FRAME_IND_UL_UNACK)100 * ------------------------------------------------------ %

sum(RLC_DATA_BLOCKS_UL_UNACK + BAD_FRAME_IND_UL_UNACK)


Figure 57. Unack. CS1 RLC UL block error rate, S9PS (rlc_4a)

Ack. CS1 RLC UL block error rate, S9PS (rlc_5a)

sum(BAD_FRAME_IND_UL_CS2)100 * -------------------------------------------------- %

sum(RLC_DATA_BLOCKS_UL_CS2 + BAD_FRAME_IND_UL_CS2)


Figure 58. Ack. CS1 RLC UL block error rate), S9PS (rlc_5a)

UL CS1 RLC data share, S9PS (rlc_6a)

Use: Indicates how big a share as a percentage UL CS1 datacomprises out of all RLC payload data.

sum(RLC_data_blocks_UL_CS1*20)100 * ----------------------------------------------------------%

sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30+ RLC_data_blocks_DL_CS1*20+RLC_data_blocks_DL_CS2*30)




Unit: %

Figure 59. UL CS1 RLC data share, S9PS (rlc_6a)

UL CS1 ack RLC data share, S9PS (rlc_6b)

Use: Indicates how big a share as a percentage UL CS1 ack datacomprises out of all RLC payload data.

sum(RLC_data_blocks_UL_CS1-RLC_data_blocks_UL_UNACK)*20100 * ----------------------------------------------------------%



Unit: %

Figure 60. UL CS1 ack RLC data share, S9PS (rlc_6b)

UL CS1 unack RLC data share, S9PS (rlc_6c)

Use: Indicates how big a share as a percentage DL CS1 unack datacomprises out of all RLC payload data.

sum(RLC_data_blocks_UL_UNACK)*20100 * ----------------------------------------------------------%



Figure 61. UL CS1 unack RLC data share, S9PS (rlc_6c)

UL CS2 RLC data share, S9PS (rlc_7a)

Use: Indicates how big a share as a percentage UL CS2 datacomprises out of all RLC payload data.

sum(RLC_data_blocks_UL_CS2*30)100 * ----------------------------------------------------------%





Unit: %

Figure 62. UL CS2 RLC data share, S9PS (rlc_7a)

DL CS1 RLC data share, S9PS (rlc_8a)

Use: Indicates how big a share as a percentage DL CS1 datacomprises out of all RLC payload data.

sum(RLC_data_blocks_DL_CS1*20)100 * ----------------------------------------------------------%



Unit: %

Figure 63. DL CS1 RLC data share, S9PS (rlc_8a)

DL CS1 ack RLC data share, S9PS (rlc_8b)

Use: Indicates how big a share as a percentage DL CS1 ack datacomprises out of all RLC payload data.

sum(RLC_data_blocks_DL_CS1 - RLC_data_blocks_DL_UNACK)*20100 * ----------------------------------------------------------%

sum(RLC_data_blocks_UL_CS1*20 + RLC_data_blocks_UL_CS2*30+ RLC_data_blocks_DL_CS1*20 + RLC_data_blocks_DL_CS2*30)


Unit: %

Figure 64. DL CS1 ack RLC data share, S9PS (rlc_8b)

DL CS1 unack RLC data share, S9PS (rlc_8c)

Use: Indicates how big a share as a percentage DL CS1 unack datacomprises out of all RLC payload data.

sum(RLC_data_blocks_DL_UNACK)*20100 * ---------------------------------------------------------- %





Unit: %

Figure 65. DL CS1 unack RLC data share, S9PS (rlc_8c)

DL CS2 RLC data share, S9PS (rlc_9a)

Use: Indicates how big a share as a percentage DL CS2 datacomprises out of all RLC payload data.

sum(RLC_data_blocks_DL_CS2*30)100 * ----------------------------------------------------------%



Unit: %

Figure 66. DL CS2 RLC data share, S9PS (rlc_9a)

UL CS1 RLC block error rate, S9PS (rlc_10a)

Use: High BLER means worse radio interface conditions.Known problems: 1) The number of ignored RLC data blocks in uplink due to

BSN being in acknowledged mode should be subtracted butthere is no counter for this specifically for CS1 There is onlyone counter for CS1 and CS2 which is marked as counter xxin the formula.2) Does not show correctly if there is unack mode used.(rlc_data_blocks_ul_cs1 contains both ack and unack)

Experiences on use: UL block error rate (BLER) is normally higher than DLBLER. There can be several reasons to this:1) There is UL power control while full power is used in DL.The UL PC parameters may have been set too aggressively.2) UL BLER includes uplink as well as downlinktransmission problems (the MS needs to decode the USFcorrectly before transmitting).



3) When the MS stops responding during an UL TBF (e.g. dueto a cell change), a number of ‘bad frames’ is received by thePCU before it detects that the radio contact has been lost.These bad frames increase the counters c072070 or c072071,and thus the UL BLER.4) Also when resources are allocated for an UL TBF duringone phase access, but the MS does not respond, bad frames arereceived. Such an allocation without any blocks may occur,for example, if the MS has sent the channel request more thanonce during the access, or if there is a collision, i.e. thenetwork initiates DL TBF establishment at the same time, theMS reacts on that and ignores the UL TBF.

sum(bad_frame_ind_UL_CS1)100 * -------------------------------------------------%

sum(rlc_data_blocks_UL_CS1+ bad_frame_ind_UL_CS1+xx)

xx = RLC CS1 blocks ignored due to incorrect BSN (missing counter approximated)

sum(rlc_data_blocks_ul_cs1)= ------------------------------- *sum(ignor_rlc_data_bl_ul_due_bsn)

sum(rlc_data_blocks_ul_cs1+ rlc_data_blocks_ul_cs2)


Figure 67. UL CS1 RLC block error rate, S9PS (rlc_10a)

UL CS1 ACK RLC block error rate, S9PS (rlc_10b)


BSN being in acknowledged mode should be subtracted butthere is no counter for this specifically for CS1 There is onlyone counter for CS1 and CS2 which is marked as counter xxin the formula.2) Does not show correctly if there is unack mode used.(rlc_data_blocks_ul_cs1 contains both ack and unack)

Experiences on use: UL block error rate (BLER) is normally higher than DLBLER. There can be several reasons to this:1) There is UL power control while full power is used in DL.The UL PC parameters may have been set too aggressively.2) UL BLER includes uplink as well as downlinktransmission problems (the MS needs to decode the USFcorrectly before transmitting).



3) When the MS stops responding during an UL TBF (e.g. dueto a cell change), a number of ’bad frames’ is received by thePCU before it detects that the radio contact has been lost.These bad frames increase the counters c072070 or c072071,and thus the UL BLER.4) Also when resources are allocated for an UL TBF duringone phase access, but the MS does not respond, bad frames arereceived. Such an allocation without any blocks may occur,for example, if the MS has sent the channel request more thanonce during the access, or if there is a collision, i.e. thenetwork initiates DL TBF establishment at the same time, theMS reacts on that and ignores the UL TBF.

sum(bad_frame_ind_UL_CS1)100 * --------------------------------------------------------------------%

sum(rlc_data_blocks_UL_CS1- rlc_data_blocks_UL_unack !ack CS1 data blocks+ bad_frame_ind_UL_CS1+ xx !RLC CS1 blocks ignored due to incorrect BSN (estimate)

where

xx =

sum(rlc_data_blocks_ul_cs1- rlc_data_blocks_UL_unack)

xx= ----------------------------- *sum(ignor_rlc_data_bl_ul_due_bsn)sum(rlc_data_blocks_ul_cs1

- rlc_data_blocks_UL_unack+ rlc_data_blocks_ul_cs2)


Figure 68. UL CS1 RLC block error rate, S9PS (rlc_10b)

UL CS1 ACK RLC block error rate, S9PS (rlc_10c)

Use: High BLER means worse radio interface conditions.Known problems: 1)

SPARE72107=RETRA_RLC_DATA_BLOCKS_UL_CS1.The spare counter will be replaced by an official counter lateron.2) Impact of "Ignore due to BSN" not considered.3) See also rlc_10b.

Experiences on use: See rlc_10a.

sum(SPARE072107)100 * -----------------------------------------%

sum(rlc_data_blocks_UL_CS1)




Figure 69. UL CS1 ACK RLC block error rate, S9PS (rlc_10c)

UL CS2 ARLC block error rate, S9PS (rlc_11a)


BSN being in acknowledged mode should be subtracted butthere is no counter for this specifically for CS2. There is onlyone counter for CS1 and CS2 the estimated value of which iscounted as xx below.2) Does not work if unack mode is used, too.

Experiences on use: See rlc_10a.

sum(bad_frame_ind_ul_cs2)100 * -------------------------------------------------%

sum(rlc_data_blocks_ul_cs2+ bad_frame_ind_ul_cs2+ xx)

xx = RLC CS2 blocks ignored due to incorrect BSN (missing counter approximated)

sum(rlc_data_blocks_ul_cs2)= -----------------------------------------------------* sum(ignor_rlc_data_bl_ul_due_bsn)

sum(rlc_data_blocks_ul_cs1 + rlc_data_blocks_ul_cs2)


Figure 70. UL CS2 ARLC block error rate, S9PS (rlc_11a)

UL CS2 ACK RLC block error rate, S9PS (rlc_11b)

Use: High BLER means worse radio interface conditions.Known problems: 1) The number of ignored RLC data blocks in downlink due

to BSN being in acknowledged mode should be subtracted butthere is no counter for this separately for CS2. There is onlyone counter for CS1 and CS2, the estimated value of which iscounted as xx below.

Experience on use: See rlc_10a

sum(bad_frame_ind_ul_cs2)100 * ---------------------------------------------- %

sum(rlc_data_blocks_ul_cs2- rlc_data_blocks_UL_unack ! CS2 ack blocks+ bad_frame_ind_ul_cs2+ xx ! RLC CS2 blocks ignored due to incorrect BSN (estimated))

where

sum(rlc_data_blocks_ul_cs2)xx = ------------------------------ *sum(ignor_rlc_data_bl_ul_due_bsn)



sum(rlc_data_blocks_ul_cs1--rlc_data_blocks_UL_unack+rlc_data_blocks_ul_cs2)


Figure 71. UL CS2 RLC block error rate, S9PS (rlc_11b)

UL CS2 ACK RLC block error rate, S9PS (rlc_11c)


BSN being in acknowledged mode should be subtracted butthere is no counter for this separately for CS2. There is onlyone counter for CS1 and CS2, the estimated value of which iscounted as xx below.

Experiences on use: See rlc_10b.

sum(bad_frame_ind_ul_cs2)100 * ----------------------------------------------------------------- %

sum(rlc_data_blocks_ul_cs2 + bad_frame_ind_ul_cs2+ xx) ! RLC CS2 blocks ignored due to incorrect BSN estimated

where

sum(rlc_data_blocks_ul_cs2)xx = ----------------------------- *sum(ignor_rlc_data_bl_ul_due_bsn)

sum(rlc_data_blocks_ul_cs1- rlc_data_blocks_UL_unack+ rlc_data_blocks_ul_cs2)


Figure 72. UL CS2 ACK RLC block error rate, S9PS (rlc_11c)

UL CS2 ACK RLC block error rate, S10.5PS(rlc_11d)

Use: High BLER means worse radio interface conditions.Known problems: 1)SPARE72108=RETRA_RLC_DATA_BLOCKS_UL_CS.

The spare counter will be replaced by an official counter lateron.Impact of blocks “Ignored due to BSN” not covered.

Experience on use: See rlc_10b.

sum(SPARE072108)100 * -------------------------------------------%

sum(rlc_data_blocks_ul_cs2)




Figure 73. UL CS2 ACK RLC block error rate, S10.5PS(rlc_11d)

DL CS1 RLC block error rate, S9PS (rlc_12)

Use: High BLER means worse radio interface conditions.Known problems: Does not work if the unack mode is used, too.

sum(retra_rlc_data_blocks_dl_cs1)100 * ---------------------------------------------------------- %

sum(rlc_data_blocks_dl_cs1 + retra_rlc_data_blocks_dl_cs1)


Figure 74. DL CS1 RLC block error rate, S9PS (rlc_12)

DL CS1 ACK RLC block error rate, S9PS (rlc_12a)

Use: High BLER means worse radio interface conditions.

sum(retra_rlc_data_blocks_dl_cs1)100 * --------------------------------------------------------------- %

sum(rlc_data_blocks_dl_cs1 - rlc_data_blocks_dl_unack ; ack CS1+ retra_rlc_data_blocks_dl_cs1)


Figure 75. DL CS1 ACK RLC block error rate, S9PS (rlc_12a)

DL CS2 RLC block error rate, S9PS (rlc_13)

Use: High BLER means worse radio interface conditions.

sum(retra_rlc_data_blocks_dl_cs1)100 * --------------------------------------------------------------- %

sum(rlc_data_blocks_dl_cs1 - rlc_data_blocks_dl_unack ; ack CS1+ retra_rlc_data_blocks_dl_cs1)


Figure 76. DL CS2 RLC block error rate, S9PS (rlc_13)

UL RLC blocks, S9PS (rlc_14)

Use: Total UL data volume as the number of RLC blocks.



sum(rlc_data_blocks_ul_cs1 + rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul + bad_frame_ind_ul_cs1+ bad_frame_ind_ul_cs2 + bad_frame_ind_ul_unack+ ignor_rlc_data_bl_ul_due_bsn)


Figure 77. UL RLC blocks, S9PS (rlc_14)

DL RLC blocks, S9PS (rlc_15)

Use: Total DL data volume as the number of RLC blocks.

sum(rlc_data_blocks_dl_cs1 + rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl + retra_rlc_data_blocks_dl_cs1+ retra_rlc_data_blocks_dl_cs2)


Figure 78. DL RLC blocks, S9PS (rlc_15)

UL ACK EGPRS block error ratio S10.5PS (rlc_18)

Use: Ratio of retransmitted UL blocks to all blocks sent in EGPRScoding schemes.

Sum over MCS1 to 9 (RETRANS_RLC_DATA_BLOCKS_UL)100 * --------------------------------------------------------------------------

Sum over MCS1 to 9 (UL_RLC_BLOCKS_IN_ACK_MODE + RETRANS_RLC_DATA_BLOCKS_UL)

Counters from table(s):p_nbsc_coding_scheme

Unit: %

Figure 79. UL ACK EGPRS block error ratio S10.5PS (rlc_18)

DL ACK EGPRS block error ratio S10.5PS (rlc_19)

Use: Ratio of retransmitted DL blocks to all blocks sent in EGPRScoding schemes.

Sum over MCS1 to 9 (RETRANS_RLC_DATA_BLOCKS_DL)100 * --------------------------------------------------------------------------

Sum over MCS1 to 9 (DL_RLC_BLOCKS_IN_ACK_MODE + RETRANS_RLC_DATA_BLOCKS_DL)




Unit: %

Figure 80. DL ACK EGPRS block error ratio S10.5PS (rlc_19)

UL ACK EGPRS block error ratio MCS-n, S10.5PS (rlc_20)

Use: Used for UL ACK EGPRS block error ratio of any MCS 1 to9.

Sum over MCSn (RETRANS_RLC_DATA_BLOCKS_UL)100 * ----------------------------------------------------------------------

Sum over MCSn (UL_RLC_BLOCKS_IN_ACK_MODE + RETRANS_RLC_DATA_BLOCKS_UL)

where n can be from 1 to 9


Unit: %

Figure 81. UL ACK EGPRS block error ratio MCS-n, S10.5PS (rlc_20)

DL ACK EGPRS block error ratio MCS-n, S10.5PS (rlc_21)

Use: For DL ACK EGPRS block error ratio of any MCS 1 to 9.

Sum over MCSn (RETRANS_RLC_DATA_BLOCKS_DL)100 * ----------------------------------------------------------------------

Sum over MCSn (DL_RLC_BLOCKS_IN_ACK_MODE + RETRANS_RLC_DATA_BLOCKS_DL)



Unit: %

Figure 82. DL ACK EGPRS block error ratio MCS-n, S10.5PS (rlc_21)

UL ACK RLC data share MCS-n, S10.5PS (rlc_22)

Use: Indicates how big a share as a percentage UL ACK RLC datacomprises out of all RLC payload data.

Sum over MCS-n (UL_RLC_BLOCKS_IN_ACK_MODE)100* --------------------------------------------

Sum over MCS-n (UL_RLC_BLOCKS_IN_ACK_MODE+ UL_RLC_BLOCKS_IN_UNACK_MODE+ DL_RLC_BLOCKS_IN_ACK_MODE+ DL_RLC_BLOCKS_IN_UNACK_MODE)





p_nbsc_coding_scheme

Unit: %

Figure 83. UL ACK RLC data share MCS-n, S10.5PS (rlc_22)

UL UNACK RLC data share MCS-n, S10.5PS (rlc_23)

Use: Indicates how big a share as a percentage UL UNACK RLCdata comprises out of all RLC payload data.

Sum over MCS-n (UL_RLC_BLOCKS_IN_UNACK_MODE)100* ----------------------------------------------




Unit: %

Figure 84. UL UNACK RLC data share MCS-n, S10.5PS (rlc_23)

DL ACK RLC data share MCS-n, S10.5PS (rlc_24)

Use: Indicates how big a share as a percentage DL ACK RLC datacomprises out of all RLC payload data.

Sum over MCS-n (DL_RLC_BLOCKS_IN_ACK_MODE)100* --------------------------------------------

Sum over MCS-n (UL_RLC_BLOCKS_IN_ACK_MODE+ UL_RLC_BLOCKS_IN_UNACK_MODE+ DL_RLC_BLOCKS_IN_ACK_MODE++ DL_RLC_BLOCKS_IN_UNACK_MODE)



Unit: %

Figure 85. DL ACK RLC data share MCS-n, S10.5PS (rlc_24)

DL UNACK RLC data share MCS-n, S10.5PS (rlc_25)

Use: Indicates how big a share as a percentage DL UNACK RLCdata comprises out of all RLC payload data.

Sum over MCS-n (DL_RLC_BLOCKS_IN_UNACK_MODE)100* ----------------------------------------------






Unit: %

Figure 86. DL UNACK RLC data share MCS-n, S10.5PS (rlc_25)

GMSK RLC data block share, S10.5PS (rlc_39)

Use: Share of RLC blocks with MCS1..4 out of all used MCSs.

Sum over MCS 1..4 (UL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_UL +BAD_RLC_VALID_HDR_UL_UNACK +UL_RLC_BLOCKS_IN_UNACK_MODE +DL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_DL +DL_RLC_BLOCKS_IN_UNACK_MODE)------------------------------- * 100Sum over MCS 1..9 (UL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_UL +BAD_RLC_VALID_HDR_UL_UNACK +UL_RLC_BLOCKS_IN_UNACK_MODE +DL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_DL +DL_RLC_BLOCKS_IN_UNACK_MODE)


Unit: %

Figure 87. GMSK RLC data block share, S10.5PS (rlc_39)

GMSK RLC data share, S10.5PS (rlc_41)

Use: Share of RLC data bytes with MCS1..4 out of all used MCSs.

(sum over MCS-1 (xx)* 22+sum over MCS-2 (xx)* 28+sum over MCS-3 (xx)* 37+sum over MCS-4 (xx)* 44)

------------------------------ * 100(sum over MCS-1 (xx)*30+sum over MCS-2 (xx)*36+sum over MCS-3 (xx)*45+sum over MCS-4 (xx)*52+sum over MCS-5 (xx)*63+sum over MCS-6 (xx)*81+



sum over MCS-7 (xx/2)*123+sum over MCS-8 (xx/2)*147+sum over MCS-9 (xx/2)*159)

where xx =UL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_UL +BAD_RLC_VALID_HDR_UL_UNACK +UL_RLC_BLOCKS_IN_UNACK_MODE +DL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_DL +DL_RLC_BLOCKS_IN_UNACK_MODE


Unit: %

Figure 88. GMSK RLC data share, S10.5PS (rlc_41)

GPRS UL ACK RLC data share, S10.5PS (rlc_42)

Use: Share of GPRS UL ACK RLC data in data for total data.

sum(RLC_data_blocks_UL_CS1 - RLC_data_blocks_UL_UNACK)*20+ sum(RLC_data_blocks_UL_CS2*30

100 * ------------------------------------------------------------ %sum(RLC_data_blocks_UL_CS1*20 + RLC_data_blocks_UL_CS2*30

+ RLC_data_blocks_DL_CS1*20 + RLC_data_blocks_DL_CS2*30)+( sum over MCS-1 (xx)* 22

+ sum over MCS-2 (xx)* 28+ sum over MCS-3 (xx)* 37+ sum over MCS-4 (xx)* 44+ sum over MCS-5 (xx)* 56+ sum over MCS-6 (xx)* 74+ sum over MCS-7 (xx/2)*112+ sum over MCS-8 (xx/2)*136+ sum over MCS-9 (xx/2)*148)

Where xx=(UL_RLC_BLOCKS_IN_ACK_MODE + UL_RLC_BLOCKS_IN_UNACK_MODE

+ DL_RLC_BLOCKS_IN_ACK_MODE + DL_RLC_BLOCKS_IN_UNACK_MODE)


Unit: %

Figure 89. GPRS UL ACK RLC data share, S10.5PS (rlc_42)

GPRS UL UNACK RLC data share, S10.5PS (rlc_43)

Use: Share of GPRS UL UNACK RLC data in data for total data.

sum(RLC_data_blocks_UL_UNACK)*20100 * ----------------------------------------------------------%

sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30+ RLC_data_blocks_DL_CS1*20+RLC_data_blocks_DL_CS2*30)+( sum over MCS-1 (xx)* 22







Unit: %

Figure 90. GPRS UL UNACK RLC data share, S10.5PS (rlc_43)

GPRS DL ACK RLC data share, S10.5PS (rlc_44)

Use: Share of GPRS DL ACK RLC data in data for total data.

sum(RLC_data_blocks_DL_CS1-RLC_data_blocks_DL_UNACK)*20+ sum(RLC_data_blocks_DL_CS2*30

100 * ----------------------------------------------------------%sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30+ RLC_data_blocks_DL_CS1*20+RLC_data_blocks_DL_CS2*30)+( sum over MCS-1 (xx)* 22




Counters from table(s):p_nbsc_packet_control_unitp_nbsc_coding_schemeUnit: %

Figure 91. GPRS DL ACK RLC data share, S10.5PS (rlc_44)

GPRS DL UNACK RLC data share, S10.5PS (rlc_45)

Use: Share of GPRS DL UNACK RLC data in data for total data.

sum(RLC_data_blocks_DL_UNACK)*20100 * ----------------------------------------------------------%

sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30



+ RLC_data_blocks_DL_CS1*20+RLC_data_blocks_DL_CS2*30)+( sum over MCS-1 (xx)* 22




Counters from table(s):p_nbsc_packet_control_unitp_nbsc_coding_schemeUnit: %

Figure 92. GPRS DL UNACK RLC data share, S10.5PS (rlc_45)

EGPRS UL ACK RLC data share, S10.5PS (rlc_46)

Use: Share of EGPRS UL ACK RLC data in total data.

( sum over MCS-1 (yy)* 22+ sum over MCS-2 (yy)* 28+ sum over MCS-3 (yy)* 37+ sum over MCS-4 (yy)* 44+ sum over MCS-5 (yy)* 56+ sum over MCS-6 (yy)* 74+ sum over MCS-7 (yy/2)*112+ sum over MCS-8 (yy/2)*136+ sum over MCS-9 (yy/2)*148)

100 * ----------------------------------------------------------%sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30





Where yy= UL_RLC_BLOCKS_IN_ACK_MODECounters from table(s):p_nbsc_packet_control_unitp_nbsc_coding_schemeUnit: %

Figure 93. EGPRS UL ACK RLC data share, S10.5PS (rlc_46)



EGPRS UL UNACK RLC data share, S10.5PS (rlc_47)

Use: Share of EGPRS UL UNACK RLC data in total data.


100 * ---------------------------------------------------------- %sum(RLC_data_blocks_UL_CS1*20 + RLC_data_blocks_UL_CS2*30

+ RLC_data_blocks_DL_CS1*20 + RLC_data_blocks_DL_CS2*30)+( sum over MCS-1 (xx)* 22


Where xx=(UL_RLC_BLOCKS_IN_ACK_MODE + UL_RLC_BLOCKS_IN_UNACK_MODE+ DL_RLC_BLOCKS_IN_ACK_MODE + DL_RLC_BLOCKS_IN_UNACK_MODE)

Where yy= UL_RLC_BLOCKS_IN_UNACK_MODECounters from table(s):p_nbsc_packet_control_unitp_nbsc_coding_schemeUnit: %

Figure 94. EGPRS UL UNACK RLC data share, S10.5PS (rlc_47)

EGPRS DL ACK RLC data share, S10.5PS (rlc_48)

Use: Share of EGPRS DL ACK RLC data in total data.









Where yy= DL_RLC_BLOCKS_IN_ACK_MODE


Unit: %

Figure 95. EGPRS DL ACK RLC data share, S10.5PS (rlc_48)

EGPRS DL UNACK RLC data share, S10.5PS (rlc_49)

Use: Share of EGPRS DL UNACK RLC data in total data.







Where yy= DL_RLC_BLOCKS_IN_UNACK_MODE


Unit: %

Figure 96. EGPRS DL UNACK RLC data share, S10.5PS (rlc_49)



2.6 Frame relay (frl)

Kbytes in sent frames, S9PS (frl_1)

Use: Total data volume over all DLCI. Frame relay signalling isdelivered in DLCI 0, so counters do not include thesemessages. This KPI includes NS and BSSGP signalling whichmeans that the counters start to get pegged when these layershave been created.

sum(dlci_1_bytes_sent+dlci_2_bytes_sent+dlci_3_bytes_sent+dlci_4_bytes_sent+dlci_5_bytes_sent)

Counters from table(s):p_nbsc_frame_relayUnit: Kbyte

Figure 97. Kbytes in sent frames, S9PS (frl_1)

Kbytes in received frames, S9PS (frl_2)

Use: Total data volume over all DLCI. Frame relay signalling isdelivered in DLCI 0, so counters do not include thesemessages. This KPI includes NS and BSSGP signalling whichmeans that the counters start to get pegged when these layershave been created.

sum(dlci_1_bytes_rec+dlci_2_bytes_rec+dlci_3_bytes_rec+dlci_4_bytes_rec+dlci_5_bytes_rec)


Figure 98. Kbytes in received frames, S9PS (frl_2)

’Wrong check seq.’ errors per Mbyte, S9PS (frl_3)

Use: Quality indicator of the HDLC layer.

sum(FRMS_WRONG_CHECK_SEQ_)1000 * ---------------------------------------------

sum(DLCI_1_BYTES_REC + DLCI_1_BYTES_DISC_REC+ DLCI_2_BYTES_REC + DLCI_2_BYTES_DISC_REC+ DLCI_3_BYTES_REC + DLCI_3_BYTES_DISC_REC+ DLCI_4_BYTES_REC + DLCI_4_BYTES_DISC_REC+ DLCI_5_BYTES_REC + DLCI_5_BYTES_DISC_REC)




p_nbsc_frame_relayUnit: errors per Mbyte

Figure 99. ‘Wrong check seq.’ errors per Mbyte, S9PS (frl_3)

‘Other’ errors per Mbyte, S9PS (frl_4)

Use: Quality indicator of the HDLC layer.

sum(OTHER_FRAME_ERROR)1000 * ---------------------------------------------

sum(DLCI_1_BYTES_REC + DLCI_1_BYTES_DISC_REC+ DLCI_2_BYTES_REC + DLCI_2_BYTES_DISC_REC+ DLCI_3_BYTES_REC + DLCI_3_BYTES_DISC_REC+ DLCI_4_BYTES_REC + DLCI_4_BYTES_DISC_REC+ DLCI_5_BYTES_REC + DLCI_5_BYTES_DISC_REC)

Counters from table(s):p_nbsc_frame_relayUnit: errors per Mbyte

Figure 100. ‘Other’ errors per Mbyte, S9PS (frl_4)

Bytes in discarded sent frames, S9PS (frl_5)

sum(dlci_1_bytes_sent+dlci_2_bytes_sent+dlci_3_bytes_sent+dlci_4_bytes_sent+dlci_5_bytes_sent)

Counters from table(s):p_nbsc_frame_relay

Figure 101. Bytes in discarded sent frames, S9PS (frl_5)

Bytes in discarded received frames, S9PS (frl_6)

sum(dlci_1_bytes_disc_rec+dlci_2_bytes_disc_rec+dlci_3_bytes_disc_rec+dlci_4_bytes_disc_rec+dlci_5_bytes_disc_rec)


Figure 102. Bytes in discarded received frames, S9PS (frl_6)



Maximum sent load %, S9PS (frl_7)

Use: Indicates the load % of the frame relay bearer for outgoingdata to SGSN.

Known problems: The access rate is taken from configuration data andrepresents only the current setting. This may cause errorswhen used in historical perspective if the settings have beenchanged.

max per bearer_id(8*(dlci_1_bytes_sent

+ dlci_2_bytes_sent+ dlci_3_bytes_sent+ dlci_4_bytes_sent+ dlci_5_bytes_sent)/(period_duration*60))

100 * ------------------------------------------------ %sum per frbc over all unlocked child nsvc

(c_nsvc.committed_info_rate*16);

frbc object_instance = bearer_id in p_nbsc_frame_relay


Figure 103. Maximum sent load %, S9PS (frl_7)

Maximum received load %, S9PS (frl_8)

Use: Indicates the load % of the frame relay bearer for incomingdata from SGSN.

Known problems: The access rate is taken from configuration data andrepresents only the current setting. This may cause errorswhen used in historical perspective if the settings have beenchanged.

max per bearer_id(8*(dlci_1_bytes_rec

+ dlci_2_bytes_rec+ dlci_3_bytes_rec+ dlci_4_bytes_rec+ dlci_5_bytes_rec)/(period_duration*60))

100 * ------------------------------------------------ %sum per frbc over all unlocked child nsvc(c_nsvc.committed_info_rate*16);

frbc object_instance = bearer_id in p_nbsc_frame_relay


Figure 104. Maximum received load %, S9PS (frl_8)

Sent frames, S9PS (frl_9)

sum(dlci_1_sent_frms+dlci_2_sent_frms



+dlci_3_sent_frms+dlci_4_sent_frms+dlci_5_sent_frms)


Figure 105. Sent frames, S9PS (frl_9)

Received frames, S9PS (frl_10)

sum(dlci_1_rec_frms+dlci_2_rec_frms+dlci_3_rec_frms+dlci_4_rec_frms+dlci_5_rec_frms)


Figure 106. Received frames, S9PS (frl_10)

Discarded sent frames, S9PS (frl_11)

sum(dlci_1_disc_sent_frms+dlci_2_disc_sent_frms+dlci_3_disc_sent_frms+dlci_4_disc_sent_frms+dlci_5_disc_sent_frms)


Figure 107. Discarded sent frames, S9PS (frl_11)

Discarded received frames, S9PS (frl_12)

sum(dlci_1_disc_rec_frms+dlci_2_disc_rec_frms+dlci_3_disc_rec_frms+dlci_4_disc_rec_frms+dlci_5_disc_rec_frms)


Figure 108. Discarded received frames, S9PS (frl_12)



Discarded bytes, UL NS-VC congestion S9PS (frl_13a)

sum(dlci_1_disc_ul_ns_udata+dlci_2_disc_ul_ns_udata+dlci_3_disc_ul_ns_udata+dlci_4_disc_ul_ns_udata+dlci_5_disc_ul_ns_udata)

Counters from table(s):p_nbsc_frame_relayUnit: bytes

Figure 109. Discarded bytes, UL NS-VC congestion S9PS (frl_13a)

2.7 HSCSD (hsd)

Throughput ratio, S7HS (hsd_15)

Use: Indicates the percentage of the offered data that is put through.

Sum(96*(ONE_TCH_SEIZ_TIME_9600 + TWO_TCH_SEIZ_TIME_9600+ THREE_TCH_SEIZ_TIME_9600 + FOUR_TCH_SEIZ_TIME_9600)

+ 144*(ONE_TCH_SEIZ_TIME_14400 + TWO_TCH_SEIZ_TIME_14400+ THREE_TCH_SEIZ_TIME_14400 + FOUR_TCH_SEIZ_TIME_14400))

100*-------------------------------------------------------------------- %Sum(96*(ONE_TCH_REQ_TIME_9600 + TWO_TCH_REQ_TIME_9600

+ THREE_TCH_REQ_TIME_9600 + FOUR_TCH_REQ_TIME_9600)+144*(ONE_TCH_REQ_TIME_14400 + TWO_TCH_REQ_TIME_14400

+ THREE_TCH_REQ_TIME_14400 + FOUR_TCH_REQ_TIME_14400))

Figure 110. Throughput ratio, S7HS (hsd_15)

Bps traffic share, S7HS (hsd_49)

Use: Indicates the share of 9600 bps traffic out of all traffic.

Sum(ONE_TCH_SEIZ_TIME_9600+TWO_TCH_SEIZ_TIME_9600+THREE_TCH_SEIZ_TIME_9600+FOUR_TCH_SEIZ_TIME_9600)

100*------------------------------------------------------------ %Sum(ONE_TCH_SEIZ_TIME_9600+TWO_TCH_SEIZ_TIME_9600

+THREE_TCH_SEIZ_TIME_9600+FOUR_TCH_SEIZ_TIME_9600+ONE_TCH_SEIZ_TIME_14400+TWO_TCH_SEIZ_TIME_14400+THREE_TCH_SEIZ_TIME_14400+FOUR_TCH_SEIZ_TIME_14400)

Figure 111. Bps traffic share, S7HS (hsd_49)

Bps traffic share, S7HS (hsd_50)

Use: Indicates the share of share of 14400 bps traffic out of alltraffic.

Sum(ONE_TCH_SEIZ_TIME_14400+TWO_TCH_SEIZ_TIME_14400



+THREE_TCH_SEIZ_TIME_14400+FOUR_TCH_SEIZ_TIME_14400)100*------------------------------------------------------------- %

Sum(ONE_TCH_SEIZ_TIME_9600+TWO_TCH_SEIZ_TIME_9600+THREE_TCH_SEIZ_TIME_9600+FOUR_TCH_SEIZ_TIME_9600+ONE_TCH_SEIZ_TIME_14400+TWO_TCH_SEIZ_TIME_14400+THREE_TCH_SEIZ_TIME_14400+FOUR_TCH_SEIZ_TIME_14400)

Figure 112. Bps traffic share, S7HS (hsd_50)

2.8 Dynamic Abis Pool (dap)

Average usage of DL Dynamic Abis Pool, S10.5PS (dap_1)

Use: Percentage of average usage of DL Dynamic Abis Pool fromthe total amount of sub-timeslots in DL EDAP. Indicates theusage of pool resources.

Sum(AVERAGE_DL_EDAP_USAGE_SUM)-------------------------------Sum(AVERAGE_EDAP_USAGE_DEN)

Counters from table(s):p_nbsc_dynamic_abis

Unit: %

Figure 113. Average usage of DL Dynamic Abis Pool, S10.5PS (dap_1)

Average usage of UL Dynamic Abis Pool, S10.5PS (dap_2)

Use: Percentage of average usage of UL Dynamic Abis Pool fromthe total amount of sub-timeslots in UL EDAP. Indicates theusage of pool resources.

Sum(AVERAGE_UL_EDAP_USAGE_SUM)-------------------------------Sum(AVERAGE_EDAP_USAGE_DEN)


Unit: %

Figure 114. Average usage of UL Dynamic Abis Pool, S10.5PS (dap_2)

Average Available PCM Sub-TSL, S10.5PS (dap_3)

Use: Average available amount of sub-TSLs in EDAP. Indicatestotal sub-TSLs in EDAP.

Sum(TOTAL_PCM_SUBTSLS_IN_EDAP)



-------------------------------Sum(AVERAGE_EDAP_USAGE_DEN)


Unit: sub-TSLs

Figure 115. Average Available PCM Sub-TSL, S10.5PS (dap_3)

2.9 Random access (rach)

Average RACH slot, S1 (rach_1)

Use: Indicates the capacity of BTS for RACH burst handling.Normally shows a constant value because it is dependent onthe BTS configuration which does not often change.

avg(ave_rach_slot/res_acc_denom1)

Counters from table(s):p_nbsc_res_access

Figure 116. Average RACH slot, S1 (rach_1)

Peak RACH load, average, S1 (rach_2)

Use: Indicates the absolute peak value during a measurementperiod. Correlates strongly with UL interference.

Experiences on use: High values may suggest that MSs have problems inaccessing the BTS. High values do not mean high load onSDCCH because SDCCH is needed only if the RACH passesthe detection in BTS.

Known problems: The peak value does not indicate yet how many times therehave been other peaks during the measurement period.

Open questions: How serious the high values really are from the MS point ofview?

avg(peak_rach_load)


Figure 117. Peak RACH load, average, S1 (rach_2)

Peak RACH load %, S1 (rach_3)

Use: This PI indicates how close to full capacity the peak use ofRACH has been during the measurement period.



Experiences on use: It is quite normal that the momentary (peak) load can behigh. Average RACH load is a more meaningful indicator.

max(peak_rach_load)100 * ---------------------------------- %

max(ave_rach_busy/res_acc_denom1)


Figure 118. Peak RACH load %, S1 (rach_3)

Average RACH load %, S1 (rach_4)

Use: This PI indicates how high the RACH load is on average.Experiences on use: If the value is to the order of tens of per cent, there probably

are access problems and MS users get, more often than usual,3 beeps when trying to start calls. The probable reason is ULinterference.

avg(ave_rach_busy/res_acc_denom3)100 * ------------------------------------ %

avg(ave_rach_slot/res_acc_denom1)


Figure 119. Average RACH load %, S1 (rach_4)

Average RACH busy, S1 (rach_5)

Use: This PI indicates roughly the average of the used RACH slots.If the average approaches the ‘average RACH slot’ (rach_1)there probably are access problems and MS users get, moreoften than usual, 3 beeps when trying to start calls.

avg(ave_rach_busy/res_acc_denom3)


Figure 120. Average RACH busy, S1 (rach_5)

RACH rejected due to illegal establishment, S5 (rach_6)

Use: Most of the rejections are ghost accesses. Note that part of theghosts have legal establishment cause and get further toSDCCH.Note that the actual ghost filtering is in BTS.

sum(ghost_ccch_res - rej_seiz_att_due_dist)




Figure 121. RACH rejected due to illegal establishment, S5 (rach_6)

Total RACH rejection ratio, S7 (rach_7)

Use: Ratio of all RACH rejections to total number of channelrequired messages received.Note that the counter ghost_ccch_res contains both ghosts andrejections due to distance checking. The latter one is anoptional feature of BSC.

sum(ghost_ccch_res - rej_seiz_att_due_dist; illegal establ. cause+ bcsu_overload_lower_limit+ bcsu_overload_upper_limit+ bcsu_overload_deleted_rach)

100 * --------------------------------------------------------------------- %sum(ch_req_msg_rec)


Figure 122. Total RACH rejection ratio, S7 (rach_7)

2.10 SDCCH drop failures (sd)

Ghosts detected on SDCCH and other failures, S1 (sd_1)

Use: This part of ghost RACH accesses comprises:- ghosts which have an occasionally valid establishmentcause. These should comprise statistically 5/8 of all ghosts inGSM phase 1. Another 3/8 of ghosts are detected alreadybefore SDCCH based on some invalid establishment cause. InGSM2 the ratio 5/8 and 3/8 is no longer valid.- multiple seizures of SDCCH.

Known problems: This counter includes also IMSI detaches which do not havea counter of their own.

sum(a.sdcch_assign)- sum(b.succ_seiz_term+ b.succ_seiz_orig+ b.sdcch_loc_upd+ b.succ_emerg_call+ b.sdcch_call_re_est)




a = p_nbsc_trafficb = p_nbsc_res_access

Figure 123. Ghosts detected on SDCCH and other failures, S1 (sd_1)

Ghosts detected on SDCCH and other failures, S1 (sd_1a)

Use: This part of ghost RACH accesses comprises:- ghosts which have an occasionally valid establishmentcause. These should comprise statistically 5/8 of all ghosts inGSM phase 1. Another 3/8 of ghosts are detected alreadybefore SDCCH based on some invalid establishment cause. InGSM2 the ratio 5/8 and 3/8 is no longer valid.- multiple seizures of SDCCH.

Known problems: This counter includes also IMSI detaches which do not havea counter of their own.

sum(a.sdcch_assign)- sum(b.succ_seiz_term

+ b.succ_seiz_orig+ b.sdcch_loc_upd+ b.sdcch_emerg_call+ b.sdcch_call_re_est+ imsi_detach_sdcch)

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_res_access

Figure 124. Ghosts detected on SDCCH and other failures, S1 (sd_1a)

2.10.1 SDCCH drop counters

SDCCH drop calls are counted as the sum of the following counters:



Table 4. SDCCH Drop Counters

ID Name Description

1003 SDCCH_RADIO_FAIL A coverage problem, for example

• MS moves out from timing advance

Common in connection with coverage problems.

Triggered also if the MS user clears the call in the SDCCHphase.

1004 SDCCH_RF_OLD_HO Transactions have ended due to an old channel failure in HO.

For instance, a failure in Directed Retry drops the call andtriggers this counter in the source cell.

1075 SDCCH_ABIS_FAIL_CALL Transactions have ended due to Abis problems. Missingchannel activation ack or if no indication of call establishmenthas been received. Augmented when the BSC receives anEstablish indication the contents of which are corrupted, ormore commonly when a timer (T3101, default 3 sec) expireswhile waiting for the Establish indication. The Establishindication is the first message sent from the BTS to the BSCafter the MS has successfully accessed the SDCCH.

• Ghost seizures which accidentally have a validestablishment cause and are detected on SDCCH,increment this counter.

• Multiple SDCCH seizures may cause these failures. If theMS has to send multiple random accesses for a call orlocation update, it is possible that there will be multiplereservations of SDCCH for one mobile naturally the mobilecan use only one of these and the rest will eventually timeout and result in sdcch_abis_fail_call. Onereason for multiple SDCCH seizures can be DLinterference.

• Too short frequency&BSIC reuse distance may cause HOburst from one cell to be interpreted as RACH bursts inanother cell causing false SDCCH seizures. This reasonmay be suspected if there are short, 2 - 3 second peakswith high blocking rate on SDCCH.

• A more rare yet possible reason are failing LUs.

1076 SDCCH_ABIS_FAIL_OLD Same as above but when trying to return back to the oldchannel in HO.

1078 SDCCH_A_IF_FAIL_CALL Transactions have ended due to A interface problems. A highvalue can be related to IMSI detaches (in S3).

See #1 for all possible causes. SDCCH observation may beused to diagnose the cause on cell level. If this occurs in theentire network or BSS areas, use Clear Code measurementfor cause analysis.



To see what DX causes can trigger the counters above, see #2.

2.10.2 Problems with the SDCCH drop counters

Phantoms affect SDCCH_abis_fail_*

SDCCH drop ratio counts the ratio of all SDCCH failures to SDCCH seizures.Normally most of the seizures are caused because of phantoms which are countedas SDCCH_abis_fail_call and SDCCH_abis_fail_old. In practice, thelatter case does not practically occur because SDCCH handovers are usually notused and particularly because phantoms do not perform handover.

1079 SDCCH_A_IF_FAIL_OLD Same as above but when trying to return back to the oldchannel in HO.

1035 SDCCH_LAPD_FAIL Transactions have ended due to LAPD problems (a call is lostwhen LAPD goes down).

TRX/TSL is blocked with cause lapd_fail due to a signallinglink fault or a PCM fault.

Even if it occurs, the share is normally very low because thesituation is transient.

1036 SDCCH_BTS_FAIL Transactions have ended due to BTS problems. A call is lostwhen TRX/TSL is blocked with cause bts_fail due to FUor CU or BCF fault or BTS or BCF reset.


1038 SDCCH_BCSU_RESET Transactions have ended due to BSCU reset (calls are lostwhen BSCU is reset).


1037 SDCCH_USER_ACT Transactions have ended due to user actions. A timeslot orTRX is locked by the user via the Top-level User Interface orBSC MML.


1039 SDCCH_NETW_ACT Transactions have ended due to a change in the radio networkconfiguration (BCCH swap to another TRX) initiated by theBSC. The cause for the configuration change fails or locallyblocked BCCH TRX.


Table 4. SDCCH Drop Counters (Continued)

ID Name Description



The percentage of SDCCH_abis_fail_call used to be very high in a low trafficnetwork (even tens of per cent) whereas in a high traffic network the percentagesettled down to around 18-20 per cent. In BTS B9, BTS RACH detection wasimproved, and figures well under 10 per cent are now typical.

A interface blocking not shown

SDCCH failures do not include A interface blocking. In A interface blocking, anMSC clears the call without a request from a BSC. The failure is not in the BSCwhere the principle has been that MSC failures are not counted. Yet from an MSuser’s point of view, A interface blocking ends up in a failed call attempt. Youcan detect A interface blocking from the NSS statistics for circuit groups.

2.11 SDCCH drop ratio (sdr)

SDCCH drop %, S3 (sdr_1a)

Use: To follow up the performance of SDCCH from a technicalpoint of view.

Experiences on use: 1) High SDCCH drop rates usually result from ghostaccesses. A BTS decodes them from environmental orbackground noise and filters out most of them. However, allof them cannot be filtered out and the RACH request is passedon to the BSC for processing and for the allocation of aSDCCH channel.The counter ghost_ccch_res (3030) is updated each time achannel required is rejected because of an invalidestablishment cause. In GSM ph.1 there exist altogether eightestablishment causes, three of which are undefined as invalid,for example,resulting in that this counter shows only 3/8 of allthe ghost accesses the BTS has decoded. For the rest, aSDCCH is allocated and this will result insdcch_abis_fail_call failure. Because of ghost attemptsthe SDCCH drop ratio is high with low traffic. As the amountof call attempts increases, the influence of ghosts becomessmaller and the drop ratio approaches its real value.2) The rate of ghosts coming to SDCCH dropped when BTSB9 with improved ghost filtering was taken into use.

Known problems: In SDCCH failure counters it is not possible to separate LUand call seizures.

sum(sdcch_radio_fail+sdcch_rf_old_ho+sdcch_user_act+sdcch_bcsu_reset+sdcch_netw_act+sdcch_abis_fail_call+sdcch_abis_fail_old+sdcch_bts_fail+sdcch_lapd_fail+sdcch_a_if_fail_call+sdcch_a_if_fail_old)

100 * ------------------------------------------------------------------------ %sum(sdcch_assign+sdcch_ho_seiz)



Counters from table(s):p_nbsc_traffic

Figure 125. SDCCH drop %, S3 (sdr_1a)

SDCCH drop %, abis fail excluded, S3 (sdr_2)

Known problems: SDCCH_ABIS_CALL does not necessarily refer to ghosts butalso, for example, to failing location updates.

sum(sdcch_radio_fail+sdcch_rf_old_ho+sdcch_user_act+sdcch_bcsu_reset+sdcch_netw_act +sdcch_bts_fail+sdcch_lapd_fail+sdcch_a_if_fail_call+sdcch_a_if_fail_old)

100 * ------------------------------------------------------------------- %sum(sdcch_assign+sdcch_ho_seiz)

- sum(sdcch_abis_fail_call+sdcch_abis_fail_old)


Figure 126. SDCCH drop %, abis fail excluded, S3 (sdr_2)

Illegal establishment cause % (sdr_3b)

Use: This PI gives you the number of ghost accesses which try toseize SDCCH but are rejected before seizing SDCCH due toan illegal establishment cause.

sum(a.ghost_CCCH_res-a.rej_seiz_att_due_dist)100 * ------------------------------------------------- %

sum(b.sdcch_assign+b.sdcch_ho_seiz)%

Counters from table(s):a = p_nbsc_res_accessb = p_nbsc_traffic

Figure 127. Illegal establishment cause % (sdr_3b)

SDCCH drop ratio without timer T3101 expiry % (sdr_4)

Use: This PI shows the ratio of dropped SDCCH withoutconsidering the drops caused by timer T3101. See sdr_1a.

Known problems: See sdr_1a.

sum(a.sdcch_radio_fail+ a.sdcch_rf_old_ho+ a.sdcch_user_act+ a.sdcch_bcsu_reset++ a.sdcch_netw_act+ a.sdcch_abis_fail_call+ a.sdcch_abis_fail_old+ a.sdcch_bts_fail+ a.sdcch_lapd_fail+ a.sdcch_a_if_fail_call



+ a.sdcch_a_if_fail_old- b. T3101_EXPIRED)

100 * -------------------------------- %sum(sdcch_assign+sdcch_ho_seiz)

a = p_nbsc_trafficb = p_nbsc_service

Counters from table(s):p_nbsc_trafficp_nbsc_service

Figure 128. SDCCH drop ratio without timer T3101 expiry % (sdr_4)

2.12 Setup success ratio (cssr)

SDCCH, TCH setup success %, S4 (cssr_2)

Use: This PI shows the setup success ratio, including SDCCH andTCH. It works also in the case of DR.Possible fault cases:- faulty DSP in BTS TRX

Experiences on use: Fits for general quality monitoring. Values between 2.5 and4%, for example.

Known problems: ’B no answer’ is also counted as a successful call.Includes also SMSs and LUs which do not use TCH at all.This causes problems in special cases when there are manyLUs but few calls. The problems in calls are hidden by a greatnumber of LUs which receive SDCCH successfully.

Troubleshooting: You can use SDCCH and TCH observations to see which oneis failing. However, note that this is a time-consuming task.

sum(call_setup_failure)100* ( 1 - ----------------------------------------) %

sum(setup_succ+call_setup_failure)

Counters from table(s):p_nbsc_service

Figure 129. SDCCH, TCH setup success %, S4 (cssr_2)

2.13 TCH drop failures

2.13.1 TCH drop call counters

TCH drop calls are counted as the sum of the following counters:



Table 5. TCH drop call counters

ID Name Description

1011 TCH_RADIO_FAIL • Radio link timeout

• Release indication from MS

• MS moves out from timing advance

• TCH assignment failure where an Establish Indication hasbeen received but Assignment Complete has not beenreceived

This counter is typical in connection with coverage problems. Thisfailure type is usually the dominating one.

1014 TCH_RF_OLD_HO Same as above but when trying to return to the old channel in HO.

1084 TCH_ABIS_FAIL_CALL • missing ack of channel activation

• missing establishment indication

• reception of error indication

• corruption of messages

• measurement results no longer received from BTS

• excessive timing advance

• missing HO detection

• T3107 (assignment completely missing) expiry

• T3109 expiry. As in this case the drop happens in the releasephase, the MS user cannot see the situation as a drop call.

The BTS suffering from this failure can be faulty or their TCH TRXsuffers from bad interference (TCH assignment fails).

See #1 for all possible causes that trigger this counter.

If FACCH call setup is used, we may expect that we start to seeghost seizures incrementing this counter because the signallingtries to use SDCCH instead of TCH.

1085 TCH_ABIS_FAIL_OLD Same as above but when trying to return to the old channel in HO.

1087 TCH_A_IF_FAIL_CALL • A clear command from the MSC during the call setup phasebefore the assignment from MS is complete

• Abnormal clear received due to A-interface (reset circuit,SCCP clear). For example, a BSU reset in MSC makes MSCsend a "reset circuit" message.

There have been cases where the MSC of another vendor in inter-MSC handovers has caused high values even though thehandovers have been successful.

See #1 for all possible causes that trigger this counter.



1088 TCH_A_IF_FAIL_OLD Same as above but when trying to return to the old channel in HO.

Can be updated when GSM timer T8 expires in the source BSCduring an external handover.

1029 TCH_TR_FAIL • Transcoder failure during a call attempt

This counter is updated only when BTS sends a "connectionfailure" with cause "remote trascoder failure" and the call isreleased due to this.

If this failure is related to a transcoder, you can see its share to behigh for one BSC. Another possibility is that the problem lies in aBTS. Also interruptions of the transmission may cause this failure(alarms may be filtered out in a BSC or OMC to reduce thenumber of alarms due to disturbance).

In analysing the problem, you may find it helpful to check thepattern over a longer period of time.

In S6 the portion of this failure has decreased due toimprovements in transcoders.

1030 TCH_TR_FAIL_OLD Same as above but when trying to return to the old channel in HO.

1046 TCH_LAPD_FAIL TRX is blocked due to a LAPD failure (signalling link failure orPCM failure).

Even if it occurs, the share is very small because only ongoingcalls are dropped when the LAPD fails.

1047 TCH_BTS_FAIL TRX is blocked by a BTS failure.

(FU fault, CU fault, BTS reset, BCF reset, CU and FU fault, BCFfault).

Even if it occurs, the share is very small because only ongoingcalls are dropped when a BTS fails.

1049 TCH_BCSU_RESET TRX is blocked by BCSU reset.

Even if it occurs, the share is very small because only ongoingcalls are dropped when BCSU resets.

Table 5. TCH drop call counters (Continued)

ID Name Description



To see what DX causes can trigger the counters above, see #2.

The problem especially with failure classes Abis and Aif is that they are triggeredby many different causes. To analyse the case in question, TCH observations maybe used. As there are limitations on how to set the observations, the analysis ismore time-consuming.

If the problem is not cell specific but related to the entire network or BSS areas,you can also use the Clear Code measurement.

*_OLD counters are related to the handover situation when returning to the oldchannel fails causing a call to drop. Thus, these counters reflect the amount of calldrops in handovers.

2.13.2 Drop call ratio

Drop call ratio is counted as the ratio of the sum of the above named counters toall TCH seizures for a new call. This ratio is used in reports for network ormaintenance region level. See dcr_3*.

2.13.3 Drop-out ratio

Drop-out ratio is counted as the ratio of the sum of the above named counters toall TCH seizures. This ratio is used on cell level reports where the concept of callis not applicable.

1048 TCH_USER_ACT Busy TSL or TRX blocked by MML command (blocked by user).

Even if it occurs, the share is very small because only ongoingcalls are dropped when the blocking command is given.

1050 TCH_NETW_ACT TRX is blocked by a fault leading to reconfiguration (blocked by thesystem).

Even if occurs, the share is very small because only ongoing callsare dropped when reconfiguration is executed.

1081 TCH_ACT_FAIL_CALL Channel activation nack received.

Table 5. TCH drop call counters (Continued)

ID Name Description



2.13.4 Problems with the drop call counters

TCH Tr, Abis failures

These failures can contain also situations when timer T3109 (8 to 15 s, default 12s) expires in a BSC in the call release phase while waiting for the Releaseindication. With BTS software 6.0 these were seen as TC failures, whereas sinceBTS software 6.1 they have been Abis failures. The MS user does not see thesefailures as real drop calls.

If you detect a high ratio of TC or Abis failures, check the BTS release and thetimer.

TCH_A_IF_OLD high

There have been cases when this counter has been showing high values whileanother vendor’s MSC has cleared the call with the cause CLR_CMD in the case ofa successful inter-MSC HO.

2.14 Drop call failures (dcf)

TCH drop calls in HO, S2 (dcf_2)

Open questions: Claims of cases when the TCH_A_IF_OLD has not been a dropcall have been made.

sum(tch_rf_old_ho+ tch_abis_fail_old+tch_a_if_fail_old+tch_tr_fail_old)


Figure 130. TCH drop calls in HO, S2 (dcf_2)

TCH drop calls in BSC outgoing HO, S3 (dcf_3)

Known problems: Accuracy is not good.

sum(bsc_o_drop_calls)

Counters from table(s):p_nbsc_ho

Figure 131. TCH drop calls in BSC outgoing HO, S3 (dcf_3)

TCH drop calls in intra-cell HO, S3 (dcf_4)


sum(cell_drop_calls)




Figure 132. TCH drop calls in intra-cell HO, S3 (dcf_4)

TCH drop calls in intra-BSC HO, S3 (dcf_6)

Known problems: Use on the BTS level. On the area level causes doublecounting. Accuracy is not good.

sum(bsc_i_drop_calls+bsc_o_drop_calls+cell_drop_calls)


Figure 133. TCH drop calls in intra BSC HO, S3 (dcf_6)

Drop calls in BSC incoming HO, S3 (dcf_7)


sum(bsc_i_drop_calls)


Figure 134. Drop calls in BSC incoming HO, S3 (dcf_7)

TCH drop calls in HO, S7 (dcf_11)

Known problems: On the area level causes double counting. Accuracy is notgood.

sum(msc_o_call_drop_ho+bsc_i_drop_calls+bsc_o_drop_calls+cell_drop_calls)


Figure 135. TCH drop calls in HO, S7 (dcf_11)

2.15 TCH drop call % (dcr)

TCH drop call %, area, S3 (dcr_3c)

Use: Used on the area level.Experiences on use: See dcr_3. The best value reported for area: 2.3



Known problems: 1) Some failures in release phase are included in this formula(tch_abis_fail_call) but are, in fact, not perceived as drop callsby the MS user.2) tch_norm_seiz does not mean that the MS is on TCH. Itmeans that TCH has been successfully seized. Some mobilesnever appear to the TCH because2a) the call is cleared by user (probability is higher if callsetup takes a long time, and thus DR and queuing canincrease this share) or2b) the mobile fails or2c) something else goes wrong.3) TCH failure counters are not triggered if call is cleared bypre-emption (1st priority call requested to be established, allTCH seized, lower priority calls on) whereas p_nbsc-service.dropped_call is triggered.

sum(tch_radio_fail+ tch_rf_old_ho+ tch_abis_fail_call+ tch_abis_fail_old+ tch_a_if_fail_call+ tch_a_if_fail_old+ tch_tr_fail+ tch_tr_fail_old+ tch_lapd_fail+ tch_bts_fail+ tch_user_act+ tch_bcsu_reset+ tch_netw_act+ tch_act_fail_call)

100 * ------------------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch + c.bsc_i_sdcch_tch) ;(calls started via DR)+ sum(a.tch_seiz_due_sdcch_con) ;calls started as FACCH call setup

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_res_accessc = p_nbsc_ho

Figure 136. TCH drop call %, area, S3 (dcr_3c)

TCH drop call %, area, real, after re-establishment, S3 (dcr_3f)

Use: Used on the area level.



Experiences on use: See dcr_3g. Call re-establishments can markedly improvethe drop call ratio (for example, from 2.5 to 2.0%). Since thisis an improvement from the MS user’s point of view, thisfigure suits better to management reports.In good networks where optimisation has been done alreadyfor two to three years, values have been around 2 to 3 per cent(and in networks in which no optimisation has been done yetthe values remain even above 10 per cent. A value of 5 percent is achievable in many networks despite their bad initialcoverage planning. Interference also raises the figure. Becareful when setting the target values since the factors(whether caused by the customer or Nokia) can be time-consuming and expensive to prove.If used on the cell level, the values can be even over 100 percent if a cell takes handovers in but then drops them.

Known problems: 1) See dcr_3g.2) It is assumed that call re-establishments happen on TCH. Infact they may happen also on SDCCH.3) The counters used to compensate re-establishments are theones that indicate re-establishment attempts, not thesuccessful re-establishments. In S7/T11 re-establishments canbe considered accurately (see dcr_3j).4) On cell level it can happen that the call is re-established ina different cell than where it was dropped, resulting ininaccuracy.

100 - csf_4p =

sum(a.tch_radio_fail+ a.tch_rf_old_ho+ a.tch_abis_fail_call+ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

- sum(b.sdcch_call_re_est + b.tch_call_re_est) ;call re-establ.100 * ------------------------------------------------------------------- %

sum(a.tch_norm_seiz) ;calls started directly in the cell+ sum(c.msc_i_sdcch_tch

+ c.bsc_i_sdcch_tch+ c.cell_sdcch_tch) ;DR calls

+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls- sum(b.sdcch_call_re_est+b.tch_call_re_est) ;call re-establ.

Counters from table(s):a = p_nbsc_traffic



b = p_nbsc_res_accessc = p_nbsc_ho

Figure 137. TCH drop call %, area, real, after re-establishment S3 (dcr_3f)

TCH drop call %, area, real, before re-establishment, S3 (dcr_3g)

Use: Used on the area level.Experiences on use: In good networks where optimisation has been done already

for two to three years, values have been around 2 to 3 per cent(and in networks in which no optimisation has been done yet,the values remain even above 10 per cent). A value of 5 percent is achievable in many networks despite their bad initialcoverage planning.Interference also raises the figure.Be careful when setting the target values since the factors(whether caused by the customer or Nokia) can be time-consuming and expensive to prove.If used on cell level, the values can be even over 100 per centif a cell takes HOs in but then drops them.

Known problems: Some failures in the release phase are included in this formula(tch_abis_fail_call) but are, in fact, not perceived asdrop calls by the MS user.tch_norm_seiz does not mean that the MS is on TCH. Itmeans that TCH has been successfully seized. Some mobilesnever appear to the TCH because:• the call is cleared by the user (probability is higher if

call setup takes a long time, and thus DR and queuingcan increase this share) or

• the mobile fails or• something else goes wrong.TCH failure counters are not triggered if a call is cleared bypre-emption (1st priority call requested to be established, allTCH seized, lower priority calls on) whereas p_nbsc-service_dropped_call is triggered.

sum(tch_radio_fail+ tch_rf_old_ho+ tch_abis_fail_call+ tch_abis_fail_old+ tch_a_if_fail_call+ tch_a_if_fail_old+ tch_tr_fail+ tch_tr_fail_old+ tch_lapd_fail+ tch_bts_fail+ tch_user_act+ tch_bcsu_reset+ tch_netw_act+ tch_act_fail_call)

100 * ------------------------------------------------------------------ %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch



+ c.bsc_i_sdcch_tch+ c.cell_sdcch_tch) ;(DR calls)

+ sum(a.tch_seiz_due_sdcch_con) ;calls started as FACCH call setup

Counters from table(s):a = p_nbsc_trafficc = p_nbsc_ho

Figure 138. TCH drop call %, area, real, before re-establishment, S3 (dcr_3g)

TCH drop call %, area, real, after re-establishment, S7 (dcr_3h)

Use: Used on the area level.Experiences on use: See dcr_3. Call re-establishments can markedly improve

the drop call ratio (for example, from 2.5 to 2.0%). Since thisis an improvement from the MS user’s point of view, thisfigure suits better to management reports.

Known problems: See dcr_3g. It is assumed that call re-establishments occuron TCH. In fact they may occur also on SDCCH.

sum(tch_radio_fail+ tch_rf_old_ho+ tch_abis_fail_call+ tch_abis_fail_old+ tch_a_if_fail_call+ tch_a_if_fail_old+ tch_tr_fail+ tch_tr_fail_old+ tch_lapd_fail+ tch_bts_fail+ tch_user_act+ tch_bcsu_reset+ tch_netw_act+ tch_act_fail_call

- sum(b.tch_re_est_assign) ;(call re-establishments)100 * ------------------------------------------------------------------- %

sum(a.tch_norm_seiz) ;(normal calls)+ sum(c.msc_i_sdcch_tch


+ sum(a.tch_seiz_due_sdcch_con) ;calls started as FACCH call setup- sum(b.tch_re_est_assign) ;(call re-establishments)

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_servicec = p_nbsc_ho

Figure 139. TCH drop call %, area, real, after re-establishment, S7 (dcr_3h)

TCH drop call %, area, real, before re-establishment S3 (dcr_3i)

Use: Used on the area level.This KPI indicates how many calls are dropped after TCHseizure.



Experiences on use: In good networks where optimisation has been done alreadyfor two to three years, values have been around 2 to 3 per cent(and in networks in which no optimisation has been done yetthe values remain even above 10 per cent. A value of 5 percent is achievable in many networks despite their bad initialcoverage planning. Interference also raises the figure. Becareful when setting the target values since the factors(whether caused by the customer or Nokia) can be time-consuming and expensive to prove.If used on cell level, the values can be even over 100 per centif a cell takes HOs in but then drops them.

Known problems: 1) Some failures in release phase are included in this formula(tch_abis_fail_call) but are, in fact, not perceived asdrop calls by the MS user.2) tch_norm_seiz does not mean that the MS is on TCH. Itmeans that TCH has been successfully seized. Some mobilesnever appear to the TCH because2a) the call is cleared by the user (probability is higher if callsetup takes a long time, and thus DR and queuing canincrease this share) or2b) the mobile fails or2c) something else goes wrong.3) TCH failure counters are not triggered if a call is cleared bypre-emption (1st priority call requested to be established, allTCH seized, lower priority calls on), whereas p_nbsc-service.dropped_call is triggered.

100-csf_4u =

sum(a.tch_radio_fail+ a.tch_rf_old_ho+ a.tch_abis_fail_call++ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

100 * ----------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch+ c.bsc_i_sdcch_tch+ c.cell_sdcch_tch) ;(DR calls)

- sum(a.tch_succ_seiz_for_dir_acc ) ;ref.2+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls


Figure 140. TCH drop call %, area, real, before re-establishment, S3 (dcr_3i)



Ref.2. Compensation is needed, since in case of Direct Access to super reuse TRXthe tch_norm_seiz is triggered in parallel with cell_sdcch_tch.

TCH drop call %, area, real, after re-establishment, S7 (dcr_3j)

Use: On the area level.Experiences on use: In good networks where optimisation has been done already

for two to three years, values are around 2 to 3 per cent. Innetworks in which no optimisation has been done yet thevalues are even as high as 10 per cent. A value of 5 per cent isachievable in many networks despite their bad initialcoverage planning.The values in the best networks are below 1.5%.Interference also raises the figure.Be careful when you give promises concerning quality sincethe factors (whether caused by the customer or Nokia) can betime-consuming and expensive to prove.If used on cell level, the values can be even over 100 per centif a cell takes handovers in but then drops them.Call re-establishments can markedly improve the drop callratio (for example, from 2.3 to 2.0 %). Since this is animprovement from the MS user's point of view, this figuresuits better to management reports.The biggest reason for having low figures usually is in basicnetwork planning. If coverage is not adequate, this KPI cannotshow good values.

Known problems: 1) Some failures in the release phase are included in thisformula (tch_abis_fail_call) but are, in fact, notperceived as drop calls by the MS user.2) tch_norm_seiz does not mean that the MS is on TCH.It means that TCH has been successfully seized. Somemobiles never appear to the TCH because2a) the call is cleared by user (probability is higher if callsetup takes a long time, and thus DR and queuing canincrease this share) or2b) the mobile fails or2c) something else goes wrong3) TCH failure counters are not triggered if call is cleared bypre-emption (first priority call requested to be established, allTCH seized, lower priority calls on) whereas p_nbsc-service.dropped_call is triggered.4) It is assumed that call re-establishments happen on TCH. Infact they may happen also on SDCCH.5) On the cell level it can happen that the call is re-establishedin a different cell than it was dropped resulting in inaccuracy.

100-csf_4v =

sum(a.tch_radio_fail+ a.tch_rf_old_ho+ a.tch_abis_fail_call



+ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

- sum(b.tch_re_est_assign) ;call re-establishments100 * ----------------------------------------------------------- %



- sum(a.tch_succ_seiz_for_dir_acc) ;ref.1+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls- sum(b.tch_re_est_assign) ;call re-establishments


Figure 141. TCH drop call %, area, real, after re-establishment, S7 (dcr_3j)


TCH drop-out %, BTS level, before call re-establishment, S3 (dcr_4c)

Use: Used on the BTS level. To rank cells by the share of TCH dropcall failures per TCH seizure (normal or HO). Intra-cell HOexcluded which is meaningful in the case of IUO, forexample.

Known problems: See dcr_3g.


100 * ----------------------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)


+ sum(a.tch_seiz_due_sdcch_con) ;(FACCH call setup calls)+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch) ;(TCH-TCH HO from other cells)




a = p_nbsc_trafficc = p_nbsc_ho

Figure 142. TCH drop-out %, BTS level, before call re-establishment, S3(dcr_4c)

TCH drop-out %, BTS level, before call re-establishment, S3 (dcr_4d)

Use: Used on the BTS level. To rank cells by the share of TCH dropcall failures per TCH seizure (normal or HO). Intra-cell HO isexcluded, which is meaningful in the case of IUO. Inter-cellHOs are counted only as a net value.



100 * ------------------------------------------------------------ %sum(a.tch_norm_seiz) ;(normal calls)


+ sum(a.tch_seiz_due_sdcch_con) ;(FACCH call setup calls)+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch)- sum(c.msc_o_tch_tch

+ c.bsc_o_tch_tch) ;(TCH-TCH HO net in from other cells)


Figure 143. TCH drop-out %, BTS level, before call re-establishment, S3(dcr_4d)

TCH drop-out %, BTS level, before call re-establishment, S7 (dcr_4e)

Use: Used on the BTS level. To rank cells by the share of TCH dropcall failures per TCH seizure (normal or HO). Intra-cell HO isexcluded, which is meaningful in the case of IUO, forexample.


100 - csf_4y =

sum(a.tch_radio_fail+ a.tch_rf_old_ho



+ a.tch_abis_fail_call+ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

100 * ------------------------------------------------------ %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch+ c.bsc_i_sdcch_tch+ c.cell sdcch_tch) ;(DR calls)

- sum(a.tch_succ_seiz_for_dir_acc ) ;ref.2+ sum(a.tch_seiz_due_sdcch_con) ; FACCH call setup calls+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch) ;(TCH-TCH HO from other cells)


Figure 144. TCH drop-out %, BTS level, before call re-establishment, S7(dcr_4e)


TCH drop-out %, BTS level, before call re-establishment, S7 (dcr_4f)

Use: Used on the BTS level. To rank cells by the share of TCH dropcall failures per TCH seizure (normal or HO). Intra-cell HO isexcluded, which is meaningful in the case of IUO. Inter-cellHOs are counted only as a net value.



100 * ------------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)


- sum(a.tch_succ_seiz_for_dir_acc) ;ref.2+ sum(a.tch_seiz_due_sdcch_con) ;(FACCH call setup calls)+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch)- sum(c.msc_o_tch_tch



+ c.bsc_o_tch_tch) ;(TCH-TCH HO net in from other cells)


Figure 145. TCH drop-out %, BTS level, before call re-establishment, S7 (dcr_4f)


TCH drop call (dropped conversation) %, BSC level, S4 (dcr_5)

Use: Used on the area level. Tells the ratio of calls dropped whileA and B are talking, that is after conn_ack.Theoretically should always be less than dcr_3f or dcr_3g.Results from networks 2 to 6 %.

Known problems: 1) Does not work on the BTS level (handovers). Accurate onthe BSC and PLMN levels after a bug was corrected.2) If call re-establishment is active and occurs, theconver_started is triggered once and dropped_callsonly once. After the first call re-establishment thedropped_calls counter is no longer incremented in this callno matter if the call stays or drops. This means that in this caseseen from counters, the call looks like a dropped call. In fact,from the MS user’s point of view it is impossible to knowwhether it was dropped or not (does callre_establishment save the call or not).3) Subscriber clear during HO is counted as a dropped call.4) Due to an error in the mapping table also blocking in thecase of an external HO has been counted as a dropped call.5) External HOs (inter BSC handovers) trigger theconver_started counter in a target cell. Therefore onnetwork level the ratio does not correctly illustrate thedropped conversation ratio from the MS’s point of view.

sum(dropped_calls)100 * -------------------- %

sum(conver_started)


Figure 146. TCH drop call (dropped conversation) %, BSC level, S4 (dcr_5)



TCH dropped conversation %, area, re-establishment considered, S7(dcr_5b)

Use: Used on the area level. Tells the ratio of calls dropped whileA and B are talking, that is after conn_ack. Inter BSChandovers are subtracted in the denominator because theytrigger conver_started. Compensation is 100% true onlyif the area has no inter-BSC handovers from outside the area.Theoretically should always be less than dcr_3f or dcr_3g.

Known problems: See dcr_5.1) conver_started is not triggered for call re-establishments.dropped_calls is triggered once for the first re-establishment. After that setup_failure is triggered if thecall is a dropped call.If the call is ’saved’ by re-establishment multiple times,setup_failure will be triggered several times accordingly.2) Drop call by pre-emption (1st priority call request to beestablished, all TCHs seized, lower priority calls on) triggersdropped_calls. Therefore this counter does not indicateonly technical drops.

sum(b.dropped_calls) - sum(tch_re_est_release)100 * -------------------------------------------------- %

sum(b.conver_started) - sum(a.msc_i_tch_tch)

Counters from table(s):a = p_nbsc_hob = p_nbsc_service

Figure 147. TCH dropped conversation %, area, re-establishment considered,S7 (dcr_5b)

TCH drop call %, after TCH assignment, without RE, area level, S10.5(dcr_8c)

Use: This formula is developed to better match with the formulasof other vendors. It does not consider the impact of the call re-establishment.

Known problems: 1) The formula is not reliable on hourly level because assignsand releases can happen in different measurement periods. Inthe worst case this can cause a negative value.2) Not good for BTS level because the denominator countsonly started new calls (there can be a lot of handovers in, too).3) The following bugs that affect the formula have beencorrected in S9:



3a) Counter TCH_NORM_RELEASE (c57035) is notupdated if during the call there has been a MSC controlled HOwith cause 'pre-emption' or 'traffic'.3b) TCH_NEW_CALL_ASSIGN (c57033) is not updated inthe case of MSC controlled SDCCH-TCH HO with cause'pre-emption' or 'traffic'.

Drops after TCH assignment100 * ------------------------------ % =

TCH assignments for new calls

sum(spare057044)100 * ------------------------ %

sum(tch_new_call_assign)


Figure 148. TCH drop call %, after TCH assignment, without RE, area level, S10.5 (dcr_8c)

TCH drop call %, after TCH assignment, with RE, area level, S10.5 (dcr_8e)

Description: TCH drop call%, after TCH assignment, considering re-establishments.

Use: This formula is developed to better match with the formulasof other vendors.

Known problems: See dcr_8c.

Drops after TCH assignment considering re-establishments100 * -------------------------------------------------------- % =

TCH assignments for new calls

sum(spare057044 - tch_re_est_release)100 * ------------------------------------- %

sum(tch_new_call_assign)


Figure 149. TCH drop call %, after TCH assignment, with RE, area level, S10.5(dcr_8e)

Drops per erlang, before re-establishment, S4 (dcr_10)

Use: Used on the area and BTS level.Known problems: Works for the 60 min period.

Drops-------------------------- =Traffic (Erlang hours sum)

sum(a.tch_radio_fail+ a.tch_rf_old_ho+ a.tch_abis_fail_call



+ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

---------------------------------------------------------------sum(b.ave_busy_tch/b.res_av_denom14) / (60/avg(period_duration))

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_res_avail

Figure 150. Drops per erlang, before re-establishment, S4 (dcr_10)

Drops per erlang, after re-establishment, S4 (dcr_10a)


The counters used to compensate re-establishments are theones that indicate re-establishment attempts, not thesuccessful re-establishments.

Drops- re-establishments-------------------------- =Traffic (Erlang hours sum)


- sum(c.sdcch_call_re_est + c.tch_call_re_est) ;call re-establishments-------------------------------------------------------------------------

sum(b.ave_busy_tch/b.res_av_denom14) / (60/avg(period_duration))

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_res_availc = p_nbsc_res_access

Figure 151. Drops per erlang, after re-establishment, S4 (dcr_10a)



Drops per erlang, after re-establishment, S7 (dcr_10b)


Drops - re-establishments-------------------------- =Traffic (Erlang hours sum)


- sum(c.tch_re_est_assign) ;call re-establishments--------------------------------------------------------------------

sum(b.ave_busy_tch/b.res_av_denom14) / (60/avg(b.period_duration))

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_res_availc = p_nbsc_service

Figure 152. Drops per erlang, after re-establishment, S7 (dcr_10b)

TCH Rf loss in HO - ratio, IUO (dcr_14)

sum(TCH_FAIL_CALL_HO)100 * ---------------------- %

sum(tch_succ_seiz)

Counters from table(s):p_nbsc_underlay

Figure 153. TCH Rf loss in HO - ratio, IUO (dcr_14)

Transcoder failure ratio, FR (dcr_16)

sum(TCH_ENDED_DUE_TRANSC_FR_RATE1)100 * ---------------------------------- %

sum(TCH_FULL_SEIZ_SPEECH_VER1)


Figure 154. Transcoder failure ratio, FR (dcr_16)



Transcoder failure ratio, EFR (dcr_17)




Figure 155. Transcoder failure ratio, EFR (dcr_17)

Transcoder failure ratio, HR (dcr_18)

sum(TCH_ENDED_DUE_TRANSC_HR_RATE1)100 * ---------------------------------- %

sum(TCH_HALF_SEIZ_SPEECH_VER1)


Figure 156. Transcoder failure ratio, HR (dcr_18)

Transcoder failure ratio, AMR FR (dcr_19)




Figure 157. Transcoder failure ratio, AMR FR (dcr_19)

Transcoder failure ratio, AMR HR (dcr_20)

sum(TCH_ENDED_DUE_TRANSC_HR_RATE3)100 * ---------------------------------- %

sum(TCH_HALF_SEIZ_SPEECH_VER3)


Figure 158. Transcoder failure ratio, AMR HR (dcr_20)

Transcoder failure ratio (dcr_21)

sum(TCH_ENDED_DUE_TRANSC_FR_RATE1+TCH_ENDED_DUE_TRANSC_HR_RATE1+TCH_ENDED_DUE_TRANSC_FR_RATE2+TCH_ENDED_DUE_TRANSC_FR_RATE3+TCH_ENDED_DUE_TRANSC_HR_RATE3")



100 * ----------------------------------------------------------------- %sum(TCH_FULL_SEIZ_SPEECH_VER1+TCH_HALF_SEIZ_SPEECH_VER1

+TCH_FULL_SEIZ_SPEECH_VER2+TCH_FULL_SEIZ_SPEECH_VER3+TCH_HALF_SEIZ_SPEECH_VER3)


Figure 159. Transcoder failure ratio (dcr_21)

Call failures share of transcoder failures (dcr_22)

Use: Indicates the percentage of transcoding failures during TCHseizure for a call. These are call drops.

Sum(TCH_TR_FAIL)100 * ------------------------------------------------ %

sum(TCH_TR_FAIL+TCH_TR_FAIL_OLD+TCH_TR_FAIL_NEW)


Figure 160. Call failures share of transcoder failures (dcr_22)

HO target share of transcoder failures (dcr_23)

Use: Indicates the percentage of transcoding failures in the targetcell during TCH seizure for HO. These are not call drops.

Sum(TCH_TR_FAIL_NEW)100 * ------------------------------------------------ %



Figure 161. HO target share of transcoder failures (dcr_23)

HO source share of transcoder failures (dcr_24)

Use: Indicates as a percentage the transcoding failures that happenin the source cell in HO when MS fails to return to the old celland a call drops.

Sum(TCH_TR_FAIL_OLD)100 * ------------------------------------------------ %



Figure 162. HO source share of transcoder failures (dcr_24)



Transcoder failures (dcr_25)



Figure 163. Transcoder failures (dcr_25)

2.16 Adaptive Multirate (amr)

Codec set upgrade attempts, S10 (amr_1)

Sum(SUCC_AMR_CODEC_SET_UPGR+UNSUCC_AMR_CODEC_SET_UPGR)


Figure 164. Codec set upgrade attempts, S10 (amr_1)

Codec set downgrade attempts, S10 (amr_2)

Sum(SUCC_AMR_CODEC_SET_DOWNGR+UNSUCC_AMR_CODEC_SET_DOWNGR)


Figure 165. Codec set downgrade attempts, S10 (amr_2)

Codec set upgrade failure ratio, S10 (amr_3)

Sum(UNSUCC_AMR_CODEC_SET_UPGR)100 * ------------------------------------------------------ %

Sum(SUCC_AMR_CODEC_SET_UPGR+UNSUCC_AMR_CODEC_SET_UPGR)


Figure 166. Codec set upgrade failure ratio, S10 (amr_3)

Codec set downgrade failure ratio, S10 (amr_4)

Sum(UNSUCC_AMR_CODEC_SET_DOWNGR)100 * ---------------------------------------------------------- %

Sum(SUCC_AMR_CODEC_SET_DOWNGR+UNSUCC_AMR_CODEC_SET_DOWNGR)




Figure 167. Codec set downgrade failure ratio, S10 (amr_4)

2.17 Position based services (pbs)

Failure ratio of location calculations for external LCS clients, S10 (pbs_1a)

Sum(SUCC_LOC_CALC_BY_LCS_REQ)100 - 100*--------------------------------- %

Sum(NBR_OF_LOC_REQ_FROM_LCS)

Counters from table(s):p_nbsc_pbs

Figure 168. Failure ratio of location calculations for external LCS clients, S10(pbs_1a)

Failure ratio of location calculations for emergency calls, S10 (pbs_2a)

Sum(SUCC_LOC_CALC_EMERGENCY)100 - 100*------------------------------- %

Sum(NBR_OF_LOC_REQ_EMERGENCY)


Figure 169. Failure ratio of location calculations for emergency calls, S10(pbs_2a)

Failure ratio of E-OTD location calculations, S10 (pbs_3)

Sum(SUCC_LOC_CALC_E_OTD)100 - 100 * --------------------------------------------------- %

Sum(NBR_OF_E_OTD_CALCULATIONS + SUCC_LOC_CALC_E_OTD)


Figure 170. Failure ratio of E-OTD location calculations, S10 (pbs_3)

Failure ratio of E-OTD location calculations, S10 (pbs_3a)

Sum(SUCC_LOC_CALC_E_OTD)100 - 100*------------------------------- %

Sum(NBR_OF_E_OTD_CALCULATIONS)




Figure 171. Failure ratio of E-OTD location calculations, S10 (pbs_3a)

Failure ratio of location calculations for MS, S10 (pbs_4a)

Sum(SUCC_LOC_CALC_BY_MS_REQ)100 - 100*------------------------------- %

Sum(NBR_OF_LOC_REQ_FROM_MS)


Figure 172. Failure ratio of location calculations for MS, S10 (pbs_4a)

Failure ratio of location calculations for operator, S10 (pbs_5a)

Sum(SUCC_LOC_CALC_BY_OPER_REQ)100 - 100*-------------------------------- %

Sum(NBR_OF_LOC_REQ_FROM_OPER)


Figure 173. Failure ratio of location calculations for operator, S10 (pbs_5a)

Failure ratio of location calculations using stand-alone GPS, S10 (pbs_6)

Sum(SUCC_LOC_CALC_STAND_ALONE_GPS)100 - 100 * ----------------------------------------------------------------- %

Sum(NBR_LOC_CALC_STAND_ALONE_GPS + SUCC_LOC_CALC_STAND_ALONE_GPS)


Figure 174. Failure ratio of location calculations using stand-alone GPS, S10(pbs_6)

Failure ratio of location calculations using stand-alone GPS, S10 (pbs_6a)

Sum(SUCC_LOC_CALC_STAND_ALONE_GPS)100 - 100*-------------------------------------%

Sum(NBR_LOC_CALC_STAND_ALONE_GPS)




Figure 175. Failure ratio of location calculations using stand-alone GPS, S10(pbs_6a)

Unspecified LCS requests, S10 (pbs_8)

Use: Indicates the requests for which the client type is unspecified.

Sum(NBR_OF_LOC_REQ_FROM_LCS-NBR_OF_LOC_REQ_EMERGENCY-NBR_OF_LOC_REQ_FROM_MS-NBR_OF_LOC_REQ_FROM_OPER)


Figure 176. Unspecified LCS requests, S10 (pbs_8)

2.18 Handover (ho)

Return from super TRXs to regular TRX, S4 (ho_1)

sum(ho_succ_to_reg_freq)100 * -------------------------- %

sum(ho_succ_from_reg_freq)


Figure 177. Return from super TRXs to regular TRX, S4 (ho_1)

HO attempts from regular TRXs to super, S4 (ho_2)

sum(ho_att_from_reg_freq)


Figure 178. HO attempts from regular TRXs to super, S4 (ho_2)

HO attempts from super to regular, S4 (ho_3)

sum(ho_att_to_reg_freq)




Figure 179. HO attempts from super to regular, S4 (ho_3)

Share of HO attempts from super to regular due to DL quality, S4 (ho_4)

sum(att_from_super_dl_qual)100 * -------------------------------------------------- %

sum(att_from_super_dl_qual + att_from_super_dl_if+att_from_super_ul_if + att_from_super_bad_ci)


Figure 180. Share of HO attempts from super to regular due to DL quality, S4(ho_4)

Share of HO attempts from super to regular due to DL interference, S4(ho_5)

sum(att_from_super_dl_if)100 * -------------------------------------------------- %



Figure 181. Share of HO attempts from super to regular due to DL interference,S4 (ho_5)

Share of HO attempts from super to regular due to UL interference, S4(ho_6)

sum(att_from_super_ul_if)100 * -------------------------------------------------- %



Figure 182. Share of HO attempts from super to regular due to UL interference,S4 (ho_6)

Share of HO attempts from super to regular due to bad C/I, S4 (ho_7)

sum(att_from_super_bad_ci)



100 * -------------------------------------------------- %sum(att_from_super_dl_qual + att_from_super_dl_if

+att_from_super_ul_if + att_from_super_bad_ci)


Figure 183. Share of HO attempts from super to regular due to bad C/I, S4(ho_7)

MSC incoming HO attempts (ho_8)

sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch_at)


Figure 184. MSC incoming HO attempts (ho_8)

MSC outgoing HO attempts (ho_9)

sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at)


Figure 185. MSC outgoing HO attempts (ho_9)

BSC incoming HO attempts (ho_10)

sum(bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch_at)


Figure 186. BSC incoming HO attempts (ho_10)

BSC outgoing HO attempts (ho_11)

sum(bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at)


Figure 187. BSC outgoing HO attempts (ho_11)



Intra-cell HO attempts, S6 (ho_12a)

sum(cell_tch_tch_at+cell_sdcch_at+cell_sdcch_tch_at)


Figure 188. Intra-cell HO attempts, S6 (ho_12a)

HO attempts, S3 (ho_13a)

sum(msc_i_tch_tch_at + msc_i_sdcch_tch_at + msc_i_sdcch_at)+ sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at + msc_o_sdcch_at)+ sum(bsc_i_tch_tch_at + bsc_i_sdcch_tch_at + bsc_i_sdcch_at)+ sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at + bsc_o_sdcch_at)+ sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 189. HO attempts, S3 (ho_13a)

HO attempts, outgoing and intra-cell, S5 (ho_13b)

sum(cause_up_qual+ cause_up_level+ cause_down_qual+ cause_down_lev+ cause_distance+ cause_msc_invoc+ cause_intfer_up+ cause_intfer_dwn+ cause_umbr+ cause_pbdgt+ cause_omc+ cause_ch_adm+ cause_traffic+ cause_dir_retry+ cause_pre_emption+ cause_field_drop+ cause_low_distance+ cause_bad_CI+ cause_good_CI+ ho_due_slow_mov_ms)


p_nbsc_ho

Figure 190. HO attempts, outgoing and intra-cell, S5 (ho_13b)

HO attempts, outgoing and intra-cell, S3 (ho_13e)

sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at + msc_o_sdcch_at)



+ sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at + bsc_o_sdcch_at)+ sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 191. HO attempts, outgoing and intra-cell, S3 (ho_13e)

HO attempts, outgoing and intra-cell, S9, (ho_13g)

Sum(cause_up_qual+ cause_up_level+ cause_down_qual+ cause_down_lev+ cause_distance+ cause_msc_invoc+ cause_intfer_up+ cause_intfer_dwn+ cause_umbr+ cause_pbdgt+ cause_omc+ cause_traffic+ cause_dir_retry+ cause_pre_emption+ cause_field_drop+ cause_low_distance+ cause_bad_CI+ cause_good_CI+ ho_due_slow_mov_ms+ ho_due_ms_slow_speed ; new S5,S6,S7 causes+ ho_due_ms_high_speed+ ho_att_due_switch_circ_pool+ ho_att_due_erfd+ ho_att_due_to_bsc_contr_trho.; new S8,S9 causes+ ho_att_due_to_dadlb+ ho_att_due_to_gprs+ ho_att_due_to_hscsd)


Figure 192. HO attempts, outgoing and intra-cell, S9, (ho_13g)

TCH requests for HO (ho_14a)

sum(tch_req-tch_call_req-tch_fast_req)


Figure 193. TCH requests for HO (ho_14a)



TCH requests for HO (ho_14b)

Note: When you are using IUO, you can see that the number of TCHrequests due to HO attempts goes up (even tenfold).

sum(a.tch_req-a.tch_call_req-tch_fast_req)-sum(b.bsc_i_unsucc_a_int_circ_type+b.msc_controlled_in_ho

+b.ho_unsucc_a_int_circ_type)

Counters from table(s):a = p_nbsc_trafficb = p_nbsc_ho

Figure 194. TCH requests for HO (ho_14b)

TCH seizures for HO (ho_15)

sum(tch_ho_seiz)


Figure 195. TCH seizures for HO (ho_15)

TCH-TCH HO attempts (ho_16)

Use: Used on the BTS level. If used on the area level, it wouldresult in double counting of inter-cell HOs.

sum( msc_o_tch_tch_at + msc_i_tch_tch_at+ bsc_o_tch_tch_at + bsc_i_tch_tch_at+ cell_tch_tch_at)


Figure 196. TCH-TCH HO attempts (ho_16)

SDCCH-TCH HO attempts (ho_17)


sum( msc_o_sdcch_tch_at + msc_i_sdcch_tch_at+ bsc_o_sdcch_tch_at + bsc_i_sdcch_tch_at+ cell_sdcch_tch_at)


Figure 197. SDCCH-TCH HO attempts (ho_17)



SDCCH-SDCCH HO attempts (ho_18)


sum( msc_o_sdcch_at + msc_i_sdcch_at+ bsc_o_sdcch_at + bsc_i_sdcch_at+ cell_sdcch_at)


Figure 198. SDCCH-SDCCH HO attempts (ho_18)

TCH-TCH HO success (ho_19)


sum( msc_o_tch_tch + msc_i_tch_tch+ bsc_o_tch_tch + bsc_i_tch_tch+ cell_tch_tch)


Figure 199. TCH-TCH HO successes (ho_19)

SDCCH-TCH HO success (ho_20)


sum( msc_o_sdcch_tch + msc_i_sdcch_tch+ bsc_o_sdcch_tch + bsc_i_sdcch_tch+ cell_sdcch_tch)


Figure 200. SDCCH-TCH HO successes (ho_20)

SDCCH-SDCCH HO success (ho_21)


sum( msc_o_sdcch + msc_i_sdcch+ bsc_o_sdcch + bsc_i_sdcch+ cell_sdcch)




Figure 201. SDCCH-SDCCH HO successes (ho_21)

MSC controlled HO attempts (ho_22)


sum( msc_o_tch_tch_at + msc_i_tch_tch_at+ msc_o_sdcch_tch_at + msc_i_sdcch_tch_at+ msc_o_sdcch_at + msc_i_sdcch_at)


Figure 202. MSC controlled HO attempts (ho_22)

BSC controlled HO attempts (ho_23)


sum( bsc_o_tch_tch_at + bsc_i_tch_tch_at+ bsc_o_sdcch_tch_at + bsc_i_sdcch_tch_at+ bsc_o_sdcch_at + bsc_i_sdcch_at)


Figure 203. BSC controlled HO attempts (ho_23)

Intra-cell HO attempts (ho_24)

sum(cell_tch_tch_at+ cell_sdcch_tch_at+ cell_sdcch_at)


Figure 204. Intra-cell HO attempts (ho_24)

MSC controlled HO success (ho_25)


sum( msc_o_tch_tch + msc_i_tch_tch+ msc_o_sdcch_tch + msc_i_sdcch_tch+ msc_o_sdcch_ + msc_i_sdcch)




Figure 205. MSC controlled HO successes (ho_25)

BSC controlled HO success (ho_26)


sum( bsc_o_tch_tch + bsc_i_tch_tch+ bsc_o_sdcch_tch + bsc_i_sdcch_tch+ bsc_o_sdcch + bsc_i_sdcch)


Figure 206. BSC controlled HO successes (ho_26)

Intra-cell HO success (ho_27)

sum(cell_tch_tch + cell_sdcch_tch + cell_sdcch)


Figure 207. Intra-cell HO successes (ho_27)

MSC incoming HO success (ho_28)

sum(msc_i_tch_tch+msc_i_sdcch_tch+msc_i_sdcch)


Figure 208. MSC incoming HO successes (ho_28)

MSC outgoing HO success (ho_29)

sum(msc_o_tch_tch+msc_o_sdcch_tch+msc_o_sdcch)


Figure 209. MSC outgoing HO successes (ho_29)



BSC incoming HO success (ho_30)

sum(bsc_i_tch_tch+bsc_i_sdcch_tch+bsc_i_sdcch)


Figure 210. BSC incoming HO successes (ho_30)

BSC outgoing HO success (ho_31)

sum(bsc_o_tch_tch+bsc_o_sdcch_tch+bsc_o_sdcch)


Figure 211. BSC outgoing HO successes (ho_31)

Incoming HO success (ho_32)

sum(msc_i_succ_ho+bsc_i_succ_ho)


Figure 212. Incoming HO success (ho_32)

Outgoing HO success (ho_33)

sum(msc_o_succ_ho+ bsc_o_succ_ho)


Figure 213. Outgoing HO successes (ho_33)

Outgoing HO attempts (ho_34)

sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at + msc_o_sdcch_at+bsc_o_tch_tch_at + bsc_o_sdcch_tch_at + bsc_o_sdcch_at)


Figure 214. Outgoing HO attempts (ho_34)



Incoming HO attempts (ho_35)

sum(msc_i_tch_tch_at + msc_i_sdcch_tch_at + msc_i_sdcch_at+bsc_i_tch_tch_at + bsc_i_sdcch_tch_at + bsc_i_sdcch_at)


Figure 215. Incoming HO attempts (ho_35)

Outgoing SDCCH-SDCCH HO attempts (ho_36)

sum(msc_o_sdcch_at+bsc_o_sdcch_at)


Figure 216. Outgoing SDCCH-SDCCH HO attempts (ho_36)

Incoming SDCCH-SDCCH HO attempts (ho_37)

sum(msc_i_sdcch_at+bsc_i_sdcch_at)


Figure 217. Incoming SDCCH-SDCCH HO attempts (ho_37)

Outgoing SDCCH-TCH HO attempts (ho_38)

sum(msc_o_sdcch_tch_at+bsc_o_sdcch_tch_at)


Figure 218. Outgoing SDCCH-TCH HO attempts (ho_38)

Incoming SDCCH-TCH HO attempts (ho_39)

sum(msc_i_sdcch_tch_at+bsc_i_sdcch_tch_at)


Figure 219. Incoming SDCCH-TCH HO attempts (ho_39)



Outgoing TCH-TCH HO attempts (ho_40)

sum(msc_o_tch_tch_at+bsc_o_tch_tch_at)


Figure 220. Outgoing TCH-TCH HO attempts (ho_40)

Incoming TCH-TCH HO attempts (ho_41)

sum(msc_i_tch_tch_at+bsc_i_tch_tch_at)


Figure 221. Incoming TCH-TCH HO attempts (ho_41)

Outgoing SDCCH-SDCCH HO success (ho_42)

sum(msc_o_sdcch+bsc_o_sdcch)


Figure 222. Outgoing SDCCH-SDCCH HO success (ho_42)

Incoming SDCCH-SDCCH HO success (ho_43)

sum(msc_i_sdcch+bsc_i_sdcch)


Figure 223. Incoming SDCCH-SDCCH HO success (ho_43)

Outgoing SDCCH-TCH HO success (ho_44)

sum(msc_o_sdcch_tch+bsc_o_sdcch_tch)


Figure 224. Outgoing SDCCH-TCH HO success (ho_44)



Incoming SDCCH-TCH HO success (ho_45)

sum(msc_i_sdcch tch+bsc_i_sdcch_tch)


Figure 225. Incoming SDCCH-TCH HO success (ho_45)

Outgoing TCH-TCH HO success (ho_46)

sum(msc_o_tch_tch+bsc_o_tch_tch)


Figure 226. Outgoing TCH-TCH HO success (ho_46)

Incoming TCH-TCH HO success (ho_47)

sum(msc_i_tch_tch+bsc_i_tch_tch)


Figure 227. Incoming TCH-TCH HO success (ho_47)

Intra-cell HO share, S1 (ho_48)

sum(cell_sdcch_tch_at+cell_tch_tch_at+cell_sdcch_at)100 * -------------------------------------------------------- %

sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch_at+msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at+bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at+bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch_at+cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)

Unit: %Counters from table(s):p_nbsc_ho

Figure 228. Intra-cell HO share, S1 (ho_48)

MSC controlled incoming HO attempts (ho_49)

sum(msc_i_tch_tch_at + msc_i_sdcch_tch_at + msc_i_sdcch_at)




Figure 229. MSC controlled incoming HO attempts (ho_49)

2.19 Handover failure % (hfr)

Total HO failure %, S1 (hfr_1)

Use: Works best on the BTS level, but is usable on both the areaand the cell level.

Experiences on use: In a good network the value can be less than 3 per cent,whereas in a very bad network values higher than 15 per centmay occur. When IUO is used, this formula shows high valuesdue to highly failing intra-cell handovers between layers incongested cells.

Known problems: This formula emphasises the non-intra-cell handovers sincethey are counted twice. This causes no problems on the celllevel, whereas on the area level problems may occur.Blocking is included. Blocking makes this indicator showhigh values especially in the case of IUO, but it does notnecessarily mean that there are problems.

HO failures100 * --------------- %

HO attempts

HO attempts - successful HOs= 100 * ------------------------------- %

HO attempts

successful HOs= 100 * (1- -------------- ) %

HO attempts

sum(msc_i_succ_ho+msc_o_succ_ho+bsc_i_succ_ho+bsc_o_succ_ho+cell_succ_ho)

= 100 * (1- -------------------------------------------------------------------)%sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch_at+

msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at+bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at+bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch_at+cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)


Figure 230. Total HO failure %, S1 (hfr_1)

Total HO failure %, S1 (hfr_2)

Use: On the area or network level.



Experiences on use: In a network gave the result of 8 % instead of 7 % of hfr_1.In a good network the value can be less than 3 per cent, whilein a very bad network values higher than 15 per cent mayoccur.If Directed Retry is enabled, the MS may, when congestion ofthe source cell SDCCH occurs, be moved from the best cell toa worse one. Then the MS tries to make a handover back butfails if the first cell is still congested. This leads toincrementation of the HO failure ratio.Common reasons for a handover to fail:- incorrect parameter settings of adjacencies- badly defined neighbours (UL coverage becomes a problem)- UL coverage in general. Cell imbalanced.- TCH blocking in the target cell- UL interference (target BTS never gets the HO access)

Known problems: Blocking is included. Blocking makes this indicator showhigh values especially in the case of IUO, but it does notnecessarily mean that there are some technical problems.Calls that are cleared by the MS user during the HO processincrement the attempt counters but cannot be compensated inthe numerator. (XX2)HO that is interrupted due to another procedure (e.g.assignment) increments the attempt counters but cannot becompensated in the numerator.(XX3)

HO failures100* --------------- %

HO attempts

HO attempts - successful HOs= 100 * --------------------------------- %

HO attempts

successful HOs= 100 * (1- -------------- ) %

HO attempts

sum(msc_o_succ_ho + bsc_o_succ_ho + cell_succ_ho) + XX2+ XX3= 100 * (1- -------------------------------------------------------------------)%

sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at+bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at+cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)


Figure 231. Total HO failure %, S1 (hfr_2)

Intra-cell HO failure share, S1 (hfr_3a)

Use: Used on the BTS level. The results are equal to hfr_3c.

Intra-cell HO failures100* (--------------------------------------) %



All HO attempts

sum(cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch-at)-sum(cell_succ_ho)

= 100* (--------------------------------------------------------------) %sum(msc_i_tch_tch_at + msc_i_sdcch_tch_at+ msc_i_sdcch_at)

+ sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at+ msc_o_sdcch_at)+ sum(bsc_i_tch_tch_at + bsc_i_sdcch_tch_at+ bsc_i_sdcch_at)+ sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at+ bsc_o_sdcch_at)+ sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 232. Intra-cell HO failure share, S1 (hfr_3a)

Intra-cell HO failure share, S1 (hfr_3b)

Use: Used on the area or network level. The results are equal tohfr_3d.


All HO attempts

sum(cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch-at)-sum(cell_succ_ho)




Figure 233. Intra-cell HO failure share, S1 (hfr_3b)

Intra-cell HO failure share, S1 (hfr_3c)

Use: Used on the BTS level. The results are equal to hfr_3a.


All HO attempts

sum(cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)-sum(cell_tch_tch+cell_sdcch_tch+cell_sdcch)






Figure 234. Intra-cell HO failure share, S1 (hfr_3c)

Intra-cell HO failure share, S1 (hfr_3d)

Use: On the area or network level. The results are equal to hfr_3b.


All HO attempts

sum(cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)-sum(cell_tch_tch+cell_sdcch_tch+cell_sdcch)

= 100* (--------------------------------------------------------------) %sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at+ msc_o_sdcch_at)

+ sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at+ bsc_o_sdcch_at)+ sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 235. Intra-cell HO failure share, S1 (hfr_3d)

Incoming MSC ctrl HO failure %, S1 (hfr_4)

MSC controlled incoming HO successes100* (1- --------------------------------------) %

MSC controlled incoming HO attempts

sum(msc_i_tch_tch+msc_i_sdcch_tch+msc_i_sdcch)= 100* (1- -----------------------------------------------------------) %



Figure 236. Incoming MSC ctrl HO failure %, S1 (hfr_4)

Incoming MSC ctrl HO failure share, S1 (hfr_4a)

Use: On the BTS level. The results are equivalent to hfr_4c.

MSC controlled incoming HO failures100* (--------------------------------------) %

All HO attempts

sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch-at)-sum(msc_i_succ_ho)


+ sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at+ msc_o_sdcch_at)+ sum(bsc_i_tch_tch_at + bsc_i_sdcch_tch_at+ bsc_i_sdcch_at)+ sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at+ bsc_o_sdcch_at)



+ sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 237. Incoming MSC ctrl HO failure share, S1 (hfr_4a)

Incoming MSC ctrl HO failure share, S1 (hfr_4b)



All HO incoming attempts

sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch-at)-sum(msc_i_succ_ho)


+ sum(bsc_i_tch_tch_at + bsc_i_sdcch_tch_at+ bsc_i_sdcch_at)+ sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 238. Incoming MSC ctrl HO failure share, S1 (hfr_4b)

Incoming MSC ctrl HO failure share, S1 (hfr_4c)



All HO attempts

sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch_at)- sum(msc_i_tch_tch+msc_i_sdcch_tch+msc_i_sdcch)




Figure 239. Incoming MSC ctrl HO failure share, S1 (hfr_4c)

Incoming MSC ctrl HO failure share, S1 (hfr_4d)

Use: Used on the area or network level. The results are equal tohfr_4b.




All HO incoming attempts

sum(msc_i_tch_tch_at+msc_i_sdcch_tch_at+msc_i_sdcch_at)- sum(msc_i_tch_tch+msc_i_sdcch_tch+msc_i_sdcch)




Figure 240. Incoming MSC ctrl HO failure share, S1 (hfr_4d)

Outgoing MSC ctrl HO failure ratio %, S1 (hfr_5)

sum(msc_o_tch_tch+msc_o_sdcch_tch+msc_o_sdcch)100 - 100 * ------------------------------------------------------------ %

sum(nvl(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at,0)


Figure 241. Outgoing MSC ctrl HO failure ratio %, S1 (hfr_5)

Outgoing MSC ctrl HO failure share %, S1 (hfr_5a)


MSC controlled outgoing HO failures100* (--------------------------------------) %

All HO attempts

sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch-at)-sum(msc_o_succ_ho)




Figure 242. Outgoing MSC ctrl HO failure share %, S1 (hfr_5a)

Outgoing MSC ctrl HO failure share %, S1 (hfr_5b)





All HO outgoing attempts

sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch-at)-sum(msc_o_succ_ho)




Figure 243. Outgoing MSC ctrl HO failure share %, S1 (hfr_5b)

Outgoing MSC ctrl HO failure share %, S1 (hfr_5c)



All HO attempts

sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at)- sum(msc_o_tch_tch+msc_o_sdcch_tch+msc_o_sdcch)




Figure 244. Outgoing MSC ctrl HO failure share %, S1 (hfr_5c)

Outgoing MSC ctrl HO failure share %, S1 (hfr_5d)



All HO outgoing attempts







Figure 245. Outgoing MSC ctrl HO failure share %, S1 (hfr_5d)

Incoming BSC ctrl HO failure %, S1 (hfr_6)

BSC controlled incoming HO successes100* (1- --------------------------------------) %

BSC controlled incoming HO attempts

sum(bsc_i_tch_tch+bsc_i_sdcch_tch+bsc_i_sdcch)= 100* (1- ------------------------------------------------------------) %

sum(bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch_at)


Figure 246. Incoming BSC ctrl HO failure %, S1 (hfr_6)

Incoming BSC ctrl HO failure share %, S1 (hfr_6a)


BSC controlled incoming HO failures100* (--------------------------------------) %

All HO attempts

sum(bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch-at)-sum(bsc_i_succ_ho)




Figure 247. Incoming BSC ctrl HO failure share %, S1 (hfr_6a)

Incoming BSC ctrl HO failure %, S1 (hfr_6b)

Use: Use on the area or network level. The results are equal tohfr_6d.


All incoming HO attempts

sum(bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch-at)-sum(bsc_i_succ_ho)






Figure 248. Incoming BSC ctrl HO failure %, S1 (hfr_6b)

Incoming BSC ctrl HO failure share %, S1 (hfr_6c)



All HO attempts

sum(bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch_at)-sum(bsc_i_tch_tch+bsc_i_sdcch_tch+bsc_i_sdcch)




Figure 249. Incoming BSC ctrl HO failure share %, S1 (hfr_6c)

Incoming BSC ctrl HO failure %, S1 (hfr_6d)



All incoming HO attempts

sum(bsc_i_tch_tch_at+bsc_i_sdcch_tch_at+bsc_i_sdcch_at)-sum(bsc_i_tch_tch+bsc_i_sdcch_tch+bsc_i_sdcch)




Figure 250. Incoming BSC ctrl HO failure %, S1 (hfr_6d)

Outgoing BSC ctrl HO failure share, S1 (hfr_7)

BSC controlled outgoing HO successes100* (1- --------------------------------------) %



BSC controlled outgoing HO attempts

sum(bsc_o_tch_tch+bsc_o_sdcch_tch+bsc_o_sdcch)= 100* (1- ------------------------------------------------------------) %

sum(bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at)


Figure 251. Outgoing BSC ctrl HO failure share, S1 (hfr_7)

Outgoing BSC ctrl HO failure share, S1 (hfr_7a)


BSC controlled outgoing HO failures100* (--------------------------------------) %

All HO attempts

sum(bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch-at)-sum(bsc_o_succ_ho)




Figure 252. Outgoing BSC ctrl HO failure share, S1 (hfr_7a)

Outgoing BSC ctrl HO failure share, S1 (hfr_7b)



All HO attempts

sum(bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch-at)-sum(bsc_o_succ_ho)




Figure 253. Outgoing BSC ctrl HO failure share, S1 (hfr_7b)



Outgoing BSC ctrl HO failure share, S1 (hfr_7c)



All HO attempts

sum(bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at)-sum(bsc_o_tch_tch+bsc_o_sdcch_tch+bsc_o_sdcch)




Figure 254. Outgoing BSC ctrl HO failure share, S1 (hfr_7c)

Outgoing BSC ctrl HO failure share, S1 (hfr_7d)

Use: On the area or PLMN level. The results are equal to hfr_7b.


All HO attempts

sum(bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at)-sum(bsc_o_tch_tch+bsc_o_sdcch_tch+bsc_o_sdcch)




Figure 255. Outgoing BSC ctrl HO failure share, S1 (hfr_7d)

Internal inter HO failure %, S4 (hfr_8)

sum(int_inter_ho_source_fail)100* (-----------------------------------------------) %

sum(int_inter_ho_source_fail+int_inter_ho_succ)


Figure 256. Internal inter HO failure %, S4 (hfr_8)



Internal intra HO failure %, S4 (hfr_9)

sum(int_intra_ho_source_fail)100* (-----------------------------------------------) %

sum(int_intra_ho_source_fail+int_intra_ho_succ)


Figure 257. Internal intra HO failure %, S4 (hfr_9)

External source HO failure %, S4 (hfr_10)

sum(ext_ho_source_fail)100* (-----------------------------------------------) %

sum(ext_ho_source_fail+ ext_ho_source_succ)


Figure 258. External source HO failure %, S4 (hfr_10)

HO failure % from super to regular, S4 (hfr_12)

Use: Ratio of all other failures than ’blocked’ to all HO attemptsfrom super to regular TRX.

sum(ho_fail_to_reg_due_ret+ ho_fail_to_reg_ms_lost+ ho_fail_to_reg_freq)100* (-------------------------------------------------------------------------) %

sum(ho_att_to_reg_freq)


Figure 259. HO failure % from super to regular, S4 (hfr_12)

HO failure % from regular to super, S4 (hfr_13)

sum(ho_fail_from_reg_due_ret+ho_fail_from_reg_ms_lost+ho_fail_from_reg_freq)100*(------------------------------------------------------------------------) %

sum(ho_att_from_reg_freq)


Figure 260. HO failure % from regular to super, S4 (hfr_13)



Share of HO failures from regular to super due to return, S4 (hfr_14)

sum(ho_fail_from_reg_due_ret)100* (------------------------------------------------------------------------) %

sum(ho_fail_from_reg_due_ret+ho_fail_from_reg_ms_lost+ho_fail_from_reg_freq)


Figure 261. Share of HO failures from regular to super due to return, S4 (hfr_14)

Share of HO failures from regular to super due to MS lost, S4 (hfr_15)

Use: Ratio of ’MS Lost’ failures to all HO attempts (blocked HOsexcluded) in HOs from regular to super TRX.

sum(ho_fail_from_reg_ms_lost)100* (--------------------------------------------------------------------------)%

sum(ho_fail_from_reg_due_ret+ ho_fail_from_reg_ms_lost+ ho_fail_from_reg_freq)


Figure 262. Share of HO failures from regular to super due to MS lost, S4(hfr_15)

Share of HO failures from regular to super due to another cause, S4 (hfr_16)

Use: Ratio of any other HO failures than ’return’ and ’MS lost’ toall HO attempts (blocked HOs excluded) in HOs from regularto super TRX.

sum(ho_fail_from_reg_freq)100* (--------------------------------------------------------------------------) %

sum(ho_fail_from_reg_due_ret+ ho_fail_from_reg_ms_lost+ ho_fail_from_reg_freq)


Figure 263. Share of HO failures from regular to super due to another cause, S4(hfr_16)

Share of HO failures from super to regular due to return, S4 (hfr_17)

Use: Ratio of ’return’ HO failures to all HO attempts (blocked HOsexcluded) in HOs from super to regular TRX.

sum(ho_fail_to_reg_due_ret)100* (--------------------------------------------------------------------------) %

sum(ho_fail_to_reg_due_ret+ ho_fail_to_reg_ms_lost+ ho_fail_to_reg_freq)




Figure 264. Share of HO failures from super to regular due to return, S4 (hfr_17)

Share of HO failures from super to regular due to MS lost, S4 (hfr_18)

Use: Ratio of ’MS lost’ HO failures to all HO attempts (blockedHOs excluded) in HOs from super to regular TRX.

sum(ho_fail_to_reg_ms_lost)100* (--------------------------------------------------------------------------) %

sum(ho_fail_to_reg_due_ret+ ho_fail_to_reg_ms_lost+ ho_fail_to_reg_freq)


Figure 265. Share of HO failures from super to regular due to MS lost, S4(hfr_18)

Share of HO failures from super to regular due to another cause, S4 (hfr_19)

Experiences on use: Includes HO failures due to any other reason than ’return’and ’MS lost’.

Use: Ratio of ’Other cause’ HO failures to all HO attempts(blocked HOs excluded) in HOs from super to regular TRX.

sum(ho_fail_to_reg_freq)100*(-----------------------------------------------------------------------) %

sum(ho_fail_to_reg_due_ret+ho_fail_to_reg_ms_lost+ho_fail_to_reg_freq)


Figure 266. Share of HO failures from super to regular due to another cause, S4(hfr_19)

SDCCH-SDCCH HO failure %, S2 (hfr_20)

Experiences on use: It is better to look at MSC and BSC controlled handoverseparately.

sum(msc_i_sdcch+ msc_o_sdcch+ bsc_i_sdcch+ bsc_o_sdcch+ cell_sdcch)

-sum(msc_i_sdcch_at + msc_o_sdcch_at+ bsc_i_sdcch_at + bsc_o_sdcch_at+ cell_sdcch_at)

100* (----------------------------------------------) %sum(msc_i_sdcch_at + msc_o_sdcch_at

+ bsc_i_sdcch_at + bsc_o_sdcch_at+ cell_sdcch_at)




Figure 267. SDCCH-SDCCH HO failure %, S2 (hfr_20)

SDCCH-TCH HO failure %, S2 (hfr_21)

Use: These are Directed Retry.

sum(msc_i_sdcch_tch+ msc_o_sdcch_tch+ bsc_i_sdcch_tch+ bsc_o_sdcch_tch+ cell_sdcch_tch)

- sum(msc_i_sdcch_tch_at + msc_o_sdcch_tch_at+ bsc_i_sdcch_tch_at + bsc_o_sdcch_tch_at+ cell_sdcch_tch_at)

100* (-----------------------------------------------------) %sum(msc_i_sdcch_tch_at + msc_o_sdcch_tch_at

+ bsc_i_sdcch_tch_at + bsc_o_sdcch_tch_at+ cell_sdcch_tch_at)


Figure 268. SDCCH-TCH HO failure %, S2 (hfr_21)

TCH-TCH HO failure %, S2 (hfr_22)

sum(msc_i_tch_tch+ msc_o_tch_tch+ bsc_i_tch_tch+ bsc_o_tch_tch+ cell_tch_tch)

-sum(msc_i_tch_tch_at + msc_o_tch_tch_at+ bsc_i_tch_tch_at + bsc_o_tch_tch_at+ cell_tch_tch_at)

100* (--------------------------------------------------) %sum(msc_i_tch_tch_at + msc_o_tch_tch_at

+ bsc_i_tch_tch_at + bsc_o_tch_tch_at+ cell_tch_tch_at)


Figure 269. TCH-TCH HO failure %, S2 (hfr_22)

SDCCH-SDCCH incoming HO failure %, S2 (hfr_23)

sum(msc_i_sdcch_at+ bsc_i_sdcch_at) - sum(msc_i_sdcch + bsc_i_sdcch)100* ------------------------------------------------------------------------- %

sum(msc_i_sdcch_at+ bsc_i_sdcch_at)


Figure 270. SDCCH-SDCCH incoming HO failure %, S2 (hfr_23)



SDCCH-SDCCH outgoing HO failure ratio, S2 (hfr_24)

sum(msc_o_sdcch_at+ bsc_o_sdcch_at) - sum(msc_o_sdcch + bsc_o_sdcch)100* ------------------------------------------------------------------------ %

sum(msc_o_sdcch_at+ bsc_o_sdcch_at)


Figure 271. SDCCH-SDCCH outgoing HO failure ratio, S2 (hfr_24)

SDCCH-TCH incoming HO failure %, S2 (hfr_25)

sum(msc_i_sdcch_tch_at+ bsc_i_sdcch_tch_at)-sum(msc_i_sdcch_tch + bsc_i_sdcch_tch)

100* (------------------------------------------------) %sum(msc_i_sdcch_tch_at+ bsc_i_sdcch_tch_at)


Figure 272. SDCCH-TCH incoming HO failure %, S2 (hfr_25)

SDCCH-TCH outgoing HO failure %, S2 (hfr_26)

sum(msc_o_sdcch_tch_at+ bsc_o_sdcch_tch_at)-sum(msc_o_sdcch_tch + bsc_o_sdcch_tch)

100* ----------------------------------------------- %sum(msc_o_sdcch_tch_at+ bsc_o_sdcch_tch_at)


Figure 273. SDCCH-TCH outgoing HO failure %, S2 (hfr_26)

TCH-TCH incoming HO failure %, S2 (hfr_27)

sum(msc_i_tch_tch_at+ bsc_i_tch_tch_at) - sum(msc_i_tch_tch + bsc_i_tch_tch)100* --------------------------------------------------------------------------- %

sum(msc_i_tch_tch_at+ bsc_i_tch_tch_at)


Figure 274. TCH-TCH incoming HO failure %, S2 (hfr_27)



TCH-TCH outgoing HO failure %, S2 (hfr_28)

sum(msc_o_tch_tch_at+ bsc_o_tch_tch_at) - sum(msc_o_tch_tch + bsc_o_tch_tch)100* --------------------------------------------------------------------------- %

sum(msc_o_tch_tch_at+ bsc_o_tch_tch_at)


Figure 275. TCH-TCH outgoing HO failure %, S2 (hfr_28)

MSC ctrl HO failure %, blocking (hfr_29)

msc_o_fail_lack100* ------------------------------------------------------ %

msc_o_sdcch_at + msc_o_sdcch_tch_at + msc_o_tch_tch_at


Figure 276. MSC ctrl HO failure %, blocking (hfr_29)

MSC ctrl HO failure %, not allowed (hfr_30)

msc_o_not_allwd100* ------------------------------------------------------ %



Figure 277. MSC ctrl HO failure %, not allowed (hfr_30)

MSC ctrl HO failure %, return to old (hfr_31)

msc_o_fail_ret100* ------------------------------------------------------ %



Figure 278. MSC ctrl HO failure %, return to old (hfr_31)

MSC ctrl HO failure %, call clear (hfr_32)

msc_o_call_clr100* ------------------------------------------------------ %





Figure 279. MSC ctrl HO failure %, call clear (hfr_32)

MSC ctrl HO failure %, end HO (hfr_33)

msc_o_end_of_ho100* ------------------------------------------------------ %



Figure 280. MSC ctrl HO failure %, end HO (hfr_33)

MSC ctrl HO failure %, end HO BSS (hfr_34)

msc_o_end_ho_bss100* ------------------------------------------------------ %



Figure 281. MSC ctrl HO failure %, end HO BSS (hfr_34)

MSC ctrl HO failure %, wrong A interface (hfr_35)

Use: Relates to the congestion of A-interface pool resourcesdetected by BSC.

msc_controlled_out_ho100* ------------------------------------------------------ %



Figure 282. MSC ctrl HO failure %, wrong A interface (hfr_35)

MSC ctrl HO failure %, adjacent cell error (hfr_36)

msc_o_adj_cell_id_err_c100* ------------------------------------------------------ %





Figure 283. MSC ctrl HO failure %, adjacent cell error (hfr_36)

BSC ctrl HO failure %, blocking (hfr_37)

bsc_o_fail_lack100* ------------------------------------------------------ %

bsc_o_sdcch_at + bsc_o_sdcch_tch_at + bsc_o_tch_tch_at


Figure 284. BSC ctrl HO failure %, blocking (hfr_37)

BSC ctrl HO failure %, not allowed (hfr_38)

bsc_o_not_allwd100* ------------------------------------------------------ %



Figure 285. BSC ctrl HO failure %, not allowed (hfr_38)

BSC ctrl HO failure %, return to old (hfr_39)

bsc_o_fail_ret100* ------------------------------------------------------ %



Figure 286. BSC ctrl HO failure %, return to old (hfr_39)

BSC ctrl HO failure %, call clear (hfr_40)

bsc_o_call_clr100* ------------------------------------------------------ %





Figure 287. BSC ctrl HO failure %, call clear (hfr_40)

BSC ctrl HO failure %, end HO (hfr_41)

bsc_o_end_of_ho100* ------------------------------------------------------ %



Figure 288. BSC ctrl HO failure %, end HO (hfr_41)

BSC ctrl HO failure %, end HO BSS (hfr_42)

bsc_o_end_ho_bss100* ------------------------------------------------------ %



Figure 289. BSC ctrl HO failure %, end HO BSS (hfr_42)

BSC ctrl HO failure %, wrong A interface (hfr_43)


bsc_o_unsucc_a_int_circ_type100* ------------------------------------------------------ %



Figure 290. BSC ctrl HO failure %, wrong A interface (hfr_43)

BSC ctrl HO drop call % (hfr_44)

bsc_o_drop_calls100* ------------------------------------------------------ %





Figure 291. BSC ctrl HO drop call % (hfr_44)

Intra-cell HO failure %, cell_fail_lack (hfr_45)

cell_fail_lack100* --------------------------------------------------- %

cell_sdcch_at + cell_sdcch_tch_at + cell_tch_tch_at


Figure 292. Intra-cell HO failure %, cell_fail_lack (hfr_45)

Intra-cell HO failure %, not allowed (hfr_46)

cell_not_allwd100* --------------------------------------------------- %



Figure 293. Intra-cell HO failure %, not allowed (hfr_46)

Intra-cell HO failure %, return to old (hfr_47)

cell_fail_ret100* --------------------------------------------------- %



Figure 294. Intra-cell HO failure %, return to old (hfr_47)

Intra-cell HO failure %, call clear (hfr_48)

cell_call_clr100* --------------------------------------------------- %





Figure 295. Intra-cell HO failure %, call clear (hfr_48)

Intra-cell HO failure %, MS lost (hfr_49)

cell_fail_move100* --------------------------------------------------- %



Figure 296. Intra-cell HO failure %, MS lost (hfr_49)

Intra-cell HO failure %, BSS problem (hfr_50)

cell_fail_bss100* --------------------------------------------------- %



Figure 297. Intra-cell HO failure %, BSS problem (hfr_50)

Intra-cell HO failure %, drop call (hfr_51)

cell_drop_calls100* --------------------------------------------------- %



Figure 298. Intra-cell HO failure %, drop call (hfr_51)

HO failure % to adjacent cell (hfr_52)

sum(ho_att_to_adj - ho_succ_to_adj)100* ----------------------------------- %

sum(ho_att_to_adj)



Counters from table(s):p _nbsc_ho_adj

Figure 299. HO failure % to adjacent cell (hfr_52)

HO failure % from adjacent cell (hfr_53)

sum(ho_att_from_adj - ho_succ_from_adj)100* --------------------------------------- %

sum(ho_att_from_adj)

Counters from table(s):p _nbsc_ho_adj

Figure 300. HO failure % from adjacent cell (hfr_53)

HO failure %, blocking excluded (hfr_54a)


/* all HO attempts */sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at

+bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at+cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)

/*successful handovers */-sum(msc_o_succ_ho +bsc_o_succ_ho+cell_succ_ho)

/* handovers failing due to blocking */-sum(msc_o_fail_lack+bsc_o_fail_lack+cell_fail_lack)

100 * ------------------------------------------------------%/* all HO attempts */sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at




Figure 301. HO failure %, blocking excluded (hfr_54a)

HO failure % due to radio interface blocking (hfr_55)


/* handovers failing due to blocking */sum(msc_o_fail_lack+bsc_o_fail_lack+cell_fail_lack)

100 * -------------------------------------------------------- %/* all HO attempts */sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at

+bsc_o_tch_tch_at+bsc_o_sdcch_tch_at+bsc_o_sdcch_at



+cell_tch_tch_at+cell_sdcch_tch_at+cell_sdcch_at)


Figure 302. HO failure % due to radio interface blocking (hfr_55)

Intra-cell HO failure %, wrong A interface (hfr_56)


sum(ho_unsucc_a_int_circ_type)100 x -------------------------------------------------------- %

sum(cell_sdcch_at + cell_sdcch_tch_at + cell_tch_tch_at)


Figure 303. Intra-cell HO failure %, wrong A interface (hfr_56)

Intra-cell HO failure % (hfr_57)

Intra-cell HO successes100* (1- -------------------------) %

Intra-cell HO attempts

sum(cell_tch_tch + cell_sdcch_tch+ cell_sdcch)= 100* (1- ---------------------------------------------------------------) %

sum(cell_tch_tch_at + cell_sdcch_tch_at + cell_sdcch_at)


Figure 304. Intra-cell HO failure % (hfr_57)

HO failures to target cell, S6 (hfr_58)

Use: On the adjacency level. Gives the failure % of the real (nonblocked) HO attempts.

Known problems: Not accurate because of:1) Calls that are cleared by MS user during the HO process.The ho_att_to_adj counter is incremented and cannot becompensated in the numerator.2) HO that is interrupted due to another procedure (e.g.assignment) increments attempt counters but cannot becompensated in the numerator.

sum(ho_att_to_adj-ho_succ_to_adj-ho_fail_res_to_adj)100* -------------------------------------------------------- %

sum(ho_att_to_adj- ho_fail_res_to_adj)



Counters from table(s):p_nbsc_ho_adj

Figure 305. HO failures to target cell, S6 (hfr_58)

HO failures from target cell, S6 (hfr_59)

Use: Used on the adjacency level. Gives failure % of the real(unblocked) HO attempts.

Known problems: Not accurate because of:1. Calls that are cleared by MS user during the HO

process. The ho_att_to_adj counter is incrementedand cannot be compensated in the numerator.

2. HO that is interrupted due to other procedure (e.g.assignment) increments attempt counters but cannot becompensated in the numerator.

sum(ho_att_from_adj-ho_succ_from_adj-ho_fail_res_from_adj)100* ----------------------------------------------------------- %

sum(ho_att_from-adj- ho_fail_res_from_adj)


Figure 306. HO failures from target cell, S6 (hfr_59)

HO drop ratio (hfr_68)

Use: Defines how big a share of the started handovers is dropped.Indicates the quality of the handovers.

Sum(bsc_o_drop_calls+msc_o_call_drop_ho +cell_drop_calls)100* --------------------------------------------------------- %

/* all HO attempts */sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at


/* handovers failing due to blocking */-sum(msc_o_fail_lack+bsc_o_fail_lack+cell_fail_lack)/* handovers failing due to not allowed */

-sum(msc_o_not_allwd+bsc_o_not_allwd+cell_not_allwd)/* wrong Aif circuit type */

-sum(bsc_o_unsucc_a_int_circ_type+msc_controlled_out_ho+ho_unsucc_a_int_circ_type

)


Figure 307. HO drop ratio (hfr_68)



HO failures to target WCDMA cell, S10.5 (hfr_69)

Use: Gives the failure percentage of the non-blocked HO attempts.

sum(ho_att_wcdma_ran_cell-ho_succ_wcdma_ran_cell-ho_fail_due_res_wcdma_ran)

100 * ------------------------------- %sum(ho_att_wcdma_ran_cell

- ho_fail_due_res_wcdma_ran)

Counters from table(s):p_nbsc_utran_ho_adj_cell

Figure 308. HO failures to target WCDMA cell, S10.5 (hfr_69)

HO failures from target WCDMA cell, S10.5 (hfr_70)

Use: Gives the failure percentage of the non-blocked HO attempts.

sum(ho_att_from_wcdma_ran-ho_succ_from_wcdma_ran-ho_fail_due_res_wcdma_ran_cell)

100 * ------------------------------------ %sum(ho_att_from_wcdma_ran

-ho_fail_due_res_wcdma_ran_cell)


Figure 309. HO failures from target WCDMA cell, S10.5 (hfr_70)

Intra-Segment SDCCH-SDCCH HO failure ratio from BCCH to non-BCCH layer, BSC level, S10.5 (hfr_71)

Use: Gives intra-segment SDCCH-SDCCH HO failure ratio fromBCCH to non-BCCH layer.

Known problems: Timing difference in two measurements.

sum(b.INTRA_INTER_BAND_SDCCH_HANDOVER)100 - 100* -------------------------------------- %

sum(a.HO_ATT_INTER_BAND_SDCCH)

a = p_nbsc_hob = p_nbsc_cc_pm

Counters from table(s):p_nbsc_hop_nbsc_cc_pm

Unit: %

Figure 310. Intra-Segment SDCCH-SDCCH HO failure ratio from BCCH to non-BCCH layer, BSC level, S10.5 (hfr_71)



Intra-segment SDCCH-SDCCH HO failure ratio between BTS types, BSClevel, S10.5 (hfr_72)

Use: Gives intra-segment SDCCH-SDCCH HO failure ratiobetween BTS types.

Known problems: See hfr_71.

sum(b.INTRA_INTER_BTS_TYPE_TCH)100 - 100* -------------------------------------- %

sum(a.HO_ATT_INTER_BTS_TYPE_TCH)



Unit: %

Figure 311. IIntra-segment SDCCH-SDCCH HO failure ratio between BTStypes, BSC level, S10.5 (hfr_72)

Intra-segment TCH-TCH HO failure ratio between bands (due to load), BSClevel, S10.5 (hfr_73)

Use: Gives intra-segment TCH-TCH HO failure ratio betweenbands (due to load reason).


sum(b.INTRA_INTER_BAND_TCH)100 - 100* -------------------------------------- %

sum(a.HO_ATT_INTER_BAND_TCH)


Counters from table(s):p_nbsc_ho p_nbsc_cc_pm

Figure 312. Intra-segment TCH-TCH HO failure ratio between bands (due toload), BSC level, S10.5 (hfr_73)

Intra-segment TCH-TCH HO failure ratio between bands (due to downlinksignal level), BSC level, S10.5 (hfr_74)

Use: Gives intra-segment TCH-TCH HO failure ratio betweenbands (due to downlink signal level).


sum(b.INTRA_INTER_BAND_DUE_LEV)100 - 100* -------------------------------------- %

sum(a.HO_ATTEMPT_INTERBAND_DUE_LEVEL)

a = p_nbsc_ho



b = p_nbsc_cc_pm


Unit: %

Figure 313. Intra-segment TCH-TCH HO failure ratio between bands (due todownlink signal level), BSC level, S10.5 (hfr_74)

Intra-segment TCH-TCH HO failure ratio between BTS types (due to load),BSC level, S10.5 (hfr_75)

Use: Gives intra-segment TCH-TCH HO failure ratio betweenBTS types (due to load).


sum(b.INTRA_INTER_BTS_TYPE_TCH)100 - 100* -------------------------------------- %

sum(a.HO_ATT_INTER_BTS_TYPE_TCH)



Unit: %

Figure 314. Intra-segment TCH-TCH HO failure ratio between BTS types (due toload), BSC level, S10.5 (hfr_75)

2.20 Handover success % (hsr)

MSC controlled outgoing SDCCH-SDCCH HO success %, S1 (hsr_1)

sum(msc_o_sdcch)100 * ------------------- %

sum(msc_o_sdcch_at)


Figure 315. MSC controlled outgoing SDCCH-SDCCH HO success %, S1(hsr_1)

MSC controlled outgoing SDCCH-TCH HO success %, S1 (hsr_2)

sum(msc_o_sdcch_tch)



100 * ----------------------- %sum(msc_o_sdcch_tch_at)


Figure 316. MSC controlled outgoing SDCCH-TCH HO success %, S1 (hsr_2)

MSC controlled outgoing TCH-TCH HO success %, S1 (hsr_3)

sum(msc_o_tch_tch)100 * ----------------------- %

sum(msc_o_tch_tch_at)


Figure 317. MSC controlled outgoing TCH-TCH HO success %, S1 (hsr_3)

BSC controlled outgoing SDCCH-SDCCH HO success %, S1 (hsr_4)

sum(bsc_o_sdcch)100 * ------------------- %

sum(bsc_o_sdcch_at)


Figure 318. BSC controlled outgoing SDCCH-SDCCH HO success %, S1(hsr_4)

BSC controlled outgoing SDCCH-TCH HO success %, S1 (hsr_5)

sum(bsc_o_sdcch_tch)100 * ----------------------- %

sum(bsc_o_sdcch_tch_at)


Figure 319. BSC controlled outgoing SDCCH-TCH HO success %, S1 (hsr_5)

BSC controlled outgoing TCH-TCH HO success %, S1 (hsr_6)

sum(bsc_o_tch_tch)100 * --------------------- %

sum(bsc_o_tch_tch_at)




Figure 320. BSC controlled outgoing TCH-TCH HO success %, S1 (hsr_6)

Intra-cell SDCCH-SDCCH HO success %, S1 (hsr_7)

sum(cell_sdcch)100 * ------------------- %

sum(cell_sdcch_at)


Figure 321. Intra-cell SDCCH-SDCCH HO success %, S1 (hsr_7)

Intra-cell SDCCH-TCH HO success %, S1 (hsr_8)

sum(cell_sdcch_tch)100 * --------------------- %

sum(cell_sdcch_tch_at)


Figure 322. Intra-cell SDCCH-TCH HO success %, S1 (hsr_8)

Intra-cell TCH-TCH HO success %, S1 (hsr_9)

sum(cell_tch_tch)100 * --------------------- %

sum(cell_tch_tch_at)


Figure 323. Intra-cell TCH-TCH HO success %, S1 (hsr_9)

MSC controlled incoming SDCCH-SDCCH HO success %, S1 (hsr_10)

sum(msc_i_sdcch)100 * --------------------- %

sum(msc_i_sdcch_at)




Figure 324. MSC controlled incoming SDCCH-SDCCH HO success %, S1(hsr_10)

MSC controlled incoming SDCCH-TCH HO success %, S1 (hsr_11)

sum(msc_i_sdcch_tch)100 * --------------------- %

sum(msc_i_sdcch_tch_at)


Figure 325. MSC controlled incoming SDCCH-TCH HO success %, S1 (hsr_11)

MSC controlled incoming TCH-TCH HO success %, S1 (hsr_12)

sum(msc_i_tch_tch)100 * --------------------- %

sum(msc_i_tch_tch_at)


Figure 326. MSC controlled incoming TCH-TCH HO success %, S1 (hsr_12)

BSC controlled incoming SDCCH-SDCCH HO success %, S1 (hsr_13)

sum(bsc_i_sdcch)100 * --------------------- %

sum(bsc_i_sdcch_at)


Figure 327. BSC controlled incoming SDCCH-SDCCH HO success %, S1(hsr_13)

BSC controlled incoming SDCCH-TCH HO success %, S1 (hsr_14)

sum(bsc_i_sdcch_tch)100 * ----------------------- %

sum(bsc_i_sdcch_tch_at)




Figure 328. BSC controlled incoming SDCCH-TCH HO success %, S1 (hsr_14)

BSC controlled incoming TCH-TCH HO success %, S1 (hsr_15)

sum(bsc_i_tch_tch)100 * --------------------- %

sum(bsc_i_tch_tch_at)


Figure 329. BSC controlled incoming TCH-TCH HO success %, S1 (hsr_15)

BSC controlled incoming HO success %, S1 (hsr_16)

sum(bsc_i_succ_ho)100 * ------------------------------------------------------- %

sum(bsc_i_sdcch_at+bsc_i_sdcch_tch_at+bsc_i_tch_tch_at)


Figure 330. BSC controlled incoming HO success %, S1 (hsr_16)

MSC controlled incoming HO success %, S1 (hsr_17)

sum(msc_i_succ_ho)100 * ------------------------------------------------------- %



Figure 331. MSC controlled incoming HO success %, S1 (hsr_17)

Incoming HO success %, S1 (hsr_18)

sum(msc_i_tch_tch+bsc_i_tch_tch)100 * --------------------------------------- %

sum(msc_i_tch_tch_at +bsc_i_tch_tch_at)


Figure 332. Incoming HO success %, S1 (hsr_18)



Outgoing HO success %, S1 (hsr_19)

sum(msc_o_tch_tch+bsc_o_tch_tch)100 * --------------------------------------- %

sum(msc_o_tch_tch_at +bsc_o_tch_tch_at)


Figure 333. Outgoing HO success %, S1 (hsr_19)

Intra-cell SDCCH-SDCCH HO success %, S1 (hsr_20)

sum(cell_sdcch)100 * ------------------ %

sum(cell_sdcch_at)


Figure 334. Intra-cell SDCCH-SDCCH HO success %, S1 (hsr_20)

Intra-cell SDCCH-TCH HO success %, S1 (hsr_21)

sum(cell_sdcch_tch)100 * --------------------- %

sum(cell_sdcch_tch_at)


Figure 335. Intra-cell SDCCH-TCH HO success %, S1 (hsr_21)

Intra-cell TCH-TCH HO success %, S1 (hsr_22)

sum(cell_tch_tch)100 * --------------------- %

sum(cell_tch_tch_at)


Figure 336. Intra-cell TCH-TCH HO success %, S1 (hsr_22)



2.21 Handover failures (hof)

Outgoing HO failures due to lack of resources (hof_1)

sum(BSC_o_fail_lack+MSC_o_fail_lack)


Figure 337. Outgoing HO failures due to lack of resources (hof_1)

Incoming HO failures due to lack of resources (hof_2)

sum(BSC_i_fail_lack+MSC_i_fail_lack)


Figure 338. Incoming HO failures due to lack of resources (hof_2)

TCH HO failures when return to old channel was successful (hof_3)

Known problems: Due to the mapping of different causes the accuracy may bepoor.

HOs failed in going to new channel - HOs failed to return to old channel

= sum(tch_rf_new_ho + tch_abis_fail_new + tch_a_if_fail_new + tch_tr_fail_new)- sum(tch_rf_old_ho + tch_abis_fail_old + tch_a_if_fail_old + tch_tr_fail_old)


Figure 339. TCH HO failures when return to old channel was successful (hof_3)

SDCCH HO failures when return to old channel was successful (hof_4)

Known problems: Due to the mapping of different causes the accuracy may bepoor.

HOs failed in going to new channel - HOs failed to return to old channel

= sum(sdcch_rf_new_ho+sdcch_abis_fail_new+sdccha_if_fail_new+sdcch_tr_fail_new)-sum(sdcch_rf_old_ho+sdcch_abis_fail_old+sdccha_if_fail_old+sdcch_tr_fail_old)


Figure 340. SDCCH HO failures when return to old channel was successful(hof_4)



MSC incoming HO failures (hof_5)

HO attempts - successful HO

= sum(msc_i_tch_tch_at+msc_i_tch_tch_at+msc_i_sdcch_at- msc_i_tch_tch+msc_i_sdcch_tch+msc_i_sdcch)


Figure 341. MSC incoming HO failures (hof_5)

MSC outgoing HO failures (hof_6)



Figure 342. MSC outgoing HO failures (hof_6)

MSC outgoing HO failures (hof_6a)

sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at + msc_o_sdcch_at)- sum(msc_o_succ_ho)


Figure 343. MSC outgoing HO failures (hof_6a)

BSC incoming HO failures (hof_7)

sum(bsc_i_tch_tch_at + bsc_i_sdcch_tch_at + bsc_i_sdcch_at)- sum(bsc_i_tch_tch + bsc_i_sdcch_tch + bsc_i_sdcch)


Figure 344. BSC incoming HO failures (hof_7)

BSC incoming HO failures (hof_7a)

sum(bsc_i_tch_tch_at + bsc_i_sdcch_tch_at + bsc_i_sdcch_at-bsc_i_succ_ho)




Figure 345. BSC incoming HO failures (hof_7a)

BSC outgoing HO failures (hof_8)

sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at + bsc_o_sdcch_at)- sum(bsc_o_tch_tch + bsc_o_sdcch_tch + bsc_o_sdcch)


Figure 346. BSC outgoing HO failures (hof_8)

BSC outgoing HO failures (hof_8a)

sum(bsc_o_tch_tch_at + bsc_o_sdcch_tch_at + bsc_o_sdcch_at- bsc_o_succ_ho)


Figure 347. BSC outgoing HO failures (hof_8a)

Intra-cell HO failures (hof_9)

sum(cell_tch_tch_at + cell_sdcch_at - cell_tch_tch + cell_sdcch)


Figure 348. Intra-cell HO failures (hof_9)

Intra-cell HO failures (hof_9a)

sum(cell_tch_tch_at + cell_sdcch_at+ cell_sdcch_tch-cell_tch_tch - cell_sdcch-cell_sdcch_tch)


Figure 349. Intra-cell HO failures (hof_9a)



Failed outgoing HO, return to old (hof_10)

sum(msc_o_fail_ret + bsc_o_fail_ret)


Figure 350. Failed outgoing HO, return to old (hof_10)

Outgoing HO failures (hof_12)

Outgoing HO attempts - Outgoing HO successes

=sum(msc_o_tch_tch_at + msc_o_sdcch_tch_at + msc_o_sdcch_at

+ bsc_o_tch_tch_at + bsc_o_sdcch_tch_at + bsc_o_sdcch_at)- sum(msc_o_succ_ho + bsc_o_succ_ho)


Figure 351. Outgoing HO failures (hof_12)

Intra-cell HO failure, return to old channel (hof_13)

sum(cell_fail_ret)


Figure 352. Intra-cell HO failure, return to old channel (hof_13)

Intra-cell HO failure, drop call (hof_14)

sum(cell_drop_calls)


Figure 353. Intra-cell HO failure, drop call (hof_14)

Incoming HO failures (hof_15)

Incoming HO attempts - Incoming HO successes

=sum(msc_i_tch_tch_at + msc_i_sdcch_tch_at + msc_i_sdcch_at

+ bsc_i_tch_tch_at + bsc_i_sdcch_tch_at + bsc_i_sdcch_at)- sum(msc_i_succ_ho + bsc_i_succ_ho)




Figure 354. Incoming HO failures (hof_15)

2.22 Interference (itf)

UL interference, BTS level, S1 (itf_1)

Use: UL interference is measured as the time-out of the lowestband (band 0 in BSC terminology). Band 0 is defined byboundaries 0 and 1 which are BTS parameters. Boundary 0 isfixed, whereas boundary 1 can be set.

Experiences on use: UL interference alone is not a reliable quality factor if IUOis used. In IUO cells the UL interference can be high but thequality is still good.

Known problems: This formula is on the BTS level, whereas the interferenceproblems are met on the frequency (TRX) level. This meansthat the accuracy is not good if there is more than one TRX ina cell.If band 1 is defined as exceptionally wide, it becomes difficultto see the interference.

sum(ave_idle_f_TCH_1/res_av_denom4)100 x (1- ------------------------------------- ) %

sum(ave_idle_f_TCH_1/res_av_denom4+ ave_idle_f_TCH_2/res_av_denom5+ ave_idle_f_TCH_3/res_av_denom6+ ave_idle_f_TCH_4/res_av_denom7+ ave_idle_f_TCH_5/res_av_denom8)


Figure 355. UL interference, BTS level, S1 (itf_1)

Idle TSL percentage of time in band X, TRX level, IUO, S4 (itf_2)

Experiences on use: In IUO cells the UL interference can show high values butthe UL quality is still excellent. The bigger the values are inbands towards band 5, the worse the interference.

sum(ave_full_tch_ifX)100 x (----------------------- ) %

sum(ave_full_tch_if1+ ave_full_tch_if2+ ave_full_tch_if3+ ave_full_tch_if4+ ave_full_tch_if5)




Figure 356. Idle TSL percentage of time in band X, TRX level, IUO, S4 (itf_2)

UL interference from IUO, TRX level, S4 (itf_3)

Experiences on use: UL interference alone is not a reliable quality factor if IUOis used. In IUO cells the UL interference can be high but thequality is still good.

Known problems: There are more than one TRX in a cell.

sum(ave_full_tch_if1)100 x (1 - ----------------------- ) %

sum(ave_full_tch_if1+ ave_full_tch_if2+ ave_full_tch_if3+ ave_full_tch_if4+ ave_full_tch_if5)


Figure 357. UL interference from IUO, TRX level, S4 (itf_3)

UL interference from Power Control, TRX level, S6 (itf_4)

Use: BTS reports the interference of each TCH as a band number(0-4, where 0 is the lowest and band boundaries are defined ascell parameters). BSC sums up the band numbers(ave_sum_idle_ch_interf) as well as the number ofTCHs reported (ave_sum_idle_tch_per_trx), and fromthese figures an average interference band (0-4) can becalculated. Average interference is shifted by 1 (+1) tocomply with band numbers 1-5.

Experiences on use: Shows null value on the TRX level if the TRX is completelyin the GPRS territory, and on BTS level if all TRXs arecompletely in the GPRS territory, respectively.

sum(ave_sum_idle_ch_interf)--------------------------------sum(ave_sum_idle_tch_per_trx)+1

Counters from table(s):p_nbsc_power

Figure 358. UL interference from Power Control, TRX level, S6 (itf_4)



2.23 Congestion (cngt)

TCH congestion time, S1 (cngt_1)

Experiences on use: Useful to follow on the area level. Should give higher valuesthan SDCCH congestion.

sum(tch_cong_time/100)

Counters from table(s):p_nbsc_res_availunit: second

Figure 359. TCH congestion time, S1 (cngt_1)

SDCCH congestion time, S1 (cngt_2)

Experiences on use: Useful to follow on the area level. Should give smallervalues than TCH congestion.

sum(sdcch_cong_time/100)

Counters from table(s):p_nbsc_res_availunit: second

Figure 360. SDCCH congestion time, S1 (cngt_2)

FTCH time congestion % (cngt_3)

sum(tch_fr_radio_congestion_time)100 * ----------------------------------------- %

sum(period_duration*ave_tch_busy_full/60)


Figure 361. FTCH time congestion % (cngt_3)

FTCH time congestion % (cngt_3a)

sum(tch_fr_radio_congestion_time/100)100 * ----------------------------------------- %

sum(period_duration*ave_tch_busy_full*60)


Figure 362. FTCH time congestion % (cngt_3a)



HTCH time congestion % (cngt_4)

sum(tch_hr_radio_congestion_time)100 * ----------------------------------------- %

sum(period_duration*ave_tch_busy_half/60)


Figure 363. HTCH time congestion % (cngt_4)

HTCH time congestion % (cngt_4a)

sum(tch_hr_radio_congestion_time/100)100 * ----------------------------------------- %

sum(period_duration*ave_tch_busy_half*60)


Figure 364. HTCH time congestion % (cngt_4a)

2.24 Queuing (que)

Queued, served TCH call requests % (que_1a)

Use: Indicates the quota of TCH call requests seizing the TCHsuccessfully after queuing.

Known problems: tch_qd_call_att is triggered but unsrv_qd_call_attis not if the call is lost for some other reason (e.g. MS userhangs) before the queuing timer expires.The impact of this depends on queuing time and cellparameter Directed Retry Time Limit Min.

sum( tch_qd_call_att - removal_from_que_due_to_dr - unsrv_qd_call_att)100* ---------------------------------------------------------------------- %

sum(tch_call_req)


Figure 365. Queued, served TCH call requests % (que_1a)

Queued, served TCH HO requests % (que_2)

Known problems: tch_qd_ho_att is triggered but unsrv_qd_ho_att is notif a call for some reason is lost (the MS user is hanging, forexample) before the queuing timer expires.



sum(tch_qd_ho_att-unsrv_qd_ho_att)100* ------------------------------------------------- %

sum(tch_request-tch_call_req-tch_fast_req)


Figure 366. Queued, served TCH HO requests % (que_2)

Queued, served TCH HO requests % (que_2a)

Known problems: tch_qd_ho_att is triggered but unsrv_qd_ho_att is notif a call for some reason is lost (the MS user is hanging, forexample) before the queuing timer expires.

sum(a. tch_qd_ho_att-a.unsrv_qd_ho_att)100* ------------------------------------------------- %

sum(a.tch_request-a.tch_call_req-a.tch_fast_req)- Sum(b.bsc_i_unsucc_a_int_circ_type+b.msc_controlled_in_ho



Figure 367. Queued, served TCH HO requests % (que_2a)

Successful queued TCH requests (que_3)

sum(tch_qd_call_att-unsrv_qd_call_att)


Figure 368. Successful queued TCH requests (que_3)

Successful non-queued TCH requests (que_4)

sum(tch_norm_seiz)-sum(tch_qd_call_att-unsrv_qd_call_att)


Figure 369. Successful non-queued TCH requests (que_4)

Successful queued TCH HO requests (que_5)

sum(tch_qd_ho_att-unsrv_qd_ho_att)




Figure 370. Successful queued TCH HO requests (que_5)

Successful non-queued TCH HO requests (que_6)

sum(tch_ho_seiz) -sum(tch_qd_ho_att-unsrv_qd_ho_att)


Figure 371. Successful non-queued TCH HO requests (que_6)

Non-queued, served TCH call requests % (que_7)

Use: Indicates the quota of TCH call requests seizing the TCHsuccessfully straight without queuing. DR is excluded (itsimpact is seen in dr_3).

sum(tch_norm_seiz - (tch_qd_call_att - unsrv_qd_call_att))100 * ---------------------------------------------------------- %

sum(tch_call_req)


Figure 372. Non-queued, served TCH call requests % (que_7)

Non-queued, served TCH HO requests % (que_8)

sum(tch_ho_seiz-(tch_qd_ho_att-unsrv_qd_ho_att))100 * ------------------------------------------------- %



Figure 373. Non-queued, served TCH HO requests % (que_8)

Non-queued, served TCH HO requests % (que_8a)

Known problems: See que_2.

sum(a.tch_ho_seiz-(a.tch_qd_ho_att-a.unsrv_qd_ho_att))100 * -------------------------------------------------------- %

sum(a.tch_request-a.tch_call_req-a.tch_fast_req)- Sum(b.bsc_i_unsucc_a_int_circ_type+b.msc_controlled_in_ho





Figure 374. Non-queued, served TCH HO requests % (que_8a)

2.25 Blocking (blck)

TCH raw blocking, S1 (blck_1)

Experiences on use: Was earlier (before blck_8a) widely used on the cell andthe area level.

Known problems: This PI does not take Directed Retry into consideration.Rather, it shows only raw blocking including also HOs.Blocked HOs are normally not so serious because there arealternatives to go to. Blocked new calls can be lost calls ifDirected Retry is not in use.

sum(tch_req_rej_lack)100 * --------------------- %

sum(tch_request)


Figure 375. TCH raw blocking, S1 (blck_1)

SDCCH blocking %, S1 (blck_5)

Known problems: See csf_1.

100-csf_1 =

sum(SDCCH_busy_att)100 * -------------------- %

sum(SDCCH_seiz_att)


Figure 376. SDCCH blocking %, S1 (blck_5)

SDCCH real blocking %, S1 (blck_5a)

Known problems: See csf_1a.

100_csf_1a =

sum(SDCCH_busy_att - tch_seiz_due_sdcch_con)100 * -------------------------------------------- %



sum(SDCCH_seiz_att)


Figure 377. SDCCH real blocking %, S1 (blck_5a)

TCH raw blocking % on super TRXs, S4 (blck_6)

Use: TCH blocking % on super TRXs.Note: Cannot be calculated by a simple SQL*Plus statement.

sum over super TRXs (tch_req_rej_lack)100 * ----------------------------------------- %

sum over super TRXs (tch_request)


Figure 378. TCH raw blocking % on super TRXs, S4 (blck_6)

TCH raw blocking % on regular TRXs, S4 (blck_7)

Use: TCH Blocking % on regular layer.Note: Cannot be calculated by a simple SQL*Plus statement.

sum over regular TRX (tch_req_rej_lack)100 * ----------------------------------------- %

sum over regular TRX (tch_request)


Figure 379. TCH raw blocking % on regular TRXs, S4 (blck_7)

TCH call blocking, before DR, S2 (blck_8)

Experiences on use: Shows the blocking if DR is not used.

TCH call req. rejected due to lack of res. or routed by DR to another cell100* ------------------------------------------------------------------------ % =

all TCH call requests

sum(tch_call_req-tch_norm_seiz)100* -------------------------------- %

sum(tch_call_req)


Figure 380. TCH call blocking, before DR, S2 (blck_8)



TCH call blocking %, DR compensated, S2 (blck_8b)

Use: On the cell level should appear in the busiest cells. The cellneeds an urgent capacity extension or has lost part of capacitydue to a fault.It is the blocking rate that the customer will notice when theyare driving in the mobile environment caused by the lack ofradio resources. It is therefore one of the most critical KPIs.

Experiences on use: On the area level there is not yet a target value to give(except that 0% is the best). On the cell level, for example, 2%blocking on Busy Hour has been used as a criterion for design.

Known problems: 1) This blocking also shows situations when it is caused by afault in the BTS - not only pure blocking caused by hightraffic.2) NOTE: If Trunk Reservation is used, HO and Call blockingcannot be counted precisely (there is only one counter for thecase of Trunk Reservation Invocation Refused).3) The ratio can show too high values in the following case:TCH assignment fails if the requested channel type is notfound in the A-interface circuit pool. In this casetch_norm_seiz is not triggered but tch_call_req is, i.e.this attempt in blck_8b is considered a blocked call.Anyhow, BSC requests the MSC to change the A-if circuitpool. MSC can then decide if there is another assignmentrequest or call clear (clear_command).The second request may again fail or succeed. In BSC theTCH_REJ_DUE_REQ_CH_A_IF_CRC counter is triggeredevery time the channel request fails due to the above-mentioned reason.If Nokia MSC is used, there can only be one retry. Withanother vendor’s MSC there can be multiple retrys.The actual situation when this can be met is if EFR (enhancedfull rate) codec is the primary choice but- the selected A-if circuit supports only Full Rate- there are free circuits supporting EFR in the A interface.

TCH call requests rejected due to lack of res. or saved by DR- successful DR

100* ------------------------------------------------------------------------- % =all TCH call requests

sum(a.tch_call_req-a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch+ b.bsc_o_sdcch_tch+b.cell_sdcch_tch)

100* ---------------------------------------------------------- %sum(a.tch_call_req)


Figure 381. TCH call blocking %, DR compensated, S2 (blck_8b)



TCH call blocking %, DR and DAC compensated, EFR excluded, S5(blck_8d)

Use: Applicable on area or BTS level.Queuing and Directed Retry are the BSS features that canreduce blocking.It is the failed call attempts that the MS user will notice,caused by the lack of radio resources. It is therefore one of themost critical KPIs.On the cell level may appear in the busiest cells. The cellneeds an urgent capacity extension or has lost part of capacitydue to a fault. An MS user will usually hear three beep toneswhen the call is rejected due to blocking.

Experiences on use: On the area level there is not yet a target value to give(except that the trend should be towards a smaller value, 0%being the best). On the cell level, for example, 2% blockingon Busy Hour has been used as a criterion for design. ThisKPI can be followed statistically, for example, as the numberof cells in which the value exceeds the given threshold.

Known problems: 1) Note that if Trunk Reservation is used, HO and Callblocking cannot be counted precisely, i.e. there is only onecounter for the case of Trunk Reservation Invocation Refused.2) If dadlb_start_due_exceeded_load is triggered andthe DADLB handover fails, the counter tch_call_req willbe triggered twice. This problem will be corrected in S10.3) The formula counts also the following case of a blockedcall: a call is cleared from the other end between call requestand TCH seizure. Note that queuing prolongs this time andthus the probability of a call to be cleared.

100 - csf_3l =

sum(a.tch_call_req - a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch + b.bsc_o_sdcch_tch + b.cell_sdcch_tch); DR calls+ sum(a.tch_succ_seiz_for_dir_acc) ;ref.2- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool handover failures+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type))

100 * -------------------------------------------------------------------------- %sum(a.tch_call_req)

- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool handover failures+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type))


Ref.2. Compensation needed since in the case of Direct Access to super reuse TRXthe tch_norm_seiz is triggered in parallel with the cell_sdcch_tch.

Figure 382. TCH call blocking %, DR and DAC compensated, EFR excluded, S5(blck_8d)



Ref.2. Compensation needed since in case of Direct Access to super reuse TRXthe tch_norm_seiz is triggered in parallel with cell_sdcch_tch.

Blocked calls, S5 (blck_9b)

Use: Shows also situations when blocking is caused by a fault inthe BTS - not only blocking caused purely by high traffic.

TCH call req. rejected due to lack of res. or routed by DR to another cell- successful DR- Rejections due to Aif circuit mismatch

sum(a.tch_call_req-a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch+ b.bsc_o_sdcch_tch);inter-cell DR- sum(b.cell_sdcch_tch); intra-cell DR in IUO- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool handover failures+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type))


Figure 383. Blocked calls, S5 (blck_9b)

Blocked calls, S5 (blck_9c)

Use: Shows also situations when blocking is caused by a fault inthe BTS - not only blocking caused purely by high traffic.

TCH call req. rejected due to lack of res. or routed by DR to another cell- successful DR- Rejections due to Aif circuit mismatch

sum(a.tch_call_req-a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch+ b.bsc_o_sdcch_tch);inter-cell DR- sum(b.cell_sdcch_tch); intra-cell DR in IUO+ sum(a.succ_tch_seiz_for_dir_acc) ;ref.2- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool handover failures+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type))


Figure 384. Blocked calls, S5 (blck_9c)


Blocked TCH HOs, S2 (blck_10a)

Use: Replaces blck_10.



sum(tch_request-tch_call_req - tch_fast_req-tch_ho_seiz)


Figure 385. Blocked TCH HOs, S2 (blck_10a)

Blocked TCH HOs, S5 (blck_10b)

Use: Replaces blck_10a.

sum(a.tch_request - a.tch_call_req - a.tch_fast_req - a.tch_ho_seiz)-sum(b.bsc_i_unsucc_a_int_circ_type + b.msc_controlled_in_ho

+ b.ho_unsucc_a_int_circ_type)


Figure 386. Blocked TCH HOs, S5 (blck_10b)

TCH HO blocking, S2 (blck_11a)

Known problems: 1) Shows also the situations when blocking is caused by afault in the BTS - not only blocking caused purely by hightraffic.2) Note that if Trunk Reservation is used, HO and Callblocking cannot be counted precisely (there is only onecounter for the case of Trunk Reservation InvocationRefused).

sum(tch_request-tch_call_req-tch_fast_req-tch_ho_seiz)100 * ------------------------------------------------------- %



Figure 387. TCH HO blocking, S2 (blck_11a)

TCH HO blocking without Q, S2 (blck_11b)

Use: Shows TCH HO blocking if queuing was not in use.Known problems: See que_2 (factor XX1).

sum(tch_request - tch_call_req - tch_fast_req - tch_ho_seiz)+ sum(tch_qd_ho_att - XX1-unserv_qd_ho_att)

100 * ------------------------------------------------------------- %sum(tch_request - tch_call_req - tch_fast_req)




Figure 388. TCH HO blocking without Q, S2 (blck_11b)

TCH HO blocking, S5 (blck_11c)

Known problems: 1) Shows also the situations when blocking is caused by afault in the BTS - not only blocking caused purely by hightraffic.2) Note that if Trunk Reservation is used, HO and Callblocking cannot be counted precisely (there is only onecounter for the case of Trunk Reservation InvocationRefused).

sum(a.tch_request-a.tch_call_req-a.tch_fast_req-a.tch_ho_seiz)-sum(b.bsc_i_unsucc_a_int_circ_type+b.msc_controlled_in_ho

+b.ho_unsucc_a_int_circ_type)100 * -------------------------------------------------------------- %

sum(a.tch_request-a.tch_call_req-a.tch_fast_req)-sum(b.bsc_i_unsucc_a_int_circ_type+b.msc_controlled_in_ho



Figure 389. TCH HO blocking, S5 (blck_11c)

Blocked incoming and internal HO, S2 (blck_12)

Use: Usable with S4 and earlier.

sum(msc_i_fail_lack + bsc_i_fail_lack + cell_fail_lack)


Figure 390. Blocked incoming and internal HO, S2 (blck_12)

Blocked incoming and internal HO, S2 (blck_12a)

Use: Used on the area level with S5 and S6.

sum(msc_i_fail_lack + bsc_i_fail_lack + cell_fail_lack+ bsc_i_unsucc_a_int_circ_type+ msc_controlled_in_ho+ ho_unsucc_a_int_circ_type)




Figure 391. Blocked incoming and internal HO, S2 (blck_12a)

AG blocking, S1 (blck_13)

Use: A BSC sends to a BTS an immediate assignment orimmediate assignment rejected commands. If the AccessGrant (AG) buffer in the BTS is full, it will respond with adelete indication. Thus, the ratio of delete indications to thesum of immediate assignment and immediate assignmentrejected describes the AG blocking. After receiving the deleteindication message the BSC releases the SDCCH.

100 * sum(del_ind_msg_rec)/ sum(imm_assgn_rej+imm_assgn_sent)


Figure 392. AG blocking, S1 (blck_13)

FCS blocking, S5 (blck_14)

sum(tch_seiz_att_due_sdcch_con - tch_seiz_due_sdcch_con)100 * --------------------------------------------------- %

sum(tch_seiz_att_due_sdcch_con %)


Figure 393. FCS blocking, S5 (blck_14)

Blocked SDCCH seizure attempts, S5 (blck_15)

All blocked - seizures to FACCH setup sum(sdcch_busy_att- tch_seiz_due_sdcch_con)


Figure 394. Blocked SDCCH seizure attempts, S5 (blck_15)

HO blocking % (blck_16a)

Use: Used on the area level with S4 or earlier.




100 * --------------------------------------------------------/* all HO attempts */sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at



Figure 395. HO blocking % (blck_16a)

Handover blocking % (blck_16b)

Use: Used on the area level with S5 and S6.Known problems: If the required channel type (e.g. Full Rate) is not available in

intra-cell handover, then ho_unsucc_a_int_circ_type istriggered, but the same channel may be seized successfullyafter changing the handover for external (A interface circuitchanges).

/* handovers failing due to blocking */sum(msc_i_fail_lack+bsc_i_fail_lack + cell_fail_lack+

+bsc_i_unsucc_a_int_circ_type+msc_i_unsucc_a_int_circ_type+ho_unsucc_a_int_circ_type)

100 * -------------------------------------------------------------- %/* all HO attempts */sum(msc_o_tch_tch_at+msc_o_sdcch_tch_at+msc_o_sdcch_at



Figure 396. Handover blocking % (blck_16b)

Abis link blocking (blck_17)

There is no counter but an alarm: 2720 ’Telecom Link Overload’.

Blocked FACCH call setup TCH requests (blck_18)

sum(tch_seiz_att_due_sdcch_con - tch_seiz_due_sdcch_con)


Figure 397. Blocked FACCH call setup TCH requests (blck_18)

Handover blocking to target cell (blck_19)

sum(ho_fail_res_to_adj)100 * ---------------------- %



sum(ho_att_to_adj)


Figure 398. Handover blocking to target cell (blck_19)

Handover blocking from target cell (blck_20)

sum(ho_fail_res_from_adj)100 * ------------------------- %

sum(ho_att_from_adj)


Figure 399. Handover blocking from target cell (blck_20)

NACK ratio of p-immediate assignment, S9PS (blck_21)

Use: A negative acknowledgement (NACK) is sent from BTS toBSC after all AGCH messages that are deleted from TRXbuffers due to:- buffer overflow- maximum lead-time expiry- expired starting timeThe AGCH messages are ordered by BSC to beacknowledged. The negative acknowledgement is sentimmediately after the message has been deleted.

sum(packet_immed_ass_nack_msg)100 * ---------------------------------------------------- %

sum(packet_immed_ass_msg + packet_immed_ass_rej_msg)


Figure 400. NACK ratio of p-immediate assignment, S9PS (blck_21)

Territory upgrade rejection %, S9PS (blck_22)

Use: Indicates the lack of resources to upgrade the GPRS territory.

sum(gprs_ter_ug_rej_due_csw_tr+gprs_ter_ug_rej_due_lack_psw+gprs_ter_ug_rej_due_lack_pcu)

100 * --------------------------------- %sum(gprs_ter_upgrd_req)




Figure 401. Territory upgrade rejection %, S9PS (blck_22)

Handover blocking to target WCDMA cell, S10.5 (blck_27)

sum(ho_fail_due_res_wcdma_ran)100 * ------------------------------

sum(ho_att_wcdma_ran_cell)


Unit: %

Figure 402. Handover blocking to target WCDMA cell, S10.5 (blck_27)

Handover blocking from target WCDMA cell, S10.5 (blck_28)

sum(ho_fail_due_res_wcdma_ran_cell)100 * -----------------------------------

sum(ho_att_from_wcdma_ran)


Unit: %

Figure 403. Handover blocking from target WCDMA cell, S10.5 (blck_28)

TCH denied for Call request, Ratio, S10 (blck_29)

Use: To indicate end user experience of TCH denied for new calls.Failures during call setup are not considered.

Description: Attempts that fail to get TCH after the call setup phase.Sum (a.tch_call_req + a.tch_fast_req)

- Sum (c.tch_re_est_assign + c.tch_new_call_assign)- Sum (a.tch_rej_due_req_ch_a_if_crc; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type; Aif circuit pool handover failures+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type))

100* ----------------------------------------------------------------------- %Sum (a.tch_call_req + a.tch_fast_req)

- Sum (a.tch_rej_due_req_ch_a_if_crc; Aif type mismatch or congestion-(b.bsc_i_unsucc_a_int_circ_type; Aif circuit pool handover failures+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type))




Note

c = p_nbsc_service

Unit: %

Figure 404. TCH denied for Call request, Ratio, S10 (blck_29)

2.26 Traffic (trf)

TCH traffic sum, S1 (trf_1)

Experiences on use: If counted over one hour, erlang is shown. Counting erlangsover a longer period requires that the erlang values per hourare first counted and then averaged.

Known problems: Shows slightly different values (around 3 % higher accordingto one study) than if counted from an MSC. The reason forthis is that in a BSC one call holds two TCHs for a short periodin HOs.The sampling period is 20 s, which means that during theperiod of one hour the number of used TCHs are checked 180times. This method is not accurate if we think about shortseizures and low traffic, but statistically the results have beensatisfactory.Does not show calls going to voice mail.

The result represents technical traffic, not charged traffic because counting isstarted when BSC seizes TCH. Includes some of signalling, ringing and speech.

sum(ave_busy_tch)-------------------sum(res_av_denom14)

Counters from table(s):p_nbsc_res_availUnit: Erlang hours if the measurement period is 1 hour.

Figure 405. TCH traffic sum, S1 (trf_1)

TCH traffic sum, S1 (trf_1a)

Note: See trf_1.

sum_over_area(sum_over_BTS(ave_busy_tch)/ sum_over_BTS(res_av_denom14)

)




Figure 406. TCH traffic sum, S1 (trf_1a)

TCH traffic sum of normal TRXs, S1 (trf_1b)

Use: On BTS and area level.Known problems: See trf_1.Note: See trf_1.

sum(decode(trx_type,0,ave_busy_tch) / decode(trx_type,0,res_av_denom14))

Counters from table(s):p_nbsc_res_availUnit: Erlang hours if measurement period is 1 hour.

Figure 407. TCH traffic sum of normal TRXs, S1 (trf_1b)

TCH traffic sum of extended TRXs, S1 (trf_1c)

Use: On BTS and area level.Known problems: See trf_1.Note: See trf_1.

sum(decode(trx_type,1,ave_busy_tch) / decode(trx_type,1,res_av_denom14))

Counters from table(s):p_nbsc_res_availUnit: Erlang hours if measurement period is 1 hour.

Figure 408. TCH traffic sum of extended TRXs, S1 (trf_1c)

Average call length, S1 (trf_2b)

Use: When used on the area level this PI gives an idea about thebehaviour of the MS users.

Note: In the numerator (a.ave_busy_tch / a.res_av_denom14)represents technical traffic, not charged traffic becausecounting is started when BSC seizes TCH. Includes somesignalling, ringing and speech. The denominator includes alsounanswered calls.

total TCH use time nbr of seconds in meas.period * average busy TCH------------------- = -------------------------------------------------

number of calls number of calls

sum(period_duration*60* a.ave_busy_tch / a.res_av_denom14)= -------------------------------------------------------------------------

sum(b.tch_norm_seiz) ;normal calls+ sum(c.msc_o_sdcch_tch+ c.bsc_o_sdcch_tch + c.cell_sdcch_tch) ;DR calls



+ sum(b.tch_seiz_due_sdcch_con) ; FACCH call setup calls

Counters from table(s):a = p_nbsc_res_availb = p_bsc_trafficc = p_nbsc_ho

Figure 409. Average call length, S1 (trf_2b)

Average call length, S1 (trf_2d)

Use: On the area level gives you an idea about the behaviour of theMS users.

Note: In the numerator ( a.ave_busy_tch /a.res_av_denom14) represents technical traffic, notcharged traffic, because counting is started when BSC seizesTCH. Includes some of signalling, ringing and speech.In the denominator there are also calls that are not answered.The numerator counts both the A and B side in MS-MS calls,thus duplicating call time.

total TCH use time nbr of seconds in meas.period * average busy TCH------------------- = -------------------------------------------------number of calls number of calls

sum(period_duration*60* a.ave_busy_tch / a.res_av_denom14)= ----------------------------------------------------------

sum(b.tch_norm_seiz) ;normal calls+ sum(msc_i_sdcch_tch+ bsc_i_sdcch_tch + cell_sdcch_tch); DR calls+ sum(a.tch_seiz_due_sdcch_con) ; FACCH call setup calls

Counters from table(s):a = p_nbsc_res_availb = p_bsc_trafficc = p_nbsc_ho

Figure 410. Average call length, S1 (trf_2d)

CS territory usage, S1 (trf_3)

Use: On the area level gives you an idea of how well the capacityis used. Usable after S4 if half rate is not used. The formuladoes not comprise to the GPRS timeslots (PDTCH).

Known problems: When GPRS is used, the CS territory size (denominator) canchange according to the traffic needs and therefore theindicator is not consistent.

used TCH100 * ----------------------- %

available TCH

sum(ave_busy_tch/res_av_denom14)= 100 * -------------------------------------------- %

sum(ave_avail_full_TCH/res_av_denom2)




Figure 411. CS territory usage, S1 (trf_3)

FTCH usage, S5 (trf_3b)

Use: On the area level gives you an idea of how well the capacityis used. Use with S5 or later.

Known problems: When GPRS is used, the CS territory size (denominator) canchange according to the traffic needs and therefore theindicator is not consistent.

sum(ave_tch_busy_full)= 100 * -------------------------------------------- %

sum(ave_avail_full_TCH/res_av_denom2)


Figure 412. FTCH usage, S5 (trf_3b)

Average SDCCH holding time, S1 (trf_4)

Note: For area (or segment) level, first use the formula for all theBTSs and then sum this formula over all underlying BTSs,keeping in mind the note above.

Use: The holding time may change due to modification of thetimers or perhaps software. This time is part of the call setuptime.

Experiences on use: The counters receive the value of zero if the BTS is locked.Typically the values range from 2 to 3 seconds but over 4seconds with satellite Abis.

sum(ave_sdcch_hold_tim)------------------------ secsum(res_av_denom16)*100


Figure 413. Average SDCCH holding time, S1 (trf_4)

Average FTCH holding time, S1 (trf_5)

Use: The holding time may change due to modification of thetimers or perhaps software. You can use this PI to follow theimpact of the modifications.



Experiences on use: The counters receive the value of zero if the BTS is locked.The value is highly dependent on the number of handoversthat, again, are dependent on the network plan.

sum(ave_ftch_hold_tim)------------------------ secsum(res_av_denom17)*100


Figure 414. Average FTCH holding time, S1 (trf_5)

TCH seizures for new call (call bids), S1 (trf_6)

Use: The seizures of TCH for a new call (i.e. not HO, not DR, notFCS).

sum(p_nbsc_traffic.tch_norm_seiz)


Figure 415. TCH seizures for new call (call bids), S1 (trf_6)

SDCCH usage %, S1 (trf_7b)

Experiences on use: Dynamic SDCCH allocation can add SDCCH capacitydynamically and therefore make the counting of SDCCHusage % obsolete.

total SDCCH hold time in seconds100 * --------------------------------------------------- %

average total nbr of SDCCH * period duration in seconds

sum(SDCCH_SEIZURES)*avg(a.ave_sdcch_hold_tim/a.res_av_denom16/100)= 100 * ----------------------------------------------------- %

sum((a.ave_sdcch_sub/a.res_av_denom3 + a.ave_non_avail_sdcch)* a.period duration*60)

where SDCCH_SEIZURES = (b.sdcch_assign+b.sdcch_ho_seiz+b.tch_seiz_due_sdcch_con)

Counters from table(s):a = p_nbsc_res_availb = p_nbsc_traffic

Figure 416. SDCCH usage %, S1 (trf_7b)

SDCCH usage %, S1 (trf_7c)

Known problems: SDCCH seizures are very short to be counted using 20 ssampling time.



sum(period_duration*ave_busy_sdcch/res_av_denom15)100 * ------------------------------------------------------------------------ %

sum((ave_sdcch_sub/res_av_denom3+ave_non_avail_sdcch)*period_duration)

Figure 417. SDCCH usage %, S1 (trf_7c)

TCH traffic absorption on super, S4 (trf_8)

Note: Cannot be calculated by a simple SQL*Plus statement.

1. First, count the traffic per a TRX.

avg(ave_busy_tch)

2. Then, label the TRXs to super or regular (TRX is a super TRX if HOrelated counters for this TRX show the value of zero). Sum up the trafficfor super TRXs and for all TRXs and calculate their ratio.

traffic (super)100 x --------------- %

traffic (all)


Figure 418. TCH traffic absorption on super, S4 (trf_8)

TCH traffic absorption on super, S4 (trf_8a)

Use: IUONote: Cannot be calculated by a simple SQL*Plus statement.

sum over BTS (avg per each super TRX (ave_busy_tch))100 x ---------------------------------------------------- %

sum over BTS (avg per each TRX (ave_busy_tch))


Figure 419. TCH traffic absorption on super, S4 (trf_8a)

Average cell TCH traffic from IUO, S4 (trf_9)




1. First, count the traffic per a TRX and per hour:

avg(ave_busy_tch)

2. Then, sum up the traffic over the period and divide it by the number ofhours in the period.

sum traffic of all TRXs100 x ------------------------ %

hours


Figure 420. Average cell TCH traffic from IUO, S4 (trf_9)

Cell TCH traffic from IUO, S4 (trf_9a)


sum over BTS (avg per each TRX (ave_busy_tch))


Figure 421. Cell TCH traffic from IUO, S4 (trf_9a)

Super TRX TCH traffic, S4 (trf_10)


1. First, count the traffic per a TRX and per hour:

avg(ave_busy_tch)

2. Then, sum it up over the periodfor super TRXs and divide it by the number of hours in the period:

traffic (super)100 x --------------- 100%

hours


Figure 422. Super TRX TCH traffic, S4 (trf_10)

Sum of super TRX TCH traffic, S4 (trf_10a)




sum over BTS (avg per each super TRX (ave_busy_tch))


Figure 423. Sum of super TRX TCH traffic, S4 (trf_10a)

Average SDCCH traffic, erlang, S2 (trf_11)

Known problems: SDCCH seizures are too short to be counted by using 20 ssampling time.

sum of traffic sum(ave_busy_sdcch / res_av_denom15)--------------- = --------------------------------------nbr of records count(*)


Figure 424. Average SDCCH traffic, erlang, S2 (trf_11)

Average SDCCH traffic, erlang, S2 (trf_11b)

Known problems: SDCCH seizures are too short to be counted by using 20 ssampling if traffic is low (less than 0.5 Erl).

Note: On BTS level. For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs.

sum(ave_busy_sdcch) / sum(res_av_denom15)


Figure 425. Average SDCCH traffic, erlang, S2 (trf_11b)

Average TCH traffic, erlang, S2 (trf_12)

sum of traffic sum(ave_busy_tch / res_av_denom14)--------------- = --------------------------------------nbr of records count(*)


Figure 426. Average TCH traffic, erlang, S2 (trf_12)



Average TCH traffic, erlang, S2 (trf_12a)

Note: Gives the same results as trf_12. On BTS level. For area (orsegment) level, first use the formula for all the BTSs and thensum this formula over all underlying BTSs.

avg(ave_busy_tch / res_av_denom14)


Figure 427. Average TCH traffic, erlang, S2 (trf_12a)

Average CS traffic, erlang, S2 (trf_12b)

Use: This is a speech (circuit switched) traffic indicator. Speechtraffic is a basic indicator needed to see how much TCHcapacity is consumed. When the traffic increases without theincrease of the capacity, the probability of blocking increases.The relationship between traffic, capacity and blocking isdescribed for speech traffic in the formula known as Erlang B.This KPI includes all types of CS traffic (single TCH,HSCSD).

Known problems: If extended cells are used, the value is correct only when usedon BTS/trx_type level.

sum(ave_busy_tch) / sum(res_av_denom14)


Unit: erlang

Figure 428. Average CS traffic, erlang, S2 (trf_12b)

Handover/call % (trf_13b)

Use: Indicates how stationary or mobile the traffic is. The biggerthe number, the more mobile is the traffic. Using this KPI,cells with stationary traffic can be found. This is largelydependent on how much the coverage areas overlap.

Known problems: Includes also intra-cell handovers that are not so directlyrelated to mobility.

sum(a.tch_ho_seiz)100 * ------------------------------------------------------------------ %

sum(a.tch_norm_seiz) ;(normal calls)+ sum(c.msc_i_sdcch_tch+c.bsc_i_sdcch_tch) ;(DR inter-cell calls)+ sum(c.cell_sdcch_tch) ;(DR intra-cell calls in IOU, optional feature for S6)+ sum(a.tch_seiz_due_sdcch_con) ; calls started as FACCH call setup





Figure 429. Handover/call % (trf_13b)

Intra-cell handover/call % (trf_13c)

Use: Usually illustrates the impact of interference in a non-IUOnetwork.

Known problems: See trf_13b.

sum(c.cell_tch_tch)100 * ------------------------------------------------------------------ %



Figure 430. Intra-cell handover/call % (trf_13c)

HO / call % (trf_13d)

Use: Indicates how stationary or mobile the traffic is. The biggerthe number, the more mobile is the traffic. By using this KPIcells with stationary traffic can be found. This is largelydependent on how much the coverage areas overlap. It isusable for non-IUO network.The use of this KPI depends on factors like cell size and calllength.

Known problems: Includes also intra-cell HOs that are not so directly related tomobility.

sum(a.tch_ho_seiz)- sum(c.msc_i_sdcch_tch+c.bsc_i_sdcch_tch) ;(DR inter-cell calls)- sum(c.cell_sdcch_tch) ;(DR intra-cell calls in IOU, optional feature for S6)

100 * ------------------------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch +c.bsc_i_sdcch_tch) ;(DR inter-cell calls)+ sum(c.cell_sdcch_tch) ;(DR intra-cell calls in IOU, optional feature for S6)+ sum(a.tch_seiz_due_sdcch_con) ; calls started as FACCH call setup


Figure 431. HO / call % (trf_13d)



Handover/ call % (trf_13e)

Use: Indicates how stationary or mobile the traffic is: the bigger thenumber, the more mobile is the traffic. Using this KPI, cellswith stationary traffic can be found.This is largely dependent on how much the coverage areasoverlap.Usable for a non-IUO network.Intra-cell HOs are not included as they are not directly relatedto mobility.

sum(a.tch_ho_seiz)- sum(c.msc_i_sdcch_tch+c.bsc_i_sdcch_tch) ;(DR inter-cell calls)- sum(c.cell_sdcch_tch) ;(DR intra-cell calls in IOU, optional feature for S6)- sum(c.cell_tch_tch) ;(Intra cell HOs)

100 * ------------------------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch+c.bsc_i_sdcch_tch) ;(DR inter-cell calls)+ sum(c.cell_sdcch_tch) ;(DR intra-cell calls in IOU, optional feature for S6)+ sum(a.tch_seiz_due_sdcch_con) ; calls started as FACCH call setup


Figure 432. Handover/call % (trf_13e)

IUO, average TCH seizure length on super TRXs, S4 (trf_14b)

call time/tch seizures = average period duration * average traffic / tch seizures

avg of BTS (avg of TRX (period_duration))*60 ! Avg.call time in seconds*sum of BTS (sum of super TRX(ave_busy_tch))

= ------------------------------------------------------ secsum of BTS( sum of super TRX(tch_succ_seiz))


Figure 433. IUO, average TCH seizure length on super TRXs, S4 (trf_14b)

IUO, average TCH seizure length on regular TRXs, S4 (trf_15b)

call time/tch seizures = average period duration * average traffic / tch seizures

avg of BTS (avg of TRX (period_duration))*60*sum of BTS (sum of regular TRX(avg_trx_traf))

= ------------------------------------------------------ secsum of BTS( sum of regular TRX(tch_succ_seiz))




Figure 434. IUO, average TCH seizure length on regular TRXs, S4 (trf_15b)

Average TRX traffic, IUO, S4 (trf_16)

avg(ave_busy_tch)


Figure 435. Average TRX traffic, IUO, S4 (trf_16)

Average TRX TCH seizure length, IUO, S4 (trf_17)

count(*) avg(ave_busy_tch)*3600-------------------------------

sum(tch_succ_seiz)


Figure 436. Average TRX TCH seizure length, IUO, S4 (trf_17)

Average TRX TCH seizure length, IUO, S4 (trf_17a)

count(*) avg(ave_busy_tch)* period_duration*60-----------------------------------------------

sum(tch_succ_seiz)


Figure 437. Average TRX TCH seizure length, IUO, S4 (trf_17a)

Average TRX TCH seizure length, IUO, S4 (trf_17b)

sum(ave_busy_tch* period_duration*60)-----------------------------------------------

sum(tch_succ_seiz)

Counters from table(s):p_nbsc_underlayUnit: second

Figure 438. Average TRX TCH seizure length, IUO, S4 (trf_17b)



TCH requests for a new call, S3 (trf_18)

Known problems: A interface pool circuit type mismatch related retries areincluded.

sum(tch_call_req)


Figure 439. TCH requests for a new call, S3 (trf_18)

TCH requests for a new call, S3 (trf_18a)

sum(a.tch_call_req)- sum(a.tch_rej_due_req_ch_a_if_crc)- (b.bsc_i_unsucc_a_int_circ_type+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type)


Figure 440. TCH requests for a new call, S3 (trf_18a)

Peak busy TCH (trf_19)

Use: This PI is an important traffic load indicator on the cell level.By following the development of this indicator and reactingproactively, blocking can be avoided in cells where the trafficgrows smoothly.

max(peak_busy_tch)


Figure 441. Peak busy TCH (trf_19)

Average unit load (trf_20)

sum(load_rate)/sum(load_denom1)

Counters from table(s):p_nbsc_load

Figure 442. Average unit load (trf_20)



Call time difference between TRXs, S4 (trf_21)

Use: This PI shows as a percentage how much bigger the traffic ofthe busiest TRX of a BTS is compared to the least busy TRXof the same BTS.

100 * (max_call_samples - min_call_samples)/min_call_samples

wheremax_call_samples are the call samples of the busiest TRX of a BTS:max((ul_calls + dl_calls)/2)andmin_call_samples are the call samples of the least busy TRX of a BTS:min((ul_calls + dl_calls)/2)

ul_calls =sum(freq_ul_qual0

+ freq_ul_qual1+ freq_ul_qual2+ freq_ul_qual3+ freq_ul_qual4+ freq_ul_qual5+ freq_ul_qual6+ freq_ul_qual7)

dl_calls=sum(freq_dl_qual0

+ freq_dl_qual1+ freq_dl_qual2+ freq_dl_qual3+ freq_dl_qual4+ freq_dl_qual5+ freq_dl_qual6+ freq_dl_qual7)

Counters from table(s):p_nbsc_rx_qual

Figure 443. Call time difference between TRXs, S4 (trf_21)

Call time difference between TRXs, S4 (trf_21a)

Use: Shows how many times bigger the traffic of the busiest TRXof a BTS is compared to the least busy TRX of the same BTS.

max_call_samples/min_call_samples

wheremax_call_samples are the call samples of the busiest TRX of a BTS:max((ul_calls + dl_calls)/2)andmin_call_samples are the call samples of the least busy TRX of a BTS:min((ul_calls + dl_calls)/2)

ul_calls=sum(freq_ul_qual0

+ freq_ul_qual1+ freq_ul_qual2+ freq_ul_qual3+ freq_ul_qual4+ freq_ul_qual5+ freq_ul_qual6+ freq_ul_qual7)



dl_calls=sum(freq_dl_qual0+ freq_dl_qual1+ freq_dl_qual2+ freq_dl_qual3+ freq_dl_qual4+ freq_dl_qual5+ freq_dl_qual6+ freq_dl_qual7)


Figure 444. Call time difference between TRXs, S4 (trf_21a)

Number of mobiles located in a cell, BSC (trf_23a)

Use: If counted over an area, it could be possible to derive a KPIcalled ’Call minutes per MS’ from this formula.If used over a cell, it can give you an idea about how potentialthe cell is, for example.

How many times periodic LU has been sent = PLUS

How many times one MS sends a periodic LU in a time period =count_of_periods * period_duration/LU_period

X = number of MS

==>

X* count_of_periods * period_duration/LU_period = number ofperiodic updates (PLUS)

==> X = PLUS * LU_period/ (count_of_periods *period_duration)

sum(a.sdcch_loc_upd-nbr of incom.HO from other LA) * 0.1*b.timer_periodic_update_ms-------------------------------------------------------------------------------

count(*).a.period_duration/60

Counters from table(s):a = p_nbsc_res_accessb = c_bts

Figure 445. Number of mobiles located in a cell, BSC (trf_23a)



Note

The sum of incoming handovers from other location areas has to be counted fromp_nbsc_ho_adj using the LA info from the c_bts table.

* b.timer_periodic_update_ms and a.period_duration should be ofthe same unit, minutes for example.

Total TCH seizure time (call time in seconds) (trf_24b)

Note: The sampling takes place every 20 seconds. ave_busy_tchcounts cumulatively the number of busy TCHs.res_av_denom14 counts the number of samples taken.This is not pure conversation time but TCH seizure time. InHO there are two TCHs seized for a short time simultaneouslyand both may be counted if both seizures take place at thesampling moment.

sum(period_duration*60*ave_busy_tch/res_av_denom14)

Counters from table(s):p_nbsc_res_availunit: seconds

Figure 446. Total TCH seizure time (call time in seconds) (trf_24b)

Total TCH seizure time (call time in hours) (trf_24c)

Note: 1. See trf_24b.2. On BTS level. For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs.

sum(period_duration*ave_busy_tch/res_av_denom14/60)

Counters from table(s):p_nbsc_res_availunit: erlang hour

Figure 447. Total TCH seizure time (call time in hours) (trf_24c)

SDCCH true seizures (trf_25)

Known problems: There is no counter for IMSI detaches until in release S7.

sum(succ_seiz_term+succ_seiz_orig+sdcch_call_re_est+sdcch_emerg_call+sdcch_loc_upd)




Figure 448. SDCCH true seizures (trf_25)

SDCCH true seizures, S7 (trf_25a)

Known problems: There is no counter for supplemetary service requests until inrelease S9.

sum(succ_seiz_term+succ_seiz_orig+sdcch_call_re_est+sdcch_emerg_call+sdcch_loc_upd+imsi_detach_sdcch)


Figure 449. SDCCH true seizures, S7 (trf_25a)

SDCCH true seizures for call and SS (trf_26)

Known problems: Supplementary services cannot be separated currently on thecell level.

sum(succ_seiz_term+succ_seiz_orig+sdcch_call_re_est+sdcch_emerg_call- succ_sdcch_sms_est- unsucc_sdcch_sms_est)


Figure 450. SDCCH true seizures for call and SS (trf_26)

SDCCH true seizures for call, SMS, SS (trf_27)

Known problems: Supplementary services cannot be separated.

sum(succ_seiz_term+succ_seiz_orig+sdcch_call_re_est+sdcch_emerg_call)


Figure 451. SDCCH true seizures for call, SMS, SS (trf_27)

Peak busy SDCCH seizures (trf_28)

Use: The peak value of SDCCH usage is needed for preventivecapacity monitoring on the cell level.

max(peak_busy_sdcch)




Figure 452. Peak busy SDCCH seizures (trf_28)

IUO layer usage % (trf_29)

Use: Counted for overlay TRXs or underlay TRXs.sum(ave_busy_tch)

100 * -------------------------- %sum(ave_full_tch_if1

+ ave_full_tch_if2+ ave_full_tch_if3+ ave_full_tch_if4+ ave_full_tch_if5)

+ sum(ave_busy_tch)


Figure 453. IUO layer usage % (trf_29)

SDCCH seizures (trf_30)

Use: This figure tells the number of all events that have seizedSDCCH.

sum(sdcch_assign+sdcch_ho_seiz)


Figure 454. SDCCH seizures (trf_30)

Call time (minutes) from p_nbsc_res_avail (trf_32)

sum(period_duration * ave_busy_tch/res_av_denom14)

Counters from table(s):p_nbsc_res_availunit = minute

Figure 455. Call time (minutes) from p_nbsc_res_avail (trf_32)



Non-AMR call time from p_nbsc_rx_qual (trf_32a)

Known problems: In a high load situation (OMU link) it is possible that all calltime is not measured. In other words, call time can show alower value than it has in reality.Also, in the beginning of a call and in handover, two samplesare lost, showing a shorter time than in reality.

0.48*sum(freq_ul_qual0+ freq_ul_qual1+ freq_ul_qual2+ freq_ul_qual3+ freq_ul_qual4+ freq_ul_qual5+ freq_ul_qual6+ freq_ul_qual7)

/60

Counters from table(s):p_nbsc_rx_qualunit = minutes

Figure 456. Non-AMR call time from p_nbsc_rx_qual (trf_32a)

Call time from p_nbsc_rx_statistics (trf_32b)

Known problems: In a high load situation it is possible that all call time is notmeasured. In other words, call time can show a lower valuethan it has in reality.

0.48*sum(freq_ul_qual0+ freq_ul_qual1+ freq_ul_qual2+ freq_ul_qual3+ freq_ul_qual4+ freq_ul_qual5+ freq_ul_qual6+ freq_ul_qual7)

/60

Counters from table(s):p_nbsc_rx_statisticsunit = minutes

Figure 457. Call time from p_nbsc_rx_statistics (trf_32b)

SDCCH HO seizure % out of SDCCH seizure attempts (trf_33)

sum(sdcch_ho_seiz)100 * -------------------- %

sum(sdcch_seiz_att)




Figure 458. SDCCH HO seizure % out of SDCCH seizure attempts (trf_33)

SDCCH assignment % out of SDCCH seizure attempts (trf_34)

Formula:sum(sdcch_assign)

100 * -------------------- %sum(sdcch_seiz_att )


Figure 459. SDCCH assignment % out of SDCCH seizure attempts (trf_34)

TCH HO seizure % out of TCH HO seizure request (trf_35)

sum(tch_ho_seiz100 * ---------------------------------------------- %

sum(tch_request - tch_call_req - tch_fast_req)


Figure 460. TCH HO seizure % out of TCH HO seizure request (trf_35)

TCH norm seizure % out of TCH call request (trf_36)

sum(tch_norm_seiz)100 * --------------------------- %

sum(tch_call_req)


Figure 461. TCH norm seizure % out of TCH call request (trf_36)

TCH normal seizure % out of TCH call requests (trf_36a)

sum(tch_norm_seiz)-sum(tch_succ_seiz_for_dir_acc); ref.1

100 * --------------------------------------- %sum(tch_call_req)




Figure 462. TCH normal seizure % out of TCH call requests (trf_36a)

Ref.1 tch_norm_seiz is triggered also in case of DAC.

TCH FCS seizure % out of TCH FCS attempts (trf_37)

sum(tch_seiz_due_sdcch_con)100 * ----------------------------------- %

sum(tch_seiz_att_due_sdcch_con)


Figure 463. TCH FCS seizure % out of TCH FCS attempts (trf_37)

TCH FCS (due to SDCCH congestion) seizure % out of SDCCH seizureattempts (trf_38)

Use: Indicates the percentage of SDCCH seizures saved byFACCH call setup.

sum(tch_seiz_due_sdcch_con)100 * ----------------------------------- %

sum(sdcch_seiz_att)


Figure 464. TTCH FCS (due to SDCCH congestion) seizure % out of SDCCHseizure attempts (trf_38)

TCH seizures for new calls (trf_39)



Figure 465. TCH seizures for new calls (trf_39)



TCH seizures for new calls (trf_39a)

sum(a.tch_norm_seiz) ;(normal calls)- sum(a.succ_tch_seiz_for_dir_acc); ref.2+ sum(c.msc_i_sdcch_tch+c.bsc_i_sdcch_tch) ;(DR inter-cell calls)+ sum(c.cell_sdcch_tch) ;(DR intra-cell calls in IOU)+ sum(a.tch_seiz_due_sdcch_con) ; calls started as FACCH call setup


Figure 466. TCH seizures for new calls (trf_39a)


HTCH usage, S5 (trf_40)

Use: On the area level gives you an idea of how well the capacityis used. Use with S5 or a later version.

sum(ave_tch_busy_halfl)= 100 * ----------------------------- %

sum(ave_tch_avail_half)


Figure 467. HTCH usage, S5 (trf_40)

MOC rate, S6 (trf_41)

Known problems: Do not include SMS, SS ==> Better accuracy for speech callsthan if 3012 and 3013 were used.If SDCCH were congested and FACCH used for SMS (SS?),also SMS and SS get included.

tch_moc_seiz_att100 * ------------------------------------ %

tch_moc_seiz_att + tch_mtc_seiz_att


Figure 468. MOC rate, S6 (trf_41)



MTC rate, S6 (trf_42)

Known problems: Do not include SMS, SS ==> Better accuracy for speech callsthan if 3012 and 3013 were used.If SDCCH were congested and FACCH used for SMS (SS?)then also SMS and SS are included.

tch_moc_seiz_att100 * ------------------------------------ %

tch_moc_seiz_att + tch_mtc_seiz_att


Figure 469. MTC rate, S6 (trf_42)

TCH single band subscriber holding time, S6 (trf_43)

0.48* sum(tch_single_band_hold_time)

Counters from table(s):p_nbsc_dual_bandUnit: second

Figure 470. TCH single band subscriber holding time, S6 (trf_43)

TCH dual band subscriber holding time, S6 (trf_44)

0.48* sum(tch_dual_band_hold_time)

Unit: secondCounters from table(s):p_nbsc_dual_band

Figure 471. TCH dual band subscriber holding time, S6 (trf_44)

Average SDCCH traffic, S1 (trf_45)


SDCCH used time / period duration =

(sum(a.ave_sdcch_hold_tim) / sum(a.res_av_denom16)/100)*sum(b.sdcch_assign + b.sdcch_ho_seiz + b.tch_seiz_due_sdcch_con)-----------------------------------------------------------------sum(a.period_duration*60)

Counters from table(s):a = p_nbsc_res_avail



b = p_nbsc_trafficUnit: Erlang

Figure 472. Average SDCCH traffic, S1 (trf_45)

Share of single band traffic (trf_47)

sum(tch_single_band_hold_time)100* -------------------------------------------------------- %

sum(tch_single_band_hold_time + tch_dual_band_hold_time)

Counters from table(s):p_nbsc_dual_band.

Figure 473. Share of single band traffic (trf_47)

Share of dual band traffic (trf_48)

sum(tch_dual_band_hold_time)100* -------------------------------------------------------- %

sum(tch_single_band_hold_time + tch_dual_band_hold_time)


Figure 474. Share of dual band traffic (trf_48)

Call retries due to A interface pool mismatch (trf_49)

Use: Compensation of the blocking caused by the A interfacecircuit pool mismatch.

Aif type mismatch or congestion - Aif circuit pool handover failure =

a.tch_rej_due_req_ch_a_if_crc- (b.bsc_i_unsucc_a_int_circ_type+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type)


Figure 475. Call retries due to A interface pool mismatch (trf_49)

HO retries due to A interface pool mismatch (trf_50)

Use: Compensation of blocking caused by the A interface circuitpool mismatch.

Sum(bsc_i_unsucc_a_int_circ_type



+ msc_controlled_in_ho+ ho_unsucc_a_int_circ_type)


Figure 476. HO retries due to A interface pool mismatch (trf_50)

TCH single band subscribers’ share of holding time, S6 (trf_51)

sum(tch_single_band_hold_time)100 * ------------------------------------------------------ %

sum(tch_single_band_hold_time+tch_dual_band_hold_time)


Figure 477. TCH single band subscribers’ share of holding time, S6 (trf_51)

TCH dual band subscribers’ share of holding time, S6 (trf_52)

sum(tch_dual_band_hold_time)100 * ------------------------------------------------------ %

sum(tch_single_band_hold_time+tch_dual_band_hold_time)

Unit: secondCounters from table(s):p_nbsc_dual_band

Figure 478. TCH dual band subscribers’ share of holding time, S6 (trf_52)

Calls started as FACCH call setup (trf_53)

Use: If there is SDCCH congestion and dynamic SDCCHallocation is not capable of allocating more SDCCH, thesignalling can take place on TCH if there is free capacity, anda call can be established.

sum(tch_seiz_att_due_sdcch_con)

Counters from table(s):p_nbsc_trafficUnit: second

Figure 479. Calls started as FACCH call setup (trf_53)

SDCCH seizures (trf_54)

sum(sdcch_assign+sdcch_ho_seiz)




Figure 480. SDCCH seizures (trf_54)

TCH normal seizures (trf_55)

sum(tch_norm_seiz)-sum(tch_succ_seiz_for_dir_acc); ref.1


Figure 481. TCH normal seizures (trf_55)

Ref.1 The counter tch_norm_seiz is triggered also in the case of DAC.

Total FTCH seizure time (trf_56)

sum(period_duration*ave_tch_busy_full/60)

Counters from table(s):p_nbsc_res_availunit = hour

Figure 482. Total FTCH seizure time (trf_56)

Total HTCH seizure time (trf_57)

sum(period_duration*ave_tch_busy_half/60)


Figure 483. Total HTCH seizure time (trf_57)

Average TCH hold time for HSCSD, S7 (trf_58)

Use: The numerator counts the cumulative sum of the TCH holdingtimes for HSCSD. The numerator is the number of HSCSDTCH releases.

Known problems: Incorrect if extended TRXs are used and not counted onBTS/trx_type level.

sum(ave_tch_hold_time_hscsd_numer)----------------------------------- sec



sum(ave_tch_hold_time_hscsd_denom)*100


Figure 484. Average TCH hold time for HSCSD, S7 (trf_58)

Average number of HSCSD users, S7HS (trf_60)

sum(ave_hscsd_users_numer)--------------------------sum(ave_hscsd_users_denom)


Figure 485. Average number of HSCSD users, S7HS (trf_60)

Average HSCSD main channel traffic, S7HS (trf_60a)

Use: On BTS level. For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs.

trf_104 ; HSCSD main channel traffic,normal TRXs+ trf_105 ; HSCSD main channel traffic,extended TRXs

Unit: erlang

Figure 486. Average HSCSD main channel traffic, S7HS (trf_60a)

Total HSCSD TCH seizure time (call time, hours), S7HS (trf_61)

period_duration * ave_busy_tch_hscsd_numersum(-------------------------------------------)

ave_busy_tch_hscsd_denom/60


Figure 487.

Average upgrade pending time for HSCSD (trf_62)

sum(ave_pend_time_numer)----------------------------sum(ave_pend_time_denom)*100



Counters from table(s):p_nbsc_high_speed_dataUnit: sec

Figure 488. Average upgrade pending time for HSCSD (trf_62)

Average upgrade pending time due to congestion (trf_63)

sum(ave_pend_time_due_cong_numer)---------------------------------------sum(ave_pend_time_due_cong_denom)*100

Counters from table(s):p_nbsc_high_speed_dataUnit: sec

Figure 489. Average upgrade pending time due to congestion (trf_63)

Total reporting time of ph1 and ph2 mobiles (trf_64)

sum(rep_time_by_ph_1_ms + rep_time_by_ph_2_ms)*0,46/60

Counters from table(s):p_nbsc_ms_capabilityUnit: min

Figure 490. Total reporting time of ph1 and ph2 mobiles (trf_64)

Total TCH seizures (trf_65)

sum(tch_reserv_by_mslot_cl_1_ms + ... + tch_reserv_by_mslot_cl_18_ms+tch_reserv_by_mslot_incap_ms)

Counters from table(s):p_nbsc_ms_capability

Figure 491. Total TCH seizures (trf_65)

Net UL data traffic per timeslot, S9PS (trf_69a)

Use: Gives an idea of how effectively the GPRS timeslots are used.

data in kilobits------------------------------------------------------- =total time * average number of GPRS timeslots

sum(a.RLC_data_blocks_UL_CS1*20+a.RLC_data_blocks_UL_CS2*30)* 8/1000-------------------------------------------------------------------------------sum(b.period_duration*60)*sum(b.ave_GPRS_channels_sum)/sum(b.ave_GPRS_channels_den)



Counters from table(s):a = p_nbsc_packet_control_unitb = p_nbsc_res_availUnit: kbit/sec/tsl

Figure 492. Net UL data traffic per timeslot, S9PS (trf_69a)

Net DL data traffic per timeslot, S9PS (trf_70a)

Use: Gives an idea of how effectively the GPRS timeslots are used.

data in kilobits------------------------------------------------------- =total time * average number of GPRS timeslots

sum(a.RLC_data_blocks_DL_CS1*20+a.RLC_data_blocks_DL_CS2*30)* 8/1000-------------------------------------------------------------------------------sum(b.period_duration*60)*sum(b.ave_GPRS_channels_sum)/sum(b.ave_GPRS_channels_den)

Counters from table(s):a= p_nbsc_packet_control_unitb = p_nbsc_res_availUnit: kbit/sec/tsl

Figure 493. Net DL data traffic per timeslot, S9PS (trf_70a)

Average UL throughput per allocated timeslot, S9PS (trf_72b)

Use: Indicates the net data rate per allocated channel. The lower thevalue the more loaded is the GPRS territory and the lessservice the MS users receive.The numerator does not contain the RLC header bytes (2)neither the MAC header (1) because the aim is to count datavolume from the user’s point of view.

Known problems: 1) The formula works after S9 CD1.2, seeave_dur_UL_TBF_sum.2) The number of TBFs (MS) sharing the same timeslotsvaries.3) The data blocks of abnormally released TBFs and TBFsthat are not yet released during the measurement periodincrease the amount of data during the period, but are nottaken into account in the duration counters, i.e. they do notincrease the total duration of TBF4) Another inaccuracy is the ‘average allocated tsl’. Now eachallocation in the formula is of equal weight. To be correct,each allocation should be weighted by its duration.5) After S9 BSC CD 4.0, the ‘Delayed TBF release’modification in PCU adds the TBF holding time and thusmakes this KPI show smaller values.

data in kilobits/ TBF total time---------------------------------------- =



average allocated tsl

(sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30)* 8/1000----------------------------------------------------------------------

sum(Ave_dur_UL_TBF_sum/100)----------------------------------------------------------------------

(sum(alloc_1_TSL_UL+2*alloc_2_TSL_UL+3*alloc_3_TSL_UL +4*alloc_4_TSL_UL)------------------------------------------------------------------------

sum(alloc_1_TSL_UL+alloc_2_TSL_UL+alloc_3_TSL_UL+alloc_4_TSL_UL))


Unit: kbit/sec/tsl

Figure 494. Average UL throughput per allocated timeslot, S9PS (trf_72b)

Average effective UL throughput per used tsl, S9PS (trf_72d)

Use: Indicates net data rate per used timeslot. The lower the valuethe more loaded is the GPRS territory and the less service theMS users receive.The numerator does not contain the RLC header bytes neitherdoes the MAC header because the aim is to count data volumefrom as close to the user’s point of view as possible.The denominator is built on the fact that one timeslot cancarry 50 RLC blocks per second.

Known problems: 1) The numerator is not yet pure user data but as close to thatas we can see from BSC counters.2) Note that this KPI has correlation with DL data becauserlc_mac_cntrl_blocks_ul gets triggered for each PacketDownlink ACK/NACK. If the DL retransmissions get morefrequent (radio interface quality worse or polling parametershave been modified) the polling becomes more frequent andtherefore rlc_mac_cntrl_blocks_ul gets triggered more often.This leads to a situation where the effective UL throughputseems to get worse even though nothing really has changed inUL.

UL payload data in (kbytes)---------------------------------------- =UL time for data transfer (sec)

sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30)*8 /1000--------------------------------------------------------------------sum(rlc_data_blocks_ul_cs1

+ rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul+ BAD_FRAME_IND_UL_CS1+ BAD_FRAME_IND_UL_CS2+ BAD_FRAME_IND_UL_UNACK+ IGNOR_RLC_DATA_BL_UL_DUE_BSN) /50




p_nbsc_packet_control_unitunit: Kbps / tsl

Figure 495. Average effective UL throughput per used tsl, S9PS (trf_72d)

Average effective UL throughput per used timeslot, S10PS (trf_72f)

Use: Indicates the net data rate per used timeslot. The lower thevalue the more loaded is the GPRS territory and the lessservice the MS users receive.The numerator does not contain the RLC header bytes neitherdoes the MAC header because the aim is to count data volumefrom as close to the user’s point of view as possible.The denominator is built on the fact that one timeslot cancarry 50 RLC blocks per second.

Known problems: The numerator is not yet pure user data but as close to that aswe can see from BSC counters.

UL payload data in (kilobits)-------------------------------- =UL time for data transfer (sec)

sum(a.RLC_data_blocks_UL_CS1*20 + a.RLC_data_blocks_UL_CS2*30)*8 /1000+(sum over MCS-1 (xx)*22+sum over MCS-2 (xx)*28+sum over MCS-3 (xx)*37+sum over MCS-4 (xx)*44+sum over MCS-5 (xx)*56+sum over MCS-6 (xx)*74+sum over MCS-7 (xx)*56+sum over MCS-8 (xx)*68+sum over MCS-9 (xx)*74)*8/1000

------------------------------------------------------------------------sum(a.rlc_data_blocks_ul_cs1

+ a.rlc_data_blocks_ul_cs2+ a.rlc_mac_cntrl_blocks_ul+ a.BAD_FRAME_IND_UL_CS1+ a.BAD_FRAME_IND_UL_CS2+ a.BAD_FRAME_IND_UL_UNACK+ a.IGNOR_RLC_DATA_BL_UL_DUE_BSN) /50+ sum over MSC1?6 of (yy)/50+ sum over MSC7?9 of (yy)/2 /50

wherexx = b.ul_rlc_blocks_in_ack_mode + b.ul_rlc_blocks_in_unack_modeyy =b.ul_rlc_blocks_in_ack_mode

+b.retrans_rlc_data_blocks_ul+b.bad_rlc_valid_hdr_ul_unack+b.ul_rlc_blocks_in_unack_mode

Counters from table(s):a = p_nbsc_packet_control_unitb = p_nbsc_coding_scheme

Figure 496. Average effective UL throughput per used TSL, S10PS (trf_72f)



Average effective UL throughput per used TSL, S10PS (trf_72h)

Use: Indicates the net data rate per used timeslot. The lower thevalue the more loaded is the GPRS territory and the lessservice the MS users receive.The numerator does not contain the RLC header bytes neitherthe does the MAC header because the aim is to count datavolume from the user's point of view as close as possible.The denominator is build on the fact that one timeslot cancarry 50 RLC blocks per second.

Known problems: The numerator is not yet pure user data but as close to that aswe can see from BSC counters.

UL payload data in (kilobits)---------------------------------------- =UL time for data transfer (sec)

sum(a.RLC_data_blocks_UL_CS1*20+a.RLC_data_blocks_UL_CS2*30)*8 /1000+(sum over MCS-1 (xx)*22+sum over MCS-2 (xx)*28+sum over MCS-3 (xx)*37+sum over MCS-4 (xx)*44+sum over MCS-5 (xx)*56+sum over MCS-6 (xx)*74+sum over MCS-7 (xx)*56+sum over MCS-8 (xx)*68+sum over MCS-9 (xx)*74)*8/1000--------------------------------------------------------------------sum(a.rlc_data_blocks_ul_cs1

+ a.rlc_data_blocks_ul_cs2+ a.BAD_FRAME_IND_UL_CS1+ a.BAD_FRAME_IND_UL_CS2+ a.BAD_FRAME_IND_UL_UNACK+ a.IGNOR_RLC_DATA_BL_UL_DUE_BSN) /50

+sum over MSC1...6 of (yy)/50+sum over MSC7...9 of (yy)/2 /50

wherexx = b.ul_rlc_blocks_in_ack_mode

+ b.ul_rlc_blocks_in_unack_modeyy = b.ul_rlc_blocks_in_ack_mode

+ b.retrans_rlc_data_blocks_ul+ b.bad_rlc_valid_hdr_ul_unack+ b.ul_rlc_blocks_in_unack_mode

Counters from table(s):a=p_nbsc_packet_control_unitb=p_nbsc_coding_scheme

Figure 497. Average effective UL throughput per used TSL, S10PS (trf_72h)



Average DL throughput per allocated timeslot, S9PS (trf_73b)

Use: Indicates the net data rate per allocated channel. The lower thevalue the more loaded the GPRS territory and the less servicethe MS users receive.The numerator does not contain the RLC header bytes (2)neither the MAC header (1) because the aim is to count datavolume from the user’s point of view.

Known problems: 1)The formula works after S9 CD1.2, seeave_dur_UL_TBF_sum.2) Number of TBFs (MS) sharing the same timeslot varies.3) Abnormally released TBFs as well as TBFs that are not yetreleased during the measurement period . The data blocks ofthose increase the amount of data during the period, but theyare not taken into account in the duration counters, i.e. they donot increase the total duration of TBFs. This situation occurswhen TBF completion ratio tbf_26a shows a low value.4) Another inaccuracy is the 'average allocated tsl'. Now eachallocation in the formula has equal weight. To be correct, eachallocation should be weighed by it’s duration.5) After S9 BSC CD 4.0, the 'Delayed TBF release'modification in PCU adds the TBF holding time andtherefore causes this KPI to show smaller values.

data in kilobits/ TBF total time----------------------------------- =average allocated tsl

(sum(RLC_data_blocks_DL_CS1*20+ RLC_data_blocks_DL_CS2*30)* 8/1000

------------------------------------------sum(Ave_dur_DL_TBF_sum/100 )

--------------------------------------------------(sum(alloc_1_TSL_DL+ 2*alloc_2_TSL_DL+ 3*alloc_3_TSL_DL+ 4*alloc_4_TSL_DL)

------------------------sum(alloc_1_TSL_DL

+ alloc_2_TSL_DL+ alloc_3_TSL_DL+ alloc_4_TSL_DL))


Unit: kbit/sec/tsl

Figure 498. Average DL throughput per allocated timeslot, S9PS (trf_73b)



Average effective DL throughput per used timeslot, S9PS (trf_73d)

Use: Indicates the net data rate per used timeslot. The lower thevalue the more loaded is the GPRS territory and the lessservice the MS users receive.The numerator does not contain the RLC header bytes neitherthe MAC header because the aim is to count data volume fromthe user’s point of view as close as possible.The denominator is built on the fact that one timeslot cancarry 50 RLC blocks per second.

Known problems: 1) The numerator is not yet pure user data but as close to thatas we can see from BSC counters.2) Retransmitted blocks due to other reasons than NACK arenot counted in any of the RLC counters. In DL direction theseretransmissions occur when the TBF release is delayed.3) If there is only one TBF on a timeslot, for example, someRLC blocks can be retransmitted before an ACK is received.These blocks are not counted in any of the RLC counters.4) Counter rlc_mac_cntrl_blocks_dl also containsdummy blocks until CD.6.1.

DL ’payload’ data in (kilobit)---------------------------------------- =DL time for data transfer (sec)

sum(RLC_data_blocks_DL_CS1*20+RLC_data_blocks_DL_CS2*30)*8 /1000-----------------------------------------------------------------sum(rlc_data_blocks_dl_cs1

+ rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl+ RETRA_RLC_DATA_BLOCKS_DL_CS1+ RETRA_RLC_DATA_BLOCKS_DL_CS2) /50


Unit: Kbps / tsl

Figure 499. Average effective DL throughput per used timeslot, S9PS (trf_73d)

Average effective DL throughput per used timeslot, S10PS (trf_73f)

Use: Indicates the net data rate per used timeslot. The lower thevalue the more loaded is the GPRS territory and the lessservice the MS users receive.The numerator does not contain the RLC header bytes neitherdoes the MAC header, because the aim is to count datavolume from as close to the user’s point of view as possible.



Known problems: 1) The numerator is not yet pure user data but as close to thatas we can see from BSC counters.2) Retransmitted blocks due to other reasons than NACK arenot counted in any of the RLC-counters. In DL direction theseretransmissions occur when the TBF release is delayed.3) If there is only one TBF on a timeslot, for example, someRLC blocks can be retransmitted before an ACK is received.These blocks are not counted in any of the RLC counters.

DL payload data in (kbytes)---------------------------------- =DL time for data transfer (sec)

sum(a.RLC_data_blocks_DL_CS1*20 + a.RLC_data_blocks_DL_CS2*30)*8 /1024+(sum over MCS-1 (xx)*22+sum over MCS-2 (xx)*28+sum over MCS-3 (xx)*37+sum over MCS-4 (xx)*44+sum over MCS-5 (xx)*56+sum over MCS-6 (xx)*74+sum over MCS-7 (xx)*56+sum over MCS-8 (xx)*68+sum over MCS-9 (xx)*74)*8/1024

----------------------------------------------------------------------sum(a.rlc_data_blocks_dl_cs1

+ a.rlc_data_blocks_dl_cs2+ a.rlc_mac_cntrl_blocks_dl - a.dummy_dl_mac_blocks_sent+ a.RETRA_RLC_DATA_BLOCKS_DL_CS1+ a.RETRA_RLC_DATA_BLOCKS_DL_CS2) /50+ sum over msc1...6 of (yy)/50+ sum over msc7...9 of (yy)/2/50

wherexx = b.dl_rlc_blocks_in_ack_mode + b.dl_rlc_blocks_in_unack_modeyy =b.dl_rlc_blocks_in_ack_mode

+b.retrans_rlc_data_blocks_dl+b.dl_rlc_blocks_in_unack_mode


Unit: kbit/sec/tsl

Figure 500. Average effective DL throughput per used timeslot, S10PS (trf_73f)

Average effective DL throughput per used TSL, S10PS (trf_73g)

Use: Indicates the net data rate per used timeslot. The lower thevalue the more loaded is the GPRS territory and the lessservice the MS users receive. The numerator does not containthe RLC header bytes neither the does the MAC headerbecause the aim is to count data volume from the user's pointof view as close as possible



Known problems: 1) The numerator is not yet pure user data but as close to thatas we can see from BSC counters.2) Retransmitted blocks due to other reasons than NACK arenot counted in any of the RLC-counters. In DL direction theseretransmissions occur when the TBF release is delayed.3) If there is only one TBF on a timeslot, for example, someRLC blocks can be retransmitted before an ACK is received.These blocks are not counted in any of the RLC counters.

DL payload data in (kilobits)---------------------------------------- =DL time for data transfer (sec)

sum(a.RLC_data_blocks_DL_CS1*20+ a.RLC_data_blocks_DL_CS2*30)*8 /1000

+ (sum over MCS-1 (xx)*22++ sum over MCS-2 (xx)*28++ sum over MCS-3 (xx)*37++ sum over MCS-4 (xx)*44++ sum over MCS-5 (xx)*56++ sum over MCS-6 (xx)*74++ sum over MCS-7 (xx)*56++ sum over MCS-8 (xx)*68++ sum over MCS-9 (xx)*74)*8/1000

---------------------------------------------sum(a.rlc_data_blocks_dl_cs1

+ a.rlc_data_blocks_dl_cs2+ a.RETRA_RLC_DATA_BLOCKS_DL_CS1+ a.RETRA_RLC_DATA_BLOCKS_DL_CS2) /50

+ sum over msc1…6 of (yy)/50+ sum over msc7…9 of (yy)/2/50

wherexx = b.dl_rlc_blocks_in_ack_mode

+ b.dl_rlc_blocks_in_unack_modeyy = b.dl_rlc_blocks_in_ack_mode

+ b.retrans_rlc_data_blocks_dl+ b.dl_rlc_blocks_in_unack_mode


Unit: kbit/sec/tsl

Figure 501. Average effective DL throughput per used TSL, S10PS (trf_73g)

Total RLC data, S9PS (trf_74)

Use: Indicates the total amount of data transmitted as CS1 or CS2blocks, UL or DL.

(sum(RLC_data_blocks_UL_CS1*23 + RLC_data_blocks_UL_CS2*33+ RLC_data_blocks_DL_CS1*23 + RLC_data_blocks_DL_CS2*33) /1000




Unit: kbyte

Figure 502. Total RLC data, S9PS (trf_74)

Total RLC data, S9PS (trf_74a)

Use: Indicates the total amount of data (both ack and unack modes)transmitted as CS1 or CS2 blocks. UL or DL. MAC blocksand RLC header bytes are excluded in order to get as close aspossible to the payload data.

Known problems: The divisor should be 1024 for Kbytes.

(sum(RLC_data_blocks_UL_CS1*20+RLC_data_blocks_UL_CS2*30+ RLC_data_blocks_DL_CS1*20+RLC_data_blocks_DL_CS2*30) /1000


Unit: kbyte

Figure 503. Total RLC data, S9PS (trf_74a)

Total GPRS RLC data, S9PS (trf_74b)

Use: Indicates the total amount of data (both ack and unack modes)transmitted as CS1 or CS2 blocks. UL or DL. MAC blocksand RLC header bytes are excluded in order to get as close aspossible to the payload data.

(sum(RLC_data_blocks_UL_CS1*20 + RLC_data_blocks_UL_CS2*30+ RLC_data_blocks_DL_CS1*20 + RLC_data_blocks_DL_CS2*30) /1024

Counters from table(s):p_nbsc_packet_control_unitUnit: kbyte

Figure 504. Total GPRS RLC data, S9PS (trf_74b)

GPRS territory UL utilisation, S9PS (trf_76b)

Use: Most useful on BTS level. Used as BH (CS+PS) trendtogether with CS traffic, total TCH capacity and PS territorysize.Indicates how big a portion of the GPRS territory has beenused. When the utilisation % increases, the throughput rateperceived by the user reduces. This KPI is the way to estimatethe throughput rate reduction.



If the utilisation % is high, increasing the CDEF parametersetting (when CS traffic is low) or adding a new TRX (whenCS traffic is high) should be considered.UL and DL should be looked at the same time and the higherone of those used in dimensioning together with CS traffictrf_12b and PS territory ava_16.

Note: The denominator varies depending on the traffic situation(downgrade of territory) and therefore this KPI has no linearcorrelation with the traffic. See ava_15.

Target level: Choosing the acceptable level is a QoS issue. For example,60% utilisation would give about 70% of the maximum rateachieved.

Data blocks transmitted in UL100* ------------------------------------------------------------ % =

(available GPRS channel time in sec)* (nbr of blocks per sec)

100*(DL blocks transmitted / DL block transmission capacity) % =

sum(rlc_data_blocks_ul_cs1+ rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul+ BAD_FRAME_IND_UL_CS1+ BAD_FRAME_IND_UL_CS2+ BAD_FRAME_IND_UL_UNACK+ IGNOR_RLC_DATA_BL_UL_DUE_BSN)

100* -------------------------------------------------------------- %sum(a.ave_gprs_channels_sum/sum(a.ave_gprs_channels_den)*sum(a.period_duration*60)*50

Counters from table(s):a = p_nbsc_res_availb = p_nbsc_packet_control_unit

Figure 505. GPRS territory UL utilisation, S9PS (trf_76b)

GPRS territory DL utilisation, S9PS (trf_77a)

Use: Most useful on BTS level. Used as BH (CS+PS) trendtogether with CS traffic, total TCH capacity and PS territorysize.Indicates how big a portion of the GPRS territory has beenused. When the utilisation % increases, the throughput rateperceived by the user reduces. This KPI is the way to estimatethe throughput rate reduction.If the utilisation % is high, increasing the CDEF parametersetting (when CS traffic is low) or adding a new TRX (whenCS traffic is high) should be considered.UL and DL should be looked at the same time and the higherone of those used in dimensioning together with CS traffictrf_12b and PS territory ava_16.




Known problems: Dummy blocks on DL make this PI show too high a value(fixed in CD6.0: RLC_MAC_cntrl_blocks does not containdummy blocks any more).


Data blocks transmitted in DL and UL100* ------------------------------------------------------------ % =


100*(DL blocks transmitted / DL block transmission capacity) % =

sum(+ b.RLC_data_blocks_DL_CS1+ b.RLC_data_blocks_DL_CS2+ b.RLC_MAC_cntrl_blocks_DL+ b.retra_RLC_data_blocks_DL_CS1+ b.retra_RLC_data_blocks_DL_CS2)

100* --------------------------------------------------------------- %sum(a.ave_gprs_channels_sum/sum(a.ave_gprs_channels_den)

*sum(a.period_duration*60)*50

Counters from table(s):a = p_nbsc_res_availb = p_nbsc_packet_control_unit

Figure 506. GPRS territory DL utilisation, S9PS (trf_77a)

UL GPRS traffic, S9PS (trf_78a)

Use: Indicates the amount of resources (timeslots) the GPRS trafficdata consumes on average during the period. This informationis useful, for example, in forecasting the need for capacityextension.

Known problems: The value is optimistic because the time needed for TBFestablishment and release is not included. The delayed TBFrelease that was taken into use in a CD of S9 also adds theactual usage of the TCH but cannot be considered in thisformula. These make the value seem smaller.

Actual UL data throughput (blocks)---------------------------------------------- =max. nbr of blocks during measurement period

sum(rlc_data_blocks_ul_cs1+ rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul+ BAD_FRAME_IND_UL_CS1+ BAD_FRAME_IND_UL_CS2+ BAD_FRAME_IND_UL_UNACK+ IGNOR_RLC_DATA_BL_UL_DUE_BSN)

-----------------------------------sum(period_duration*60)*50




Unit: tsl (or erlang)

Figure 507. UL GPRS traffic, S9PS (trf_78a)

DL GPRS traffic, S9PS (trf_79a)

Use: Indicates the amount of resources (timelots) the GPRS trafficdata consumes. This information is useful, for example, inforecasting the need for capacity extension.

Known problems: 1) The MS can send an UL data block only if it has receivedits USF in the preceding DL block. If the network has nothingelse to send, it will send a Packet DL Dummy Control Blockto carry the USF. These dummy blocks are included in thisKPI until CD6.1 and make it show bigger values.2) Transferred DL blocks, whose corresponding element inthe transmit window V(B) has the value PENDING ACK, arenot counted to any of the counters.3) The time needed for TBF establishment and release is notincluded. Also the delayed TBF release that was taken intouse in a CD of S9 adds the actual usage of the TCH but cannotbe considered in this formula. These make the value seemsmaller.

Actual DLdata throughput (blocks)---------------------------------------------- =max. nbr of blocks during measurement period

sum(rlc_data_blocks_dl_cs1+ rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl+ RETRA_RLC_DATA_BLOCKS_DL_CS1+ RETRA_RLC_DATA_BLOCKS_DL_CS2)

----------------------------------sum(period_duration*60)*50

Counters from table(s):p_nbsc_packet_control_unitUnit: timeslot or erlang

Figure 508. DL GPRS traffic, S9PS (trf_79a)

TCH free margin, S9PS (trf_81)

Use: Indicates the average number of free TCH timeslots.Known problems: 1) DL MAC blocks contain dummy blocks.

2) Transferred DL blocks, whose corresponding element inthe transmit window V(B) has the value PENDING ACK, arenot counted to any of the counters.

Capacity available for CSW + dedicated PSW capacity - CSW traffic - PSW DL trafficava_15+ava_16-trf_12b-trf_79a



Counters from table(s):p_nbsc_res_availp_nbsc_packet_control_unit

Figure 509. TCH free margin, S9PS (trf_81)

Normal TCH usage % for CS (trf_83a)

Use: Indicates how many % of the total available normal TCHcapacity has been used for CS traffic on average. Used fortrend analysis.

Capacity used by CS traffic / total normal TCH capacity

trf_97= 100* ------------- %

ava_28+ava_16a

Figure 510. Normal TCH usage % for CS (trf_83a)

TCH usage % for PS, S9PS(trf_84a)

Use: Indicates how many % of the total available TCH capacity hasbeen used for PS traffic on average.

Known problems: 1) See the problems of trf_79a.2) For the absolute peak value there is no counter availableunlike for CS traffic.

Capacity used by PS traffic / total TCH capacity

trf_79a= 100* --------------- %

ava_15+ava_16

Figure 511. TCH usage % for PS, S9PS (trf_84a)

Normal TCH usage % for PS, S9PS (trf_84b)

Use: Indicates how many % of the total available normal TCHcapacity has been used for PS traffic on average. Used fortrend analysis.

trf_95= 100* ------------- %

ava_28+ava_16a

Figure 512. Normal TCH usage % for PS, S9PS (trf_84b)



Total TCH usage % for CS, S9PS(trf_85)

Use: Indicates how many % the of total TCH capacity has beenused for PS traffic.

Known problems: It is assumed that DL PS traffic is always greater than UL PStraffic.

Capacity used by CS and PS traffic / total TCH capacity

trf_12b +trf_79a= 100* --------------- %

ava_15+ava_16

Figure 513. Total TCH usage % for CS, S9PS (trf_85)

Total TCH usage % for CS and PS, S9PS (trf_85b)

Use: Indicates how many % of the total TCH capacity has beenused for PS traffic.

Known problems: It is assumed that DL PS traffic is always greater than UL PStraffic.

Capacity used by CS and PS traffic / total TCH capacity

trf_12b +trf_95= 100* --------------- %

ava_25a

Figure 514. Total TCH usage % for CS and PS, S9PS (trf_85b)

Free TCH %, S9PS (trf_86a)

Use: Most useful on BTS level in context with trf_83 and trf_84aKnown problems: Because the measurement period usually is 60min, the value

can not be used for spotting momentary problems but trends.

100- TCH usage % for CS - TCH usage % for PS= 100 - trf_83-trf_84a

Figure 515. Free TCH %, S9PS (trf_86a)

Free TCH %, S9PS (trf_86b)

Use: Most useful on BTS level in connection with trf_83 andtrf_84a. The combined (PS+CS traffic) BH value trend can beused for dimensioning.Indicates how many % of the total available TCH capacity hasnot been used on average. If free TCH % approaches 0 the MSusers start to experience call blocking and/or slowing down ofGPRS throughput.



Known problems: Because the measurement period is usually 60 min., the valuecan not be used for spotting momentary problems but only thethe trend.

100- TCH usage % for CS - TCH usage % for PS= 100 - trf_83a-trf_84b

Figure 516. Free TCH %, S9PS (trf_86b)

Free TCH %, S10.5PS (trf_86c)

Use: Most useful on BTS level in connection with trf_83a, trf_84cand trf_160. The combined (PS+CS traffic) value trend can beused for dimensioning.Indicates how many % of the total available RCH capacity hasnot been used on average. If free TCH percentage approaches0 the MS users start to experience call blocking and/orslowing down of (E)GPRS throughput.

Known problems: Because the measurement period usually is 60 minutes, thevalue can not be used for spotting momentary problems butonly the trend.

100 - TCH usage % for CS - TCH usage % for GPRS - TCH usage % for EGPRS= 100 - trf_83a - trf_84b - trf_160

Unit: %

Figure 517. Free TCH %, S10.5PS (trf_86c)

Total TCH % for PS (trf_87b)

Use: Indicates how many % of the total available normal TRXTCH capacity has been allocated for the PS territory,including additional channels.If there are now upgrades or downgrades, this value should bequite close to the value of parameter CDEF. Some differencecomes from granularity when the CDEF value is converted totimeslots in BSC.The denominator does not contain extended TRXs becausethey are not GPRS capable and are not included in theconversion of CDEF to timeslots.

Capacity allocated for PS territory-------------------------------------total TCH capacity of normal TRXs

ava_16a



= 100* --------------- %ava_28+ava_16a

Figure 518. Total TCH % for PS (trf_87b)

Total TCH % for dedicated PS, S9PS (trf_88b)

Use: Indicates how many % of the total normal TRX TCH capacityhas been allocated for the dedicated PS territory. This valueshould be quite close to the value of the CDED parameter.Some difference comes from granularity when the CDEDvalue is converted to timeslots in BSC.The denominator does not contain extended TRXs becausethey are not GPRS capable and are not included in theconversion of CDED to timeslots.

Capacity allocated for dedicated PS territory---------------------------------------------

total TCH capacity of normal TRXs

ava_17a= 100* --------------- %

ava_28+ava_16a

Figure 519. Total TCH % for dedicated PS, S9PS (trf_88b)

Average total UL throughput per used timeslot, S9PS (trf_89]

Use: Indicates the total data rate per used timeslot. This figure isaffected by the coding scheme selected by the link adaptation.

Known problems: IGNOR_RLC_DATA_BL_UL_DUE_BSN is not only CS1(23 octets) but can also be CS2 (33). The share between CS1and CS2 has to be approximated.

All UL data (kilobit)---------------------------------- =time used for UL data transfer (sec)

( sum(rlc_data_blocks_ul_cs1 *23+ rlc_data_blocks_ul_cs2 *33+ rlc_mac_cntrl_blocks_ul *23+ BAD_FRAME_IND_UL_CS1*23+ BAD_FRAME_IND_UL_UNACK*23+ BAD_FRAME_IND_UL_CS2* 33)

+ ignor_rlc_data_bl_ul_due_bsn_CS1_aprx*23+ ignor_rlc_data_bl_ul_due_bsn_CS2_aprx *33

)*8/1000-----------------------------------------------sum(period_duration)*60* trf_78a

where

ignor_rlc_data_bl_ul_due_bsn_CS1_aprx =RLC CS1 blocks ignored due to incorrect BSN (missing counter approximated) =



sum(rlc_data_blocks_ul_cs1-rlc_data_blocks_UL_unack)-------------------------------------------------*sum(ignor_rlc_data_bl_ul_due_bsn)

sum(rlc_data_blocks_ul_cs1-rlc_data_blocks_UL_unack+rlc_data_blocks_ul_cs2)

ignor_rlc_data_bl_ul_due_bsn_CS2_aprx =RLC CS2 blocks ignored due to incorrect BSN (missing counter approximated)=

sum(rlc_data_blocks_ul_cs2)------------------------------------------------------*sum(ignor_rlc_data_bl_ul_due_bsn)

sum(rlc_data_blocks_ul_cs1-rlc_data_blocks_UL_unack+rlc_data_blocks_ul_cs2)


unit: kbps / tsl

Figure 520. Average total UL throughput per used timeslot, S9PS (trf_89)

Average total UL throughput per used TSL, S10PS (trf_89a)

Use: Indicates the total data rate per used timeslot. This figure isaffected by the coding scheme selected by the link adaptation.

Known problems: IGNOR_RLC_DATA_BL_UL_DUE_BSN is not only CS1(23 octets) but can also be CS2 (33). The share between CS1and CS2 has to be approximated.

All UL data (kilobits)------------------------------------ =time used for UL data transfer (sec)

sum(a.rlc_data_blocks_ul_cs1 *23+ a.rlc_data_blocks_ul_cs2 *33+ a.rlc_mac_cntrl_blocks_ul *23+ a.BAD_FRAME_IND_UL_CS1*23+ a.BAD_FRAME_IND_UL_UNACK*23+ a.BAD_FRAME_IND_UL_CS2* 33+ a.ignor_rlc_data_bl_ul_due_bsn_CS1_aprx*23+ a.ignor_rlc_data_bl_ul_due_bsn_CS2_aprx *33

..+ )*8/1000+8*(sum over MCS-1 (xx)*30+

sum over MCS-2 (xx)*36+sum over MCS-3 (xx)*45+sum over MCS-4 (xx)*52+sum over MCS-5 (xx)*63+sum over MCS-6 (xx)*81+sum over MCS-7 (xx/2)*123+sum over MCS-8 (xx/2)*147+sum over MCS-9 (xx/2)*159)/1000

-----------------------------------------------sum(period_duration)*60* trf_78c

where1)ignor_rlc_data_bl_ul_due_bsn_CS1_aprx =RLC CS1 blocks ignored due to incorrect BSN (missing counter approximated) =



sum(rlc_data_blocks_ul_cs1- rlc_data_blocks_UL_unack)

------------------------------- *sum(ignor_rlc_data_bl_ul_due_bsn)sum(rlc_data_blocks_ul_cs1



2)ignor_rlc_data_bl_ul_due_bsn_CS2_aprx =RLC CS2 blocks ignored due to incorrect BSN (missing counter approximated)=

sum(rlc_data_blocks_ul_cs2)------------------------------- *sum(ignor_rlc_data_bl_ul_due_bsn)sum(rlc_data_blocks_ul_cs1


Counters from table(s):a=p_nbsc_packet_control_unit,b=p_nbsc_coding_scheme

3)xx = EGPRS UL RLC blocks =(b.ul_rlc_blocks_in_ack_mode + b.retrans_rlc_data_blocks_ul+b.bad_rlc_valid_hdr_ul_unack + b.ul_rlc_blocks_in_unack_mode)

Unit: kbps/tsl

Figure 521. Average total UL throughput per used TSL, S10PS (trf_89a)

Average total DL throughput per used timeslot, S9PS (trf_90)

Use: Indicates the total data rate per used timeslot.Known problems: Counter rlc_mac_cntrl_blocks_dl also contains dummy

blocks until CD6.1.

All DL data (kbits)---------------------------------- =time used for DL data transfer (sec)

sum(rlc_data_blocks_dl_cs1 *23+ rlc_data_blocks_dl_cs2 *33+ rlc_mac_cntrl_blocks_dl *23+ RETRA_RLC_DATA_BLOCKS_DL_CS1*23+ RETRA_RLC_DATA_BLOCKS_DL_CS2* 33)*8/1000

-----------------------------------------------sum(rlc_data_blocks_dl_cs1

+ rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl+ RETRA_RLC_DATA_BLOCKS_DL_CS1+ RETRA_RLC_DATA_BLOCKS_DL_CS2)/50




Unit: kbit/sec/tsl

Figure 522. Average total DL throughput per used timeslot, S9PS (trf_90)

Average total DL throughput per used timeslot, S10PS (trf_90a)

Use: Indicates the total data rate per used timeslot.Known problems: Dummy blocks are included. They can be subtracted in S10.

All GPRS DL data + all EGPRS DL data (kbits)-------------------------------------------- =time used for DL data transfer (sec)

sum(a.rlc_data_blocks_dl_cs1 *23+ a.rlc_data_blocks_dl_cs2 *33+ a.rlc_mac_cntrl_blocks_dl *23+ a.RETRA_RLC_DATA_BLOCKS_DL_CS1*23+ a.RETRA_RLC_DATA_BLOCKS_DL_CS2* 33)*8/1000+(sum over MCS-1 (xx)*30+sum over MCS-2 (xx)*36+sum over MCS-3 (xx)*45+sum over MCS-4 (xx)*52+sum over MCS-5 (xx)*63+sum over MCS-6 (xx)*81+sum over MCS-7 (xx/2)*123+sum over MCS-8 (xx/2)*147+sum over MCS-9 (xx/2)*159)*8/1000

-----------------------------------------------sum(period_duration)*60* trf_79c

Wherexx =(b.dl_rlc_blocks_in_ack_mode+b.retrans_rlc_data_blocks_dl+b.dl_rlc_blocks_in_unack_mode)

Counters from table(s):a = p_nbsc_packet_control_unit,b = p_nbsc_coding_scheme

Unit: kbps/tsl

Figure 523. Average total DL throughput per used timeslot, S10PS (trf_90a)

SDCCH true seizures for call (trf_91)

sum(succ_seiz_term+succ_seiz_orig+sdcch_call_re_est+sdcch_emerg_call- succ_sdcch_sms_est- unsucc_sdcch_sms_est-succ_seiz_supplem_serv)


Figure 524. SDCCH true seizures for call (trf_91)



Average HSCSD subchannel traffic, S7HS (trf_92)

HSCSD total traffic - HSCS main channel traffic =trf_59-trf-60 =

sum(ave_busy_tch_hscsd_numer)/sum(ave_busy_tch_hscsd_denom)-sum(ave_hscsd_users_numer)/sum(ave_hscsd_users_denom)-


Figure 525. Average HSCSD subchannel traffic, S7HS (trf_92)

Average HSCSD subchannel traffic, S7HS (trf_92a)


trf_99 ; HSCSD total traffic, normal TRXs+ trf_100 ; HSCSD total traffic, extended TRXs- trf_104 ; HSCS main channel traffic, normal TRXs- trf_105 ; HSCS main channel traffic, extended TRXs

Unit: erlang


Figure 526. Average HSCSD subchannel traffic, S7HS (trf_92a)

Voice calls on SDCCH, S1 (trf_93)

sum(sdcch_emerg_call+succ_seiz_term+succ_seiz_orig-succ_sdcch_sms_est-unsucc_sdcch_sms_est)


Figure 527. Voice calls on SDCCH, S1 (trf_93)

TCH traffic, S1 (trf_94)

Use: Possible to use on the TRX level.

Call time / period duration =

sum(freq_ul_qual0+ freq_ul_qual1+ freq_ul_qual2+ freq_ul_qual3+ freq_ul_qual4+ freq_ul_qual5



+ freq_ul_qual6+ freq_ul_qual7)/60

0.48 * ------------------------sum(period_duration)

Unit: Erlang


Figure 528. TCH traffic, S1 (trf_94)

GPRS traffic sum, S9PS (trf_95a)

Use: Indicates the amount of resources (timeslots) the GPRS trafficdata consumes during the period on average. This informationis useful for example in forecasting the need for capacityextension.

Known problems: Timeslot usage caused by DL TBF release delay is notincluded and this makes the value seem optimistic.

Time used to transmit RLC blocks---------------------------------- =time available

sum(max of (rlc_data_blocks_ul_cs1+ rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul+ BAD_FRAME_IND_UL_CS1+ BAD_FRAME_IND_UL_CS2+ BAD_FRAME_IND_UL_UNACK+ IGNOR_RLC_DATA_BL_UL_DUE_BSN,

rlc_data_blocks_dl_cs1+ rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl+ RETRA_RLC_DATA_BLOCKS_DL_CS1+ RETRA_RLC_DATA_BLOCKS_DL_CS2))

/50/3600


Unit: erlang hour

Figure 529. GPRS traffic sum, S9PS (trf_95a)



Note

GPRS territory utilisation, S9PS (trf_96a)

Use: Used on BTS level. It Indicates how big a part of the GPRSterritory has been used. If utilisation % is high, an increase ofCDEF parameter setting should be considered.

The denominator varies depending on the traffic situation (downgrade ofterritory) and therefore this KPI has no linear correlation with the traffic. Seeava_15

100*(RLC blocks transmitted / (block transmission capacity) % =

Data blocks transmitted # greater one chosen, DL or UL100* ------------------------------------------------------------ % =


sum(max of (

rlc_data_blocks_ul_cs1+ rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul+ BAD_FRAME_IND_UL_CS1+ BAD_FRAME_IND_UL_CS2+ BAD_FRAME_IND_UL_UNACK+ IGNOR_RLC_DATA_BL_UL_DUE_BSN,

rlc_data_blocks_dl_cs1+ rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl+ RETRA_RLC_DATA_BLOCKS_DL_CS1+ RETRA_RLC_DATA_BLOCKS_DL_CS2))

100*------------------------------------------------------- %ava_16a *sum(a.period_duration*60)*50

Counters from table(s):a= p_nbsc_res_availb= p_nbsc_packet_control_unit

Figure 530. GPRS territory utilisation, S9PS (trf_96a)

PS territory utilisation, S10.5PS (trf_96b)

Use: Used on BTS level. Indicates how big a portion of the PSterritory has been used. If the utilisation percentage is high,increasing the CDEF parameter setting should be considered.





Known problems: Dummy blocks on DL make this PI show too high a value(fixed in CD6.0: RLC_MAC_cntrl_blocks_DL does notcontain dummy blocks anymore).1) If there are very few timeslots in the GPRS territory, thisKPI can show a high value even if there is only one activeuser.2) The denominator is slightly incorrect if extended TRXswere used.

100*(RLC blocks transmitted / (block transmission capacity) % =

Data blocks transmitted # greater one chosen, DL or UL100* ------------------------------------------------------------- % =


sum(max of (

rlc_data_blocks_ul_cs1+ rlc_data_blocks_ul_cs2+ rlc_mac_cntrl_blocks_ul+ BAD_FRAME_IND_UL_CS1+ BAD_FRAME_IND_UL_CS2+ BAD_FRAME_IND_UL_UNACK+ IGNOR_RLC_DATA_BL_UL_DUE_BSN,+(

sum over MCS-1 (xx)+sum over MCS-2 (xx)+sum over MCS-3 (xx)+sum over MCS-4 (xx)+sum over MCS-5 (xx)+sum over MCS-6 (xx)+sum over MCS-7 (xx/2)+sum over MCS-8 (xx/2)+sum over MCS-9 (xx/2)),

rlc_data_blocks_dl_cs1+ rlc_data_blocks_dl_cs2+ rlc_mac_cntrl_blocks_dl+ RETRA_RLC_DATA_BLOCKS_DL_CS1+ RETRA_RLC_DATA_BLOCKS_DL_CS2)+(

sum over MCS-1 (yy)+sum over MCS-2 (yy)+sum over MCS-3 (yy)+sum over MCS-4 (yy)+sum over MCS-5 (yy)+sum over MCS-6 (yy)+sum over MCS-7 (yy/2)+sum over MCS-8 (yy/2)+sum over MCS-9 (yy/2)

))100*------------------------------------------ %

ava_16a *sum(a.period_duration*60)*50

Where xx=c.(UL_RLC_BLOCKS_IN_ACK_MODE

+RETRANS_RLC_DATA_BLOCKS_UL+BAD_RLCVALID_HDR_UL_UNACK+UL_RLC_BLOCKS_IN_UNACK_MODE)

yy=c.(DL_RLC_BLOCKS_IN_ACK_MODE+ RETRANS_RLC_DATA_BLOCKS_DL



+ DL_RLC_BLOCKS_IN_UNACK_MODE)

Counters from table(s):a= p_nbsc_res_availb= p_nbsc_packet_control_unitc= p_nbsc_coding_scheme

Figure 531. PS territory utilisation, S10.5PS (trf_96b)

Average CS traffic, normal TRXs, erlang, S2 (trf_97)

Use: This is a speech (circuit switched) traffic indicator. Speechtraffic is a basic indicator needed to show how much TCHcapacity is used. When traffic increases without an increase incapacity, the probability of blocking increases. Therelationship between traffic, capacity and blocking for speechtraffic is described in the formula known as Erlang B.This KPI includes all types of CS traffic (single TCH,HSCSD) on normal TRXs.

sum(decode(trx_type,0,ave_busy_tch))-----------------------------------sum(decode(trx_type,0,res_av_denom14))


Unit: erlang

Figure 532. Average CS traffic, normal TRXs, erlang, S2 (trf_97)

Average CS traffic, extended TRXs S2 (trf_98)

Use: This is a speech (circuit switched) traffic indicator. Speechtraffic is a basic indicator needed to show how much TCHcapacity is used. When the traffic increases without increasein capacity, the probability of blocking grows.The relationship between traffic, capacity and blocking inspeech traffic is described in the formula known as Erlang B.This KPI includes all types of CS traffic (single TCH,HSCSD) on extended TRXs

sum(decode(trx_type,1,ave_busy_tch))-----------------------------------sum(decode(trx_type,1,res_av_denom14))




Unit: erlang

Figure 533. Average CS traffic, extended TRXs S2 (trf_98)

Average HSCSD traffic, normal TRXs, S7HS (trf_99)

Note: HSCSD uses FR.

sum(decode(trx_type,0,ave_busy_tch_hscsd_numer))------------------------------------------------sum(decode(trx_type,0,ave_busy_tch_hscsd_denom))


Unit: erlang

Figure 534. Average HSCSD traffic, normal TRXs, S7HS (trf_99)

Average HSCSD traffic, extended TRXs, S7HS (trf_100)


sum(decode(trx_type,1,ave_busy_tch_hscsd_numer))------------------------------------------------sum(decode(trx_type,1,ave_busy_tch_hscsd_denom))


Unit: erlang

Figure 535. Average HSCSD traffic, extended TRXs, S7HS (trf_100)

Average HTCH traffic, S7HS (trf_101)

Use: Total of speech (circuit switched) single timeslot half ratetraffic over normal and extended TRXs

Note: On BTS level. For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs.

trf_102 ; HTCH traffic, normal TRXs+ trf_103 ; HTCH traffic, extended TRXs

Unit: erlang

Figure 536. Average HTCH traffic, S7HS (trf_101)



Average HTCH traffic, normal TRXs, S7HS (trf_102)

Use: Total of speech (circuit switched) in single timeslot half ratetraffic over normal TRXs

Avg(decode,trx_type,0,ave_tch_busy_half))

Unit: erlang

Figure 537. Average HTCH traffic, normal TRXs, S7HS (trf_102)

Average HTCH traffic, extended TRXs, S7HS (trf_103)

Use: Total of speech (circuit switched) in single timeslot half ratetraffic over extended TRXs

nvl(Avg(decode,trx_type,1,ave_tch_busy_half)),0)

Unit: erlang

Figure 538. Average HTCH traffic, extended TRXs, S7HS (trf_103)

Average HSCSD main channel traffic, normal TRXs, S7HS (trf_104)


sum(decode(trx_type,0,ave_hscsd_users_numer))--------------------------------------------sum(decode(trx_type,0,ave_hscsd_users_denom))


Figure 539. Average HSCSD main channel traffic, normal TRXs, S7HS (trf_104)

Average HSCSD main channel traffic, extended TRXs, S7HS (trf_105)


sum(decode(trx_type,0,ave_hscsd_users_numer))--------------------------------------------sum(decode(trx_type,0,ave_hscsd_users_denom))


Unit: erlang

Figure 540. Average HSCSD main channel traffic, extended TRXs, S7HS(trf_105)



Average FTCH single traffic, S7HS (trf_106)


trf_107 ; FTCH single traffic on normal TRXs+ trf_108 ; FTCH single traffic on extended TRXs

Unit: erlang

Figure 541. Average FTCH single traffic, S7HS (trf_106)

Average FTCH single traffic, normal TRXs, S7HS (trf_107)

trf_97 ; all CS traffic on normal TRXs-trf_102 ; single HR traffic on normal TRXs-trf_99 ; all HSCSD (FR) traffic on normal TRXs

Unit: erlang

Figure 542. Average FTCH single traffic, normal TRXs, S7HS (trf_107)

Average FTCH single traffic, extended TRXs, S7HS (trf_108)

trf_98 ; all CS traffic on extended TRXs-trf_103 ; single HR traffic on extended TRXs-trf_100 ; all HSCSD (FR) traffic on extended TRXs

Unit: erlang

Figure 543. Average FTCH single traffic, extended TRXs, S7HS (trf_108)

Peak busy TCH on normal TRXs (trf_109)

Use: This PI is an important traffic load indicator on the cell level.By following this and reacting proactively, blocking can beavoided in cells in which the traffic grows smoothly.

max(decode(trx_type,0,peak_busy_tch))


Figure 544. Peak busy TCH on normal TRXs (trf_109)



Peak busy TCH on normal TRXs (trf_110)

Use: This PI is an important traffic load indicator on the cell level.By following this and reacting proactively, blocking can beavoided in cells in which the traffic grows smoothly.

max(decode(trx_type,1,peak_busy_tch))


Figure 545. Peak busy TCH on normal TRXs (trf_110)

Normal TCH usage % for CS (trf_111)

Use: Indicates how many % of the total normal TCH capacityavailable has been used for CS traffic on average.

trf_97= 100* ----------------------- %

ava_28+ava_16a-ava_17a

Figure 546. Normal TCH usage % for CS (trf_111)

Normal TCH usage % for CS (trf_112)

Use: Indicates how many % of the total normal TCH capacityavailable has been used for CS traffic on average.

trf_98= 100* ------------ %

ava_29

Figure 547. Normal TCH usage % for CS (trf_112)

CS call samples, non-AMR call (trf_113)

Use: Indicates how many call samples (sampling interval 480 ms)of non-AMR calls have been detected.

sum(nvl(FREQ_UL_QUAL0,0)+ + nvl(FREQ_UL_QUAL7,0)

-sum(nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_UL_RXQUAL_7,0)

+ nvl(AMR_FR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_3_UL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_4_UL_RXQUAL_7,0)

nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_UL_RXQUAL_6,0)



))

Figure 548. CS call samples, non-AMR call (trf_113)

CS call samples, AMR call (trf_114)

Use: Indicates TCH use (sampling interval 480 ms) for AMR calls.

sum(nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_UL_RXQUAL_7,0)


nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_UL_RXQUAL_7,0)

+ nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_UL_RXQUAL_6,0))

Figure 549. CS call samples, AMR call (trf_114)

TCH traffic time, non-AMR calls (trf_115)

Use: Indicates TCH use (sampling interval 480 ms) for non-AMRcalls.


sum(nvl(FREQ_UL_QUAL0,0)+ + nvl(FREQ_UL_QUAL7,0)

-sum(nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_UL_RXQUAL_7,0)


nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_UL_RXQUAL_6,0)

)*0,48/3600

Unit: erlang hours

Figure 550. TCH traffic time, non-AMR calls (trf_115)

TCH traffic time, AMR calls (trf_116)

Use: Indicates TCH use (sampling interval 480 ms) for AMR calls.




trf_114*0,48/36000

Unit: erlang hour

Figure 551. TCH traffic time, AMR calls (trf_116)

TCH traffic time, FR AMR calls (trf_117)

Use: Indicates TCH use (sampling interval 480 ms) for full rateAMR.


,sum(nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_UL_RXQUAL_7,0)


)*0,48/3600

Unit: erlang hours

Figure 552. TCH traffic time, FR AMR calls (trf_117)

TCH traffic time, HR AMR calls (trf_118)


Sum(nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_UL_RXQUAL_7,0)

+ nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_UL_RXQUAL_6,0))

*0,48/3600

Unit: erlang hours

Figure 553. TCH traffic time, HR AMR calls (trf_118)



TCH traffic time, all calls (trf_119)


sum(nvl(FREQ_UL_QUAL0,0)+ .. + nvl(FREQ_UL_QUAL7,0))*0,48/3600

Unit: erlang hours

Figure 554. TCH traffic time, all calls (trf_119)

TCH traffic share of non-AMR calls (trf_120)

100*Trf_115/Trf_119

Unit: erlang hours

Figure 555. TCH traffic share of non-AMR calls (trf_120)

TCH traffic share of FR AMR calls (trf_121)

100*Trf_117/Trf_119

Unit: erlang hours

Figure 556. TCH traffic share of FR AMR calls (trf_121)

TCH traffic share of HR AMR calls (trf_122)

100*Trf_118/Trf_119

Unit: erlang hours

Figure 557. TCH traffic share of HR AMR calls (trf_122)

Average effective UL timeslot throughput per TBF, S10PS (trf_123)

Use; Indicates the net data rate per used timeslot and per TBF. Thelower, the value the more loaded is the GPRS territory and theless service the MS users receive.The numerator does not contain the RLC header bytes neitherdoes the MAC header because the aim is to count data volumefrom the user’s point of view as close as possible.



Known problems: 1) The numerator of trf_72d is not yet pure user data but asclose to that as we can see from BSC counters.2) See problems of the denominator.

UL payload data (kilobit) / UL time for data transfer (sec)------------------------------------------------------------- =

Avg UL TBF per tsl

trf_72d= ----------

tbf_37b


unit: Kbps / tsl/TBF

Figure 558. Average effective UL timeslot throughput per TBF, S10PS (trf_123)

Average effective DL timeslot throughput per TBF, S10PS (trf_124)

Use; Indicates the net data rate per used timeslot and per TBF. Thelower the value, the more loaded is the GPRS territory and theless service the MS users receive.The numerator does not contain the RLC header bytes neitherthe MAC header because the aim is to count the data volumefrom the user point of view as close as possible.

Known problems: 1) The numerator of trf_73d is not yet pure user data but asclose as we can see from BSC counters.2) Retransmitted blocks due to other reasons than NACK arenot counted in any of the RLC counters. In DL direction theseretransmissions occur when TBF release is delayed.3) If there is, for example, only one TBF on a timeslot, someRLC blocks can be retransmitted before an ACK is received.These blocks are not counted in any of the RLC counters.4) Counter rlc_mac_cntrl_blocks_dl also containsdummy blocks until CD.6.1.5) See problems of the denominator.

DL payload data (kilobit) / DL time for data transfer (sec)------------------------------------------------------------- =

Avg DL TBF per tsl

trf_73d= --------

tbf_38b


unit: Kbps / tsl / TB

Figure 559. Average effective DL timeslot throughput per TBF, S10PS (trf_124)



MS specific flowrate (trf_125)

Use; This is the flowrate of LLC PDUs. It can be counted persegment and priority class.

8/1000 * ave_ms_bssgp_flow_rate_sum-----------------------------------ave_ms_bssgp_flow_rate_den

Counters from table(s):p_nbsc_qosUnit: kbit/sec

Figure 560. MS specific flowrate (trf_125)

Total RLC payload data (Kbytes), MCS-n, S10.5PS (trf_131)

Sum over MCS-n (UL_RLC_BLOCKS_IN_ACK_MODE +UL_RLC_BLOCKS_IN_UNACK_MODE +DL_RLC_BLOCKS_IN_ACK_MODE +DL_RLC_BLOCKS_IN_UNACK_MODE) * nn / 1024

where n can be from 1 to 9 and nn is the multiplier for each Coding Scheme,i.e. RLC Data Block payload in bytes.nn for each MCS:MCS-122MCS-228MCS-337MCS-444MCS-556MCS-674MCS-756MCS-868MCS-974)


Unit: Kbytes

Figure 561. Total RLC payload data (Kbytes), MCS-n, S10.5PS (trf_131)

UL RLC data MCS-n, S10.5PS (trf_140)

sum over MCS-n (xx) *nn / 1024

(where xx =



UL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_UL +BAD_RLC_VALID_HDR_UL_UNACK +UL_RLC_BLOCKS_IN_UNACK_MODE)

where n can be from 1 to 9.

(where nn is multiplier for each Coding Scheme, i.e. RLC Data Block payload in bytesnn for each MCS:MCS-1 22MCS-2 28MCS-3 37MCS-4 44MCS-5 56MCS-6 74MCS-7 56MCS-8 68MCS-9 74)


Figure 562. UL RLC data MCS-n, S10.5PS (trf_140)

DL RLC data MCS-n, S10.5PS (trf_141)

Use: DL RLC dData MCS1

sum over MCS-n (xx) * nn / 1024

where xx =(DL_RLC_BLOCKS_IN_ACK_MODE +RETRANS_RLC_DATA_BLOCKS_DL +DL_RLC_BLOCKS_IN_UNACK_MODE)

where n can be from 1 to 9.

(where nn is multiplier for each Coding Scheme, i.e. RLC Data Block payload in bytesnn for each MCS:MCS-1 22MCS-2 28MCS-3 37MCS-4 44MCS-5 56MCS-6 74MCS-7 56MCS-8 68MCS-9 74)


Unit: kbytes

Figure 563. DL RLC data MCS-n, S10.5PS (trf_141)



Normal TCH usage % for EGPRS, S10.5PS (trf_160)

Use: Indicates how many % of the total available normal TCHcapacity has been used for EGPRS traffic on average. Usedfor trend analysis.

Capacity used by EGPRS traffic------------------------------Total TCH normal capacity

trf_158= 100 * ---------------- %

ava_28 + ava_16aUnit: %

Figure 564. Normal TCH usage % for EGPRS, S10.5PS (trf_160)

UL EGPRS traffic, S10.5PS (trf_161)

Use: Indicates the amount of resources (timeslots) that the GPRStraffic data consumes on average during the period. Thisinformation is useful, for example, in forecasting the need toextend capacity.

Known problems: See trf_78c.

Actual UL data throughput (blocks)----------------------------------------------------------------- =Number of blocks equivalent to 1 tsl full use in each BTS of area

sum over MSC1-6 of( ul_rlc_blocks_in_ack_mode+ retrans_rlc_data_blocks_ul+ BAD_RLC_VALID_HDR_UL_ACK+ bad_rlc_valid_hdr_ul_unack+ ul_rlc_blocks_in_unack_mode)

+ sum over MSC7-9 of( ul_rlc_blocks_in_ack_mode+ retrans_rlc_data_blocks_ul+ BAD_RLC_VALID_HDR_UL_ACK+ bad_rlc_valid_hdr_ul_unack+ ul_rlc_blocks_in_unack_mode)/2

----------------------------------------------------------------------------------------

sum(period_duration*60)*50 ; 50 blocks /sec /tsl


Unit: tsl (or erlang)

Figure 565. UL EGPRS traffic, S10.5PS (trf_161)



DL EGPRS traffic, S10.5PS (trf_162)

Use: Indicates the amount of resources (timeslots) the DL EGPRStraffic data consumes. This information is useful, for example,in forecasting the need to extend capacity.

Actual DL data throughput (blocks)----------------------------------------------------------------- =Number of blocks equivalent to 1 tsl full use in each BTS of area

sum over msc1…6( dl_rlc_blocks_in_ack_mode+ retrans_rlc_data_blocks_dl+ dl_rlc_blocks_in_unack_mode)

+ sum over msc7…9 of( dl_rlc_blocks_in_ack_mode+ retrans_rlc_data_blocks_dl+ dl_rlc_blocks_in_unack_mode)/2

----------------------------------------------------sum(period_duration*60)*50 ;50 blocks /sec /tsl

Counters from table(s):p_nsbc_coding_scheme

Figure 566. DL EGPRS traffic, S10.5PS (trf_162)

Total EGPRS RLC data, S9PS (trf_167)

Use: Indicates the total amount of data (both ack and unack modes)transmitted as CS1 or CS2 blocks.Used in UL or DL. MAC blocks and RLC header bytes areexcluded in order to get as close as possible to the payloaddata.

((sum over MCS-1 (xx)* 22+sum over MCS-2 (xx)* 28+sum over MCS-3 (xx)* 37+sum over MCS-4 (xx)* 44+sum over MCS-5 (xx)* 56+sum over MCS-6 (xx)* 74+sum over MCS-7 (xx/2)*112+sum over MCS-8 (xx/2)*136+sum over MCS-9 (xx/2)*148)))+

(sum over MCS-1 (yy)* 22+sum over MCS-2 (yy)* 28+sum over MCS-3 (yy)* 37+sum over MCS-4 (yy)* 44+sum over MCS-5 (yy)* 56+sum over MCS-6 (yy)* 74+sum over MCS-7 (yy/2)*112+sum over MCS-8 (yy/2)*136+sum over MCS-9 (yy/2)*148))

/1024

Where xx=(UL_RLC_BLOCKS_IN_ACK_MODE + UL_RLC_BLOCKS_IN_UNACK_MODE)

Where yy= DL_RLC_BLOCKS_IN_ACK_MODE + DL_RLC_BLOCKS_IN_UNACK_MODE




Unit: Kbytes

Figure 567. Total EGPRS RLC data, S9PS (trf_167)

Average SDCCH traffic, Area, S1 (trf_168)


SDCCH used time / period duration =

(sum(a.ave_sdcch_hold_tim) / (avg(a.res_av_denom16)* 100 * count( distinct period_start_time))

)* sum(b.sdcch_assign + b.sdcch_ho_seiz + b.tch_seiz_due_sdcch_con)------------------------------------------------------------------sum(a.period_duration*60)

Counters from table(s):a = p_nbsc_res_availb = p_nbsc_trafficUnit: Erlang

Figure 568. Average SDCCH traffic, Area, S1 (trf_168)

2.27 Traffic directions

2.27.1 Mobile originated calls (moc)

SDCCH seizures for MO calls, S2 (moc_1)

Known problems: Includes supplementary services such as call divert.Includes SMS.

sum(succ_seiz_orig)


Figure 569. SDCCH seizures for MO calls, S2 (moc_1)

Successful MO speech calls, S3 (moc_2)

Note: Triggered when a call is cleared. Excludes setup failures,TCH drops and TCH busy (congestion) cases.



Note

Known problems: The measurement is on the BSC level.

sum(nbr_of_calls)where counter_id = 44

Counters from table(s):p_nbsc_cc_pm

Figure 570. Successful MO speech calls, S3 (moc_2)

Successful MO data calls, S3 (moc_3)

Note: See moc_2.Known problems: The measurement is on the BSC level.



Figure 571. Successful MO data calls, S3 (moc_3)

MO call success ratio, S6 (moc_4)

Note: See moc_2.Known problems: The measurement is on the BSC level.

MO call attempts are counted when MOCs are found on SDCCH. The numeratorexcludes setup failures, TCH drops and TCH busy (congestion) cases.

sum(nbr_of_calls) where counter_id = 44 /* MO call completed */100 * -----------------------------------------------------------------------

sum(nbr_of_calls) where counter_id = 38 /* MO call attempt */


Figure 572. MO call success ratio, S6 (moc_4)

MO speech call attempts, S3 (moc_5)

Note: Triggered when a call is cleared. Excludes setup failures,TCH drops and TCH busy (congestion) cases.

Known problems: The measurement is on the BSC level.

sum(nbr_of_calls)



where counter_id = 38


Figure 573. MO speech call attempts, S3 (moc_5)

MO call bids, S2 (moc_6)

Known problems: Includes supplementary services such as call divert.Includes SMS.

sum(succ_seiz_orig+tch_moc)


Figure 574. MO call bids, S2 (moc_6)

2.27.2 Mobile terminated calls (mtc)

SDCCH seizures for MT calls, S2 (mtc_1)

Known problems: Includes SMS. See also moc_2.

sum(succ_seiz_term)


Figure 575. SDCCH seizures for MT calls, S2 (mtc_1)

Successful MT speech calls (mtc_2)

Note: See moc_2.Known problems: See moc_2.



Figure 576. Successful MT speech calls (mtc_2)

Successful MT data calls, S3 (mtc_3)






Figure 577. Successful MT data calls, S3 (mtc_3)

MT call success ratio, S6 (mtc_4)

Note: MT call attempts are counted when MTCs are found onSDCCH. The numerator excludes setup failures, TCH dropsand TCH busy (congestion) cases.

Known problems: See moc_2.

sum(nbr_of_calls) where counter_id = 43 /* MT call completed */100 * -----------------------------------------------------------------------

sum(nbr_of_calls) where counter_id = 37 /* MT call attempt */


Figure 578. MT call success ratio, S6 (mtc_4)

MT speech call attempts (mtc_5)




Figure 579. MT speech call attempts (mtc_5)

MT call attempts, S2 (mtc_6)

Use: Total number of calls bids with establishment cause ’MT’.Known problems: Includes SMS.

sum(succ_seiz_term+tch_mtc)


Figure 580. MT call attempts, S2 (mtc_6)



2.28 Paging (pgn)

Number of paging messages sent, S2 (pgn_1)

Known problems: The number of repagings cannot be separated.

sum(paging_msg_sent)


Figure 581. Number of paging messages sent, S2 (pgn_1)

Paging buffer size average, S1 (pgn_2)

Use: To have an indication on how close to problems the BTS hasbeen.

Known problems: It is difficult to say when the problems start. Even if thecounter 3018 does not yet show the 0 value, there may havebeen the situation in one or some of the buffers that thecapacity has run out.

avg(min_paging_buf)


Parameters related:Number of Blocks for AGCH (AG): e.g. = 2Number of MultiFrames (MFR): e.g. = 6

Formulas related:Nbr of paging groups = (3-AG)*MFR ;if combined control channelNbr of paging groups = (9-AG)*MFR ;if non-combined control channel

Paging_Buffer_Size = free buffers (max 8) * Nbr of paging groups

Min Paging Buffer (counter 3018) = min(Paging_Buffer_Space). = min(Paging_Buffer_Size/2)

Figure 582. Paging buffer size average, S1 (pgn_2)

Paging Buffer Space is sent by BTS in the CCH_Load_Ind message to a BSCevery 30 s. A BSC sends current paging load as Paging_Buffer_Size to astatistical unit. The minimum value of this is recorded as counter 3018. If MinPaging Buffer (counter 3018) equals to zero, paging blocking has occurred.

Average paging buffer space, S1 (pgn_3)

Use: Average remaining free space for paging commands in GSMbuffer area (part of GPRS buffer area). When there are nopagings, this PI shows the capacity of the buffer.



Known problems: Incorrect if CCCH load ind interval has changed during theobservation period.

avg(ave_pch_load/res_acc_denom2)


Figure 583. Average paging buffer space, S1 (pgn_3)

Average free space of paging GSM buffer area, S1 (pgn_3a)

Use: Average remaining free space for paging commands in GSMbuffer area (part of GPRS buffer area). When there are nopagings this PI shows the capacity of the buffer.

sum(ave_pch_load)/sum(res_acc_denom2)


Figure 584. Average free space of paging GSM buffer area, S1 (pgn_3a)

Paging success ratio, S1 (pgn_4)

Known problems: Due to the very dynamic behaviour it seems that this formulais not useful.

sum over all BTS in LA (succ_seiz_term + tch_mtc)100* ------------------------------------------------------- %

sum over LA(paging_msg_sent) / sum over LA (count of BTS)


Figure 585. Paging success ratio, S1 (pgn_4)

Average paging buffer air interface occupancy, S7 (pgn_5)

sum(ave_paging_buffer_capa_numer)----------------------------------sum(ave_paging_buffer_capa_denom)


Figure 586. Average paging buffer air interface occupancy, S7 (pgn_5)



Average paging buffer Gb occupancy, S7PS (pgn_6)

sum(ave_paging_gb_buf_sum)---------------------------sum(ave_paging_gb_buf_den)


Figure 587. Average paging buffer Gb occupancy, S7PS (pgn_6)

Average air interface DRX buffer load, due to paging, S7 (pgn_7)

Use: The DRX buffer handles messages that are sent in the DRXcycle, i.e. pagings and DRX access grants. This counterdescribes the DRX buffer load resulting from pagingmessages.

sum(ave_paging_load_air_sum)---------------------------sum(ave_paging_load_air_den)


Figure 588. Average air interface DRX buffer load, due to paging, S7 (pgn_7)

Average air interface DRX buffer load, due to DRX AG, S7 (pgn_8)

Use: The DRX buffer handles messages that are sent in DRX cycle,i.e. pagings and DRX access grants. This counter describesDRX buffer load resulting from DRX AG messages (e.g. anImm.Ass.for DL TBF establishment.)

sum(ave_drx_agch_load_air_sum)-------------------------------sum(ave_drx_agch_load_air_den)


Figure 589. Average air interface DRX buffer load, due to DRX AG, S7 (pgn_8)

Average air interface non-DRX buffer load due to AG, S7 (pgn_9)

Use: The non-DRX buffer handles messages that are sentimmediately. This counter describes the non-DRX buffer loadresulting from non-DRX (i.e. immediate) access grants suchas CS Imm.Ass. and UL TBF (Imm.Ass. sent as an answer toRACH).

sum(ave_non_drx_agch_load_air_sum)



-----------------------------------sum(ave_non_drx_agch_load_air_den)


Figure 590. Average air interface non-DRX buffer load due to AG, S7 (pgn_9)

Average free space of paging GPRS buffer area, S9 (pgn_10)

Use: Average remaining free space for paging commands in theGSM buffer area. When there are no pagings, this PI showsthe capacity of the GPRS buffer area.

sum(AVE_PCH_GB_LOAD_ON_CCCH_SUM)----------------------------------sum(AVE_PCH_GB_LOAD_ON_CCCH_DEN)


Figure 591. Average free space of paging GPRS buffer area, S9 (pgn_10)

2.29 Short message service (sms)

SMS establishment failure % (sms_1)

100* unsuccessful SMS establishments / all SMS establishments =

sum(unsucc_TCH_sms_est+unsucc_SDCCH_sms_est)100* ------------------------------------------------------------------ %

sum(succ_TCH_sms_est+unsucc_TCH_sms_est+succ_SDCCH_sms_est+unsucc_SDCCH_sms_est)


Figure 592. SMS establishment failure % (sms_1)

SMS TCH establishment failure % (sms_2)

sum(unsucc_TCH_sms_est)100 * ------------------------------------------- %

sum(succ_TCH_sms_est+unsucc_TCH_sms_est)


Figure 593. SMS TCH establishment failure % (sms_2)



SMS SDCCH establishment failure % (sms_3)

Use: MOC: Instead of the sending SETUP message, the MS startsSMS by sending SABM with SAPI 3 to BTS, and a newestablishment indication is generated.MTC: Instead of the sending SETUP message, the MSC startsSMS by sending a CP DATA message to BSC and BSC sendsan ESTABLISH REQUEST to BTS, then MS answers by theUA message, and the ESTABLISH CONFIRM message isgenerated.SMS fails if the message data is corrupted, timer expires whenwaiting for an establishment confirmation, or if an errorindication or release indication is received.

sum(unsucc_sdcch_sms_est)100 * -------------------------------------------- %

sum(succ_sdcch_sms_est+unsucc_sdcch_sms_est)


Figure 594. SMS SDCCH establishment failure % (sms_3)

SMS establishment attempts (sms_4)

sum(succ_tch_sms_est+unsucc_tch_sms_est+succ_sdcch_sms_est+unsucc_sdcch_sms_est)


Figure 595. SMS establishment attempts (sms_4)

SMS SDCCH establishment attempts (sms_5)

sum(succ_sdcch_sms_est+unsucc_sdcch_sms_est)


Figure 596. SMS SDCCH establishment attempts (sms_5)

SMS TCH establishment attempts (sms_6)

sum(succ_TCH_sms_est+unsucc_TCH_sms_est)




Figure 597. SMS TCH establishment attempts (sms_6)

2.30 Directed retry (dr)

DR, outgoing attempts, S3 (dr_1)

sum(msc_o_sdcch_tch_at + bsc_o_sdcch_tch_at)


Figure 598. DR, outgoing attempts, S3 (dr_1)

DR attempts, S3 (dr_1a)

Use: Includes all DR cases (to another cell and intra-cell).

sum(cause_dir_retry)


Figure 599. DR attempts, S3 (dr_1a)

DR, incoming attempts, S3 (dr_2)

sum(msc_i_sdcch_tch_at + bsc_i_sdcch_tch_at)


Figure 600. DR, incoming attempts, S3 (dr_2)

DR, outgoing success to another cell, S3 (dr_3)

sum(msc_o_sdcch_tch + bsc_o_sdcch_tch)


Figure 601. DR, outgoing success to another cell, S3 (dr_3)



DR, incoming success from another cell, S3 (dr_4)

sum(msc_i_sdcch_tch + bsc_i_sdcch_tch)


Figure 602. DR, incoming success from another cell, S3 (dr_4)

DR, intra-cell successful HO, S3 (dr_5)

Use: Triggered by• S6 feature ’TCH assignment to super-reuse in IUO’• S7 feature ’Direct access to super-reuse TRX’

sum(cell_sdcch_tch)


Figure 603. DR, intra-cell successful HO, S3 (dr_5)

% of new calls successfully handed over to another cell by DR, S3 (dr_6)

sum(msc_o_sdcch_tch + bsc_o_sdcch_tch)100 * --------------------------------------

sum(tch_call_req)


Figure 604. % of new calls successfully handed over to another cell by DR, S3(dr_6)

DR, outgoing to another cell, failed, S3 (dr_7)

sum(msc_o_sdcch_tch_at + bsc_o_sdcch_tch_at- msc_o_sdcch_tch + bsc_o_sdcch_tch)


Figure 605. DR, outgoing to another cell, failed, S3 (dr_7)



DR, intra-cell failed, S3 (dr_8)

Use: Triggered by• S6 feature ’TCH assignment to super-reuse in IUO’• S7 feature ’Direct access to super-reuse TRX’

sum(cell_sdcch_tch_at- cell_sdcch_tch)


Figure 606. DR, intra-cell failed, S3 (dr_8)

2.31 Availability (ava)

TCH availability %, S4 (ava_1a)

Use: Failures (downtime) of TRXs cause loss of TCH and affectthis KPI.

Known problems: 1)If TRXs are locked and BTSs and BCFs are unlocked, theTCHs appear as unavailable. This means that both the systemand the user can affect this KPI and make it less useful.2)This PI does not take HTCH into consideration.

available TCH100 * ------------------ %

all TCH

sum(ave_avail_full_TCH/res_av_denom2)=100 * ------------------------------------------------------------ %

sum(ave_avail_full_TCH/res_av_denom2)+sum(ave_non_avail_TCH)


Figure 607. TCH availability %, S4 (ava_1a)

TCH availability %, S9 (ava_1c)


Known problems: 1) If TRXs are locked and BTSs and BCFs are unlocked, theTCHs appear as unavailable. This means that both the systemand the user can affect this KPI and make it less useful.2) The formula leaves out the timeslots reserved for GPRS.

available TCH100 * ------------------ %

all TCH

sum(ave_avail_TCH_sum/ave_avail_TCH_den)



=100 * --------------------------------------------------------------- %sum(ave_avail_TCH_sum/ave_avail_TCH_den)+sum(ave_non_avail_TCH)


Figure 608. TCH availability %, S9 (ava_1c)

TCH availability %, S9 (ava_1d)


Known problems: If TRXs are locked and BTSs and BCFs are unlocked, theTCHs appear as unavailable. This means that both the systemand the user can affect this KPI and make it less useful.

Note: This KPI has to be counted separately for extended andnormal area. Trx type: 0 = normal, 1 = extended.

available TCH100 * ------------------------- %

all TCH (traffic and GPRS)

sum(ave_avail_TCH_sum/ave_avail_TCH_den+ ave_GPRS_channels_sum/ave_GPRS_channels_den)

=100 * ---------------------------------------------------------------- %sum(ave_avail_TCH_sum/ave_avail_TCH_den

+ ave_GPRS_channels_sum/ave_GPRS_channels_den+ave_non_avail_TCH)


Figure 609. TCH availability %, S9 (ava_1d)

Average available TCH, S1 (ava_2)

Known problems: If TRXs are locked and BTSs and BCFs are unlocked, theTCHs appear as unavailable.

Note: 1. This KPI has to be counted separately for extended andnormal area (trx_type: 0 = normal, 1 = extended).2. For area (or segment) level, first use the formula for all theBTSs and then sum this formula over all underlying BTSs,keeping in mind the note above.

sum(ave_avail_full_TCH)-------------------------sum(res_av_denom2)


Figure 610. Average available TCH, S1 (ava_2)



Average available SDCCH, S1 (ava_3)

Note: 1)This KPI has to be counted separately for extended andnormal area (trx_type: 0 = normal, 1 = extended).2)For area (or segment) level, first use the formula for all theBTSs and then sum this formula over all underlying BTSs,keeping in mind the note above.

sum(ave_sdcch_sub)/sum(res_av_denom3)


Figure 611. Average available SDCCH, S1 (ava_3)

SDCCH availability %, S4 (ava_4)

Use: Indicates how big a share of all SDCCH resources has beenavailable for traffic. Failures (downtime) of TRX containingSDCCH affect this KPI.

Known problems: Affected by locked TRX under unlocked BCF and BTS.Note: This KPI has to be counted separately for extended and

normal area. Trx type: 0 = normal, 1 = extended.

sum(ave_sdcch_sub/res_av_denom3)100 * --------------------------------------------------------- %

sum(ave_sdcch_sub/res_av_denom3)+sum(ave_non_avail_sdcch)


Figure 612. SDCCH availability %, S4 (ava_4)

Average available FTCH in area, S1 (ava_5)

sum_over_area(

sum_over_BTS(ave_avail_full_TCH)/sum_over_BTS(res_av_denom2))


Figure 613. Average available FTCH in area, S1 (ava_5)

DMR availability %, S6 (ava_6)

sum(avail_time)100 * ---------------- %

sum(total_time)



Counters from table(s):p_nbsc_dmr

Figure 614. DMR availability %, S6 (ava_6)

DN2 availability %, S6 (ava_7)

sum(avail_time)100 * --------------- %

sum(total_time)

Counters from table(s):p_nbsc_dn2

Figure 615. DN2 availability %, S6 (ava_7)

TRU availability %, S6 (ava_8)

sum(avail_time)100 * --------------- %

sum(total_time)

Counters from table(s):p_nbsc_tru_bie

Figure 616. TRU availability %, S6 (ava_8)

Average defined HTCH, S1 (ava_9)

Known problems: If TRXs are locked and BTSs and BCFs are unlocked,the TCHs appear as unavailable.

Avg(ave_tch_avail_half)


Figure 617. Average defined HTCH, S1 (ava_9)

SC ET availability %, S7 (ava_10)

sum(avail_time)100 * --------------- %

sum(total_time)



Counters from table(s):p_nbsc_et_bsc.

Figure 618. SC ET availability %, S7 (ava_10)

BSC ET availability %, S7 (ava_11)

sum(remote_avail_time)100 * --------------------- %

sum(remote_total_time)

Counters from table(s):p_nbsc_et_bsc.

Figure 619. BSC ET availability %, S7 (ava_11)

SC TCSM availability %, S7 (ava_12)

sum(avail_time)100 * --------------- %

sum(total_time)

Counters from table(s):p_nbsc_et_tcsm.

Figure 620. SC TCSM availability %, S7 (ava_12)

BSC TCSM availability %, S7 (ava_13)

sum(remote_avail_time)100 * --------------------- %

sum(remote_total_time)

Counters from table(s):p_nbsc_et_tcsm.

Figure 621. BSC TCSM availability %, S7 (ava_13)

TRE availability %, S6 (ava_14)

sum(avail_time)100 * --------------- %

sum(total_time)



Counters from table(s):p_nbsc_tre.

Figure 622. TRE availability %, S6 (ava_14)

Average CS territory, S9 (ava_15)

Use: Used on BTS level. It indicates the average number of TCHsavailable for circuit switched traffic (CS).For area (orsegment) level, first use the formula for all the BTSs and thensum this formula over all underlying BTSs.Note: This is not the same as the total capacity available forCS traffic, which is defined as ava_21. If CS traffic grows, thePS territory can diminish to what is defined as dedicatedterritory and give space for CS traffic.This figure is affected by:1)The settings of BTS-parameters CDEF and CDED2)Changes in capacity:a) TRX locked or unlocked by the operator. If TRXs arelocked and BTSs and BCFs are unlocked, the TCHs of thatTRX appear as unavailable.b) TRX disabled by BSC due to fatal faults.Upgrades and downgrades of territory by BSC according totraffic needs

Known problems: If extended cells are used, this KPI shows correct value onlyonly on the BTS/trx_type level

sum(ave_avail_TCH_sum)------------------------sum(ave_avail_TCH_den)


Unit: timeslots

Figure 623. Average CS territory, S9 (ava_15)

Average available PDTCH in BTS, S9PS (ava_16)

Use: BTS level.Shows the actual average territory available for packetswitched (PS) traffic.This figure is affected by1) The settings of BTS parameters CDEF and CDED2) Changes of capacity:a) TRX locked or unlocked by the operator. If TRXs arelocked and BTSs and BCFs are unlocked, the TCHs of thatTRX appear as unavailable.



b) TRX disabled by BSC due to fatal faults.3) Upgrades and downgrades of territory by BSC according totraffic needs

Known problems: If extended TRXs are used, the values are correct only if thereport is on BTS / TRX_type level.

sum(ave_GPRS_channels_sum)---------------------------sum(ave_GPRS_channels_den)


Unit: timeslots

Figure 624. Average available PDTCH in BTS, S9PS (ava_16)

Average PS territory, S9PS (ava_16a)

Use: BTS level.Shows the actual average territory available for packetswitched (PS) traffic.The value can be used for tuning the CDEF parameter.This figure is affected by1) The settings of BTS parameters CDEF and CDED2)Changes of capacitya) TRX locked or unlocked by the operator. If TRXs arelocked and BTSs and BCFs are unlocked, the TCHs of thatTRX appear as unavailable.b) TRX disabled by BSC due to fatal faults3) Upgrades and downgrades of territory by BSC according totraffic needs

Known problems: GPRS timeslots can only be in normal TRXs.

sum(decode(trx_type,0,ave_GPRS_channels_sum))----------------------------------------------sum(decode(trx_type,0,ave_GPRS_channels_den))


Unit: timeslot

Figure 625. Average PS territory, S9PS (ava_16a)



Average available dedicated GPRS channels, S9PS (ava_17)

Use: BTS level.Indicates the average number of channels available only fordedicated PS (GPRS) traffic. This capacity is allocated bysetting the parameter CDED. In the case that there are nodedicated GPRS channels, throughput is not guaranteed, if CStraffic needs all the capacity.

Known problems: If extended TRXs are used, the values are correct only if thereport is on the BTS/TRX type level.

sum(ave_permanent_GPRS_ch_sum)--------------------------------sum(ave_permanent_GPRS_ch_den)


Figure 626. Average available dedicated GPRS channels, S9PS (ava_17)

Average available dedicated GPRS channels, S9PS (ava_17a)

Use: BTS level. For area (or segment) level, first use the formulafor all the BTSs and then sum this formula over all underlyingBTSs.Indicates the average number of channels available only fordedicated PS (GPRS) traffic. This capacity is allocated bysetting the parameter CDED. In the case that there are nodedicated GPRS channels, throughput is not guaranteed, if CStraffic needs all the capacity.

Known problems: GPRS timeslots can only be in normal TRXs.

sum(decode(trx_type,0,ave_permanent_GPRS_ch_sum))-------------------------------------------------sum(decode(trx_type,0,ave_permanent_GPRS_ch_den))


Figure 627. Average available dedicated GPRS channels, S9PS (ava_17a)

TRE-SEL availability %, S6 (ava_20)

sum(avail_time)100 * --------------- %

sum(total_time)

Counters from table(s):p_nbsc_tre_sel

Figure 628. TRE-SEL availability %, S6 (ava_20)



Number of timeslots available for CS traffic, S9 (ava_21)

Use: BTS levelThis KPI shows all the available timeslots that are notdedicated for PS (GPRS), i.e. that can be used by CS traffic.Note that this is not the same as CS territory, which is definedas ava_15.

Known problems: Incorrect if extended TRXs were used.

sum(ave_avail_TCH_sum)/sum(ave_avail_TCH_den)+ sum(ave_GPRS_channels_sum)/sum(ave_GPRS_channels_den)- sum(ave_permanent_GPRS_ch_sum)/sum(ave_permanent_GPRS_ch_den)


Figure 629. Number of timeslots available for CS traffic, S9 (ava_21)

Number of timeslots available for CS traffic on normal TRXs, S9 (ava_21a)

Use: Used on BTS levelThis formula contains all the available timeslots that are notdedicated for PS (GPRS), i.e. that can be used by CS traffic onnormal TRXs.

ava_28 ; Pure CS TCHs on normal TRXs+ava_16a-ava_17a ; dynamic part of PS territory

Figure 630. Number of timeslots available for CS traffic on normal TRXs, S9(ava_21a)

Number of HR timeslots available, S9 (ava_22)

Use: Average number of timeslots available for half rate trafficonly

Known problems: Incorrect values if extended TRXs were used.

Sum(ave_avail_TCH_sum)/sum(ave_avail_TCH_den)-sum(ave_avail_full_TCH)/sum(res_av_denom2)


Figure 631. Number of HR timeslots available, S9 (ava_22)

Number of HR timeslots available, S9 (ava_22a)

Use: Average number of timeslots available for half rate trafficonly



Known problems: Incorrect values if extended TRXs used.

ava_30+ava_31


Unit: timeslots

Figure 632. Number of HR timeslots available, S9 (ava_22a)

Number of FR timeslots available, S9 (ava_23)

Use: Average number of timeslots available for full rate traffic only

Sum(ave_avail_TCH_sum)/sum(ave_avail_TCH_den) - avg(ave_tch_avail_half)/2


Figure 633. Number of FR timeslots available, S9 (ava_23)

Number of FR timeslots available, S9 (ava_23a)

Use: Average number of timeslots available for full rate traffic only

ava_32 ; FR timeslots on normal TRXs+ ava_33 ; FR timeslots on extended TRXs


Figure 634. Number of FR timeslots available, S9 (ava_23a)

Number of dual timeslots available, S9 (ava_24)

Use: Average number of timeslots available for dual (half rate orfull rate) traffic

Known problems: Incorrect values if extended TRXs were used.

sum(ave_avail_full_TCH)/sum(res_av_denom2) + avg(ave_tch_avail_half)/2)- Sum(ave_avail_TCH_sum)/sum(ave_avail_TCH_den)


Figure 635. Number of dual timeslots available, S9 (ava_24)



Number of dual timeslots available, S9 (ava_24a)

Use: Average number of timeslots available for dual (half rate orfull rate) traffic

ava_34 ; dual timeslots on normal TRXs+ ava_35 ; dual timeslots on extended TRXs


Figure 636. Number of dual timeslots available, S9 (ava_24a)

Average number of available TCH timeslots, S9 (ava_25a)

Use: Used on BTS level. Average number of TCH timeslots (bothCS and PS) available.For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs.

ava_31 ; HR timeslots on extended TRXs+ ava_33 ; FR timeslots on extended TRXs+ ava_35 ; dual timeslots on extended TRXs+ ava_30 ; HR timeslots on normal TRXs+ ava_32 ; FR timeslots on normal TRXs+ ava_34; ; Dual timeslots on normal TRXs+ ava_16a ; PS territory timeslots (always on normal TRXs))


Figure 637. Average number of available TCH timeslots, S9 (ava_25a)

Number of available TCH timeslots, PS and CS common, S9 (ava_26)

Use: Average number of timeslots available both CS and PStraffic.These are part of PS (GPRS) territory but if CS trafficneeds they can be taken to CS call use (territory downgrade)automatically by BSC.

Known problems: If extended TRXs were used, the values are correct only if thereport is on the BTS/TRX_type level.

ava_16-ava_17

Figure 638. Number of available TCH timeslots, PS and CS common, S9(ava_26)



Number of available TCH timeslots, PS and CS common, S9 (ava_26a)

Use: Average number of timeslots available for both CS and PStraffic. These are part of PS (GPRS) territory but if CS trafficneeds they can be taken to CS call use (territory downgrade)automatically by BSC. For area (or segment) level, first usethe formula for all the BTSs and then sum this formula overall underlying BTSs.

ava_16a-ava_17a

Figure 639. Number of available TCH timeslots, PS and CS common, S9(ava_26a)

Average CS TCH in normal TRXs, S9 (ava_28)

Use: BTS level. Indicates the average number of TCHs availablefor circuit switched traffic (CS) in normal TRXs. Thiscapacity can also be used for PS traffic if CS traffic is low.IfCS traffic grows, the PS territory can diminish to what isdefined as dedicated territory and give space for CStraffic.This figure is affected byThis figure is affected by1) The settings of BTS-parameters CDEF andCDED2) Changes of capacity:a) TRX locked or unlocked by the operator. If TRXs arelocked and BTSs and BCFs are unlocked, the TCHs of thatparticular TRX appear as unavailable.b) TRX disabled by BSC due to fatal faults.3) Upgrades and downgrades of territory by BSC according totraffic needs.

sum(decode(trx_type,0,ave_avail_TCH_sum))/sum(decode(trx_type,0,ave_avail_TCH_den))


Unit: timeslots

Figure 640. Average CS TCH in normal TRXs, S9 (ava_28)

Average available CS TCH in extended TRXs, S9 (ava_29)

Use: BTS level. Indicates the average number of TCHs availablefor circuit switched traffic (CS) in extended TRXs. Thiscapacity cannot be used for PS traffic.

nvl(sum(decode(trx_type,1,ave_avail_TCH_sum))/sum(decode(trx_type,1,ave_avail_TCH_den)),1)




p_nbsc_res_avail

Unit: timeslots

Figure 641. Average available CS TCH in extended TRXs, S9 (ava_29)

Number of HR tsls available, normal TRXs, S9 (ava_30)

Use: Average number of timeslots available for half rate trafficonly. On BTS level. For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs.

Sum(decode(trx_type,0,ave_avail_TCH_sum))/sum(decode(trx_type,0,ave_avail_TCH_den))-sum(decode(trx_type,0,ave_avail_full_TCH))/sum(decode(trx_type,0,res_av_denom2))


Figure 642. Number of HR tsls available, normal TRXs, S9 (ava_30)

Number of HR tsls available, extended TRXs S9 (ava_31)

Use: Average number of timeslots available on extended TRXs forfull rate traffic only.Used on the BTS level. If used on the area level, it shows theaverage over extended cells only. For area (or segment) level,first use the formula for all the BTSs and then sum thisformula over all underlying BTSs.

nvl(Sum(decode(trx_type,1,ave_avail_TCH_sum))/sum(decode(trx_type,1,ave_avail_TCH_den))

-sum(decode(trx_type,1,ave_avail_full_TCH))/sum(decode(trx_type,1,res_av_denom2)),0)


Figure 643. Number of HR tsls available, extended TRXs S9 (ava_31)

Number of FR timeslots available, normal TRXs, S9 (ava_32)

Use: Average number of timeslots available on normal TRXs forfull rate traffic only. On BTS level. For area (or segment)level, first use the formula for all the BTSs and then sum thisformula over all underlying BTSs.

Sum(decode(trx_type,0,ave_avail_TCH_sum))/sum(decode(trx_type,0,ave_avail_TCH_den))

- avg(decode(trx_type,0,ave_tch_avail_half))/2




Figure 644. Number of FR timeslots available, normal TRXs, S9 (ava_32)

Number of FR timeslots available, extended TRXs, S9 (ava_33)

Use: Average number of timeslots available on extended TRXs forfull rate traffic only.Used on the BTS level. If used on the area level, it shows theaverage over extended cells only. For area (or segment) level,first use the formula for all the BTSs and then sum thisformula over all underlying BTSs.

nvl(Sum(decode(trx_type,1,ave_avail_TCH_sum))/sum(decode(trx_type,1,ave_avail_TCH_den))

- avg(decode(trx_type,1,ave_tch_avail_half))/2,0)


Figure 645. Number of FR timeslots available, extended TRXs, S9 (ava_33)

Number of dual timeslots available, normal TRXs, S9 (ava_34)

Use: Average number of timeslots available for dual (half rate orfull rate) traffic. On BTS level. For area (or segment) level,first use the formula for all the BTSs and then sum thisformula over all underlying BTSs.

sum(decode(trx_type,0,ave_avail_full_TCH))/sum(decode(trx_type,0,res_av_denom2))+ avg(decode(trx_type,0,ave_tch_avail_half))/2-Sum(decode(trx_type,0,ave_avail_TCH_sum))/sum(decode(trx_type,0,ave_avail_TCH_den)

)+


Figure 646. Number of dual timeslots available, normal TRXs, S9 (ava_34)

Number of dual timeslots available, extended TRXs, S9 (ava_35)

Use: Average number of timeslots available for dual (half rate orfull rate) traffic.Used on the BTS level. If used on the area level, it shows theaverage over extended cells only. For area (or segment) level,first use the formula for all the BTSs and then sum thisformula over all underlying BTSs.



nvl(sum(decode(trx_type,1,ave_avail_full_TCH))/sum(decode(trx_type,1,res_av_denom2))+ avg(decode(trx_type,1,ave_tch_avail_half))/2-Sum(decode(trx_type,1,ave_avail_TCH_sum))/sum(decode(trx_type,1,ave_avail_TCH_den)

),0)


Figure 647. Number of dual timeslots available, extended TRXs, S9 (ava_35)

GPRS enable time %, S10 (ava_36)

Use: To indicate the GPRS availability in time.

(100*(AVE_PCH_GB_LOAD_ON_CCCH_DEN / RES_ACC_DENOM1))


Unit: %

Figure 648. GPRS enable time %, S10 (ava_36)

Number of HR TSLs available, extended TRXs, Area, S9 (ava_37)

Use: Average number of timeslots available for half rate trafficonly. Used on the BTS level. If used on the area level, it showsthe average over extended cells only. For area (or segment)level, first use the formula for all the BTSs and then sum thisformula over all underlying BTSs.

nvl((

sum(decode(trx_type,1,ave_avail_TCH_sum))--------------------------------------------(avg(decode(trx_type,1,ave_avail_TCH_den))* count(distinct period_start_time))

)-

(sum(decode(trx_type,1,ave_avail_full_TCH))

--------------------------------------------(avg(decode(trx_type,1,res_av_denom2)* count(distinct period_start_time))

),0)


Figure 649. Number of HR TSLs available, extended TRXs, Area, S9 (ava_37)



Number of FR TSLs available, extended TRXs, Area, S9 (ava_38)

Use: Average number of timeslots available on extended TRXs forfull rate traffic only.

nvl(sum(decode(trx_type,1,ave_avail_TCH_sum))

--------------------------------------------(avg(decode(trx_type,1,ave_avail_TCH_den))* count(distinct period_start_time))

-sum(decode(trx_type,1,ave_tch_avail_half * RES_AV_DENOM1))

---------------------------------------------------------------(2 * avg(decode(trx_type,1,ave_avail_TCH_den))* count(distinct period_start_time))

,0)


Figure 650. Number of FR TSLs available, extended TRXs, Area, S9 (ava_38)

Average number of available TCH TSLs, Area, S9 (ava_39)

Use: Average number of TCH timeslots (both CS and PS)available.

(ava_37 ; HR timeslots on extended TRXs

+ ava_38 ; FR timeslots on extended TRXs+ ava_40 ; Dual timeslots on extended TRXs+ ava_42 ; HR timeslots on normal TRXs+ ava_43 ; FR timeslots on normal TRXs+ ava_41 ; Dual timeslots on normal TRXs+ ava_44 ; PS territory timeslots (always on normal TRXs))


Figure 651. Average number of available TCH TSLs, Area, S9 (ava_39)

Number of dual TSLs available, extended TRXs, Area, S9 (ava_40)

Use: Average number of timeslots available for dual (half rate orfull rate) traffic.

nvl((

sum(decode(trx_type,1,ave_avail_full_TCH))--------------------------------------------(avg(decode(trx_type,1,res_av_denom2))* count(distinct period_start_time))

)+

(sum(decode(trx_type,1,ave_tch_avail_half * RES_AV_DENOM1))

---------------------------------------------------------------(2 * avg(decode(trx_type,1, RES_AV_DENOM1)))

)



-(

sum(decode(trx_type,1,ave_avail_TCH_sum))--------------------------------------------(avg(decode(trx_type,1,ave_avail_TCH_den))* count(distinct period_start_time))

),0)


Figure 652. Number of dual TSLs available, extended TRXs, Area, S9 (ava_40)

Number of dual TSLs available, normal TRXs, Area, S9 (ava_41)

Use: Average number of timeslots available for dual (half rate orfull rate) traffic.

nvl((

sum(decode(trx_type,0,ave_avail_full_TCH))--------------------------------------------(avg(decode(trx_type,0,res_av_denom2))* count(distinct period_start_time))

)+

(sum(decode(trx_type,0,ave_tch_avail_half * RES_AV_DENOM1))

--------------------------------------------------------------(2 * avg(decode(trx_type,0, RES_AV_DENOM1)))

)-

(sum(decode(trx_type,0,ave_avail_TCH_sum))

-------------------------------------------(avg(decode(trx_type,0,ave_avail_TCH_den))* count(distinct period_start_time))

),0)


Figure 653. Number of dual TSLs available, normal TRXs, Area, S9 (ava_41)

Number of HR TSLs available, normal TRXs, Area, S9 (ava_42)

Use: On the area level. Average number of timeslots available forhalf rate traffic only.


-------------------------------------------------------------------------------(avg(decode(trx_type,0,ave_avail_TCH_den))* count(distinct period_start_time))

)-(

sum(decode(trx_type,0,ave_avail_full_TCH))----------------------------------------------------------------------------



(avg(decode(trx_type,0,res_av_denom2)) * count(distinct period_start_time)))Counters from table(s):p_nbsc_res_avail

Figure 654. Number of HR TSLs available, normal TRXs, Area, S9 (ava_42)

Number of FR TSLs available, normal TRXs, Area, S9 (ava_43)

Use: Average number of timeslots available on normal TRXs forfull rate traffic only.


-------------------------------------------------------------------------------(avg(decode(trx_type,0,ave_avail_TCH_den)) * count(distinct period_start_time))

)-(

sum(decode(trx_type,0,ave_tch_avail_half * RES_AV_DENOM1))-------------------------------------------------------------------------------(2 * avg(decode(trx_type,0,RES_AV_DENOM1)) * count(distinct period_start_time))

)


Figure 655. Number of FR TSLs available, normal TRXs, Area, S9 (ava_43)

Average PS territory, Area, S9PS (ava_44)

Use: BTS level. For area (or segment) level, first use the formulafor all the BTSs and then sum this formula over all underlyingBTSs.Shows the actual average territory available for packetswitched (PS) traffic.The value can be used for tuning the CDEF parameter.This figure is affected by1) The settings of BTS parameters CDEF and CDED2) Changes of capacitya) TRX locked or unlocked by the operator. If TRXs arelocked and BTSs and BCFs are unlocked, the TCHs of thatTRX appear as unavailable.b) TRX disabled by BSC due to fatal faults3) Upgrades and downgrades of territory by BSC according totraffic needs

sum(decode(trx_type,0,ave_GPRS_channels_sum))--------------------------------------------------------------------------------(avg(decode(trx_type,0, decode(ave_GPRS_channels_sum,0,0,ave_GPRS_channels_den)))* count(distinct period_start_time))




p_nbsc_res_avail

Unit: timeslots

Figure 656. Average PS territory, Area, S9PS (ava_44)

Average available SDCCH, Area, S1 (ava_45)

sum(ave_sdcch_sub)--------------------------------------------------------(avg(res_av_denom3) * count(distinct period_start_time))


Figure 657. Average available SDCCH, Area, S1 (ava_45)

2.32 Unavailability (uav)

Average unavailable TSL per BTS, S1 (uav_1)

Use: For area (or segment) level, first use the formula for all theBTSs and then sum this formula over all underlying BTSs.


sum(ave_non_avail_tsl)/sum(res_av_denom1)


Figure 658. Average unavailable TSL per BTS, S1 (uav_1)

Average unavailable TSL per BTS, S1 (uav_1a)


avg(ave_non_avail_tsl/res_av_denom1)


Figure 659. Average unavailable TSL per BTS, S1 (uav_1a)



Average unavailable TSL per BTS, S1 (uav_1b)


sum(ave_non_avail_tsl/res_av_denom1)------------------------------------

count(*)


Figure 660. Average unavailable TSL per BTS, S1 (uav_1b)

Total outage time (uav_2)

Known problems: It should be made possible to differentiate the reasons foroutage.

Note: Alarm number changed in S7 from 2567 to 7767

sum of BCCH missing alarm durations =

sum(cancel_time-alarm_time)*24*60

where probable_cause = 7767 /* BCCH missing alarm */

Counters from table(s):fx_alarmunit = minutes

Figure 661. Total outage time (uav_2)

Number of outages (uav_3)

Note: Alarm number changed in S7 from 2567 to 7767.

number of BCCH missing alarm starts =

count(alarm_start_time)

where probable_cause = 7767 /* BCCH missing alarm */

Counters from table(s):fx_alarm

Figure 662. Number of outages (uav_3)

Share of unavailability due to user (uav_4)

Experiences on use: Locked TRXs can make this PI show high values.Known problems: The measurement is made on the BSC level. The BTS level

cannot be seen.



sum(ave_non_avail_user)100 * -------------------------------------------------------------- %

sum(ave_non_avail_user + ave_non_avail_int + ave_non_avail_ext)

Counters from table(s):p_nbsc_trx_avail

Figure 663. Share of unavailability due to user (uav_4)

Share of unavailability due to internal reasons (uav_5)

Known problems: The measurement is made on the BSC level. The BTS levelcannot be seen. This includes, for example, also an electricitybreak which, in fact, is not a BTS fault.

sum(ave_non_avail_int)100 * -------------------------------------------------------------- %



Figure 664. Share of unavailability due to internal reasons (uav_5)

Share of unavailability due to external reasons (uav_6)

Known problems: The measurement is made on the BSC level. The BTS levelcannot be seen.

sum(ave_non_avail_ext)100 * -------------------------------------------------------------- %



Figure 665. Share of unavailability due to external reasons (uav_6)

TRX unavailability time due to user (uav_7)

Experiences on use: Locked TRXs can make this PI show high values.

sum(period_duration * ave_non_avail_user)


Figure 666. TRX unavailability time due to user (uav_7)



TRX unavailability time due to internal reasons (uav_8)

Known problems: This includes, for example, also an electricity break which, infact, is not a BTS fault.

sum(period_duration * ave_non_avail_int)


Figure 667. TRX unavailability time due to internal reasons (uav_8)

TRX unavailability time due to external reasons (uav_9)

sum(period_duration * ave_non_avail_ext)


Figure 668. TRX unavailability time due to external reasons (uav_9)

Average unavailable SDCCH, S5 (uav_10)

Note: 1)This KPI has to be counted separately for extended andnormal area (trx_type: 0=normal, 1=extended).2)For area (or segment) level, first use the formula for all theBTSs and then sum this formula over all underlying BTSs.

avg(ave_non_avail_sdcch)


Figure 669. Average unavailable SDCCH, S5 (uav_10)

Average unavailable TCH, S5 (uav_11a)

Use: On BTS level. For area (or segment) level, first use theformula for all the BTSs and then sum this formula over allunderlying BTSs

Known problems: Locked TRXs are counted as unavailable TCH

uav_13+uav_14

Figure 670. Average unavailable TCH, S5 (uav_11a)



Average bearer unavailability, S9PS (uav_12)

100*sum(time_bear_oper_unoper)/sum(period_duration*60)


Figure 671. Average bearer unavailability, S9PS (uav_12)

Average unavailable TCH on normal TRXs, S5 (uav_13)

Known problems: Locked TRXs are counted as unavailable TCH.Note: For area (or segment) level, first use the formula for all the

BTSs and then sum this formula over all underlying BTSs.

avg(decode(trx_type,0,ave_non_avail_tch))


Figure 672. Average unavailable TCH on normal TRXs, S5 (uav_13)

Average unavailable TCH on extended TRXs, S5 (uav_14)

Use: On the BTS level. If used on the area level, it shows theaverage over extended cells only.

Known problems: Locked TRXs are counted as unavailable TCH. For area (orsegment) level, first use the formula for all the BTSs and thensum this formula over all underlying BTSs.

nvl(avg(decode(trx_type,1,ave_non_avail_tch)),0)


Figure 673. Average unavailable TCH on extended TRXs, S5 (uav_14)

Average unavailable TCH on normal TRXs, Area, S5 (uav_15)

Known problems: Locked TRXs are counted as unavailable TCH.

sum(decode(trx_type,0,ave_non_avail_tch * RES_AV_DENOM1))-------------------------------------------------------------------------avg(decode(trx_type,0,RES_AV_DENOM1)) * count(distinct period_start_time)


Figure 674. Average unavailable TCH on normal TRXs, Area, S5 (uav_15)



Average unavailable TCH on extended TRXs, Area, S5 (uav_16)

Known problems: Locked TRXs are counted as unavailable TCH.

nvl(sum(decode(trx_type,1,ave_non_avail_tch * RES_AV_DENOM1)),0)-------------------------------------------------------------------------avg(decode(trx_type,1,RES_AV_DENOM1)) * count(distinct period_start_time)


Figure 675. Average unavailable TCH on extended TRXs, Area, S5 (uav_16)

Average unavailable TCH, Area, S5 (uav_17)

Use: Average unavailable TCH on Area level.Known problems: Locked TRXs are counted as unavailable TCH.

uav_15 + uav_16

Figure 676. Average unavailable TCH, Area, S5 (uav_17)

2.33 Location updates (lu)

Number of LU attempts, S1 (lu_1)

sum(sdcch_loc_upd)


Figure 677. Number of LU attempts, S1 (lu_1)

Average of LU attempts per hour, S1 (lu_2)

sum(sdcch_loc_upd)--------------------------------avg(period_duration)*count(*)/60


Figure 678. Average of LU attempts per hour, S1 (lu_2)

Number of LU attempts, S1 (lu_3)

sum(nbr_of_calls)



where counter_id = 25 /* LU started */


Figure 679. Number of LU attempts, S1 (lu_3)

2.34 LU success % (lsr)

LU success %, S6 (lsr_2)

Use: Probable causes to make this KPI show bad values:interference, coverage, possibly MSC side problems.

Known problems: The measurement is made on the BSC level.The LU started (51025) is triggered from establish indication.Any problems prior to that cannot be seen. For example,interference prevents a mobile station from making a locationupdate.

Values: Good: >99% .

sum(nbr_of_calls) where counter_id = 26 /* LU completed */100 * -------------------------------------------------------------- %

sum(nbr_of_calls) where counter_id = 25 /* LU started */


Figure 680. LU success %, S6 (lsr_2)

2.35 Emergency call (ec)

Emergency calls, S6 (ec_1)

sum(nbr_of_calls)where counter_id = 35 /* Em.call started */


Figure 681. Emergency calls, S6 (ec_1)



2.36 Emergency call success % (ecs)

Emergency call success %, S6 (ecs_1)

sum(nbr_of_calls) where counter_id = 41 /* Em.call completed */100 * -----------------------------------------------------------------------

sum(nbr_of_calls) where counter_id = 35 /* Em.call started */


Figure 682. Emergency call success %, S6 (ecs_1)

2.37 Call re-establishment (re)

Call re-establishment attempts, S6 (re_1)

sum(nbr_of_calls)where counter_id = 36 /* Callreest. started */


Figure 683. Call re-establishment attempts, S6 (re_1)

Call re-establishments, S6 (re_2)

sum(sdcch_call_re_est+tch_call_re_est)


Figure 684. Call re-establishments, S6 (re_2)

2.38 Call re-establishment success % (res)

Call re-establishment success %, S6 (res_1)

sum(nbr_of_calls) where counter_id = 42 /* Call re-est. completed */100 * -----------------------------------------------------------------------

sum(nbr_of_calls) where counter_id = 36 /* Call re-est. started */




Figure 685. Call re-establishment success %, S6 (res_1)

2.39 Quality

2.39.1 Downlink quality (dlq)

DL BER, S1 (dlq_1)

Known problems: BER % is not a very easy entity for network planners.Measurement periods that have no period on TCH, havepower_denom5 = 0.

sum(ave_dl_sig_qual)------------------------ %sum(power_denom5)*100

Counters from table(s):p_nbsc_powerUnit = BER %

Figure 686. DL BER, S1 (dlq_1)

DL cumulative quality % in class X, S1 (dlq_2)

Use: This PI gives a cumulative percentage of call samples (AMR,non-AMR in classes 0 to X. X = 5 is normally used as aquality indicator. If X = 5 and this figure is 100 %, then theMS users obviously have not perceived any quality problems.

Known problems: If DL DTX is used, there is a shift to a worse quality %, but aMS does not perceive this.

sum(freq_dl_qual0 + ... + freq_dl_qualX)100 * --------------------------------------------- %

sum( freq_dl_qual0 + ... + freq_dl_qual7)


Figure 687. DL cumulative quality % in class X, S1 (dlq_2)



DL cumulative quality % in class X, S1 (dlq_2a)

Use: This PI gives a cumulative percentage of call samples inclasses 0 to X. X=5 is normally used as quality indicator. IfX=5 and this figure is 100 %, then the MS users obviouslyhave not perceived any quality problems.

sum(freq_dl_qual0 + ... + freq_dl_qualX)100 * --------------------------------------------- %

sum(freq_dl_qual0 + ... + freq_dl_qual7)

Counters from table(s):p_nbsc_rx_statistics

Figure 688. DL cumulative quality % in class X, S1 (dlq_2a)

DL quality %, FER based, S10 (dlq_3)

Use: The share (as a percentage) of call samples in FEP classes 0 to7.Classes are defined by boundaries B0 to B8. Boundaries 1 to7 can be set as a measurement parameter. Class 0 is betweenboundaries B0 (fixed 0%) and B1. Class 7 is betweenboundaries B7 and B8 (fixed 100%).

Note: In S10 all DL samples are estimated.

sum(NBR_OF_DL_FER_CL_X)100 * --------------------------------------------- %

sum(NBR_OF_DL_FER_CL_0++NBR_OF_DL_FER_CL_7)

Counters from table(s):p_nbsc_fer

Figure 689. DL quality %, FER based, S10 (dlq_3)

DL cumulative quality % in class X, HR AMR, S10 (dlq_4)

Use: This PI gives the cumulative percentage of call samples(downlink half rate AMR) in classes 0 to Z. Z = 5 is normallyused as a quality indicator.

sum(nvl(AMR_HR_MODE_1_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_DL_RXQUAL_Z,0)

+nvl(AMR_HR_MODE_2_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_DL_RXQUAL_Z,0)+nvl(AMR_HR_MODE_3_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_DL_RXQUAL_Z,0)+nvl(AMR_HR_MODE_4_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_DL_RXQUAL_Z,0))

100* ------------------------------------------------------------------------- %sum(

nvl(AMR_HR_MODE_1_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_DL_RXQUAL_7,0)+nvl(AMR_HR_MODE_2_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_DL_RXQUAL_7,0)+nvl(AMR_HR_MODE_3_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_DL_RXQUAL_7,0)+nvl(AMR_HR_MODE_4_DL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_DL_RXQUAL_7,0))




Figure 690. DL cumulative quality % in class X, HR AMR, S10 (dlq_4)

DL cumulative quality % in class X, FR AMR, S10 (dlq_5)

Use: This PI gives the cumulative percentage of call samples(downlink full rate AMR) in classes 0 to Z. Z = 5 is normallyused as a quality indicator.

sum(nvl(AMR_FR_MODE_1_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_DL_RXQUAL_Z,0

+nvl(AMR_FR_MODE_2_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_2_DL_RXQUAL_Z,0+nvl(AMR_FR_MODE_3_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_3_DL_RXQUAL_Z,0+nvl(AMR_FR_MODE_4_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_4_DL_RXQUAL_Z,0)

100* -------------------------------------------------------------------------- %sum(

nvl(AMR_FR_MODE_1_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_DL_RXQUAL_7,0+nvl(AMR_FR_MODE_2_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_2_DL_RXQUAL_7,0+nvl(AMR_FR_MODE_3_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_3_DL_RXQUAL_7,0+nvl(AMR_FR_MODE_4_DL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_4_DL_RXQUAL_7,0)


Figure 691. DL cumulative quality % in class X, FR AMR, S10 (dlq_5)

DL cumulative quality % in class X, S10 (dlq_6)

Use: This PI gives the cumulative percentage of all call samples(AMR and non-AMR) in classes 0 to Z. Z = 5 is normally usedas a quality indicator. If Z = 5 and this figure is 100 %, the MSusers obviously have not perceived any quality problems.

sum(freq_dl_qual0 + ... + freq_dl_qualZ)-sum(

nvl(AMR_HR_MODE_1_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_1_DL_RXQUAL_Z,0)+ nvl(AMR_HR_MODE_2_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_2_DL_RXQUAL_Z,0)+ nvl(AMR_HR_MODE_3_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_3_DL_RXQUAL_Z,0)+ nvl(AMR_HR_MODE_4_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_4_DL_RXQUAL_Z,0))

-sum(nvl(AMR_FR_MODE_1_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_1_DL_RXQUAL_Z,0)

+ nvl(AMR_FR_MODE_2_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_2_DL_RXQUAL_Z,0)+ nvl(AMR_FR_MODE_3_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_3_DL_RXQUAL_Z,0)+ nvl(AMR_FR_MODE_4_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_4_DL_RXQUAL_Z,0)

)100 * --------------------------------------------------------------------------- %

sum(freq_dl_qual0 + ... + freq_dl_qual7)-sum(

nvl(AMR_HR_MODE_1_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_1_DL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_2_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_2_DL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_3_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_3_DL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_4_DL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_4_DL_RXQUAL_7,0))

-sum(



nvl(AMR_FR_MODE_1_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_1_DL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_2_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_2_DL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_3_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_3_DL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_4_DL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_4_DL_RXQUAL_7,0)

)


Figure 692. DL cumulative quality % in class X,S10 (dlq_6)

DL quality 0-5 %, HR, FER based, S10 (dlq_7)

Use: FER based quality benchmark for half rate speech trafficchannel.

sum(NBR_OF_DL_FER_CL_0+ +NBR_OF_DL_FER_CL_5)100 * --------------------------------------------- %

sum(NBR_OF_DL_FER_CL_0+ +NBR_OF_DL_FER_CL_7)

codec_type = 1Counters from table(s):p_nbsc_fer

Figure 693. DL quality 0-5 %, HR, FER based, S10 (dlq_7)

DL quality 0-5 %, FR, FER based, S10 (dlq_8)

Use: FER based quality benchmark for FR calls.




Figure 694. DL quality 0-5 %, FR, FER based, S10 (dlq_8)

DL quality 0-5 % EFR, FER based, S10 (dlq_9)




codec_type = 2




Figure 695. DL quality 0-5 % EFR, FER based, S10 (dlq_9)

DL quality 0-5 % AMR HR, FER based, S10 (dlq_10)

Use: FER based quality benchmark for AMR HR calls.



codec_type = 4..9Counters from table(s):p_nbsc_fer

Figure 696. DL quality 0-5 % AMR HR, FER based, S10 (dlq_10)

DL quality 0-5 % AMR FR, FER based, S10 (dlq_11)

Use: FER based quality benchmark for AMR FR calls.




Figure 697. DL quality 0-5 % AMR FR, FER based, S10 (dlq_11)

2.39.2 Uplink quality (ulq)

UL BER, S1 (ulq_1)

Known problems: BER % is not a very easy entity for network planners.Measurement periods that have no period on TCH, havepower_denom6 = 0.

sum(ave_ul_sig_qual)------------------------ %sum(power_denom6)*100


Figure 698. UL BER, S1 (ulq_1)



UL cumulative quality % in class X, S1 (ulq_2)

Use: This PI gives a cumulative percentage of call samples (nonAMR and AMR) in classes 0 to X. X=5 is normally used asquality indicator. If X=5 and this figure is 100 %, then the MSusers obviously have not perceived any quality problems.

Known problems: Investigations in late 1997 showed that UL DTX makes ULquality seem worse than it actually is. The impact was aboutone 1% unit (1% of samples more in classes 6 and 7). Wheninvestigated with field tests, no real degradation of qualitycould be found.

sum(freq_ul_qual0 + ... + freq_ul_qualX)100 * --------------------------------------------- %

sum(freq_ul_qual0 + ... + freq_ul_qual7)


Figure 699. UL cumulative quality % in class X, S1 (ulq_2)

UL cumulative quality % in class X, S1 (ulq_2a)

Use: This PI gives the cumulative percentage of call samples inclasses 0 to X.X = 5 is normally used as a quality indicator. IfX= 5 and this figure is 100 %, the MS users obviously havenot perceived any quality problems.

sum(freq_ul_qual0 + ... + freq_ul_qualX)100 * --------------------------------------------- %

sum(freq_ul_qual0 + ... + freq_ul_qual7)


Figure 700. UL cumulative quality % in class X, S1 (ulq_2a)

UL quality %, FER based, S10 (ulq_3)

Use: The share (as a percentage) of call samples in FER classes 0to 7.Classes are defined by boundaries B0 to B8. Boundaries 1 to7 can be set as a measurement parameter. Class 0 is betweenboundaries B0 (fixed 0%) and B1. Class 7 is betweenboundaries B7 and B8 (fixed 100%).

sum(NBR_OF_UL_FER_CL_X)100 * --------------------------------------------- %

sum(NBR_OF_UL_FER_CL_0++NBR_OF_UL_FER_CL_7)




Figure 701. UL quality %, FER based, S10 (ulq_3)

UL cumulative quality % in class X, HR AMR, S10 (ulq_4)

Use: This PI gives the cumulative percentage of call samples(uplink half rate AMR) in classes 0 to Z. Z = 5 is normallyused as a quality indicator.

sum(nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_UL_RXQUAL_Z,0)

+nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_UL_RXQUAL_Z,0)+nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_UL_RXQUAL_Z,0)+nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_UL_RXQUAL_Z,0))

100* ------------------------------------------------------------------------ %sum(

nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_1_UL_RXQUAL_7,0)+nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_2_UL_RXQUAL_7,0)+nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_3_UL_RXQUAL_7,0)+nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_HR_MODE_4_UL_RXQUAL_7,0))


Figure 702. UL cumulative quality % in class X, HR AMR, S10 (ulq_4)

UL cumulative quality % in class X, FR AMR, S10 (ulq_5)

Use: This PI gives the cumulative percentage of call samples (fullrate AMR) in classes 0 to Z. Z = 5 is normally used as a qualityindicator.

sum(nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_UL_RXQUAL_Z,0

+nvl(AMR_FR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_2_UL_RXQUAL_Z,0+nvl(AMR_FR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_3_UL_RXQUAL_Z,0+nvl(AMR_FR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_4_UL_RXQUAL_Z,0)

100* -------------------------------------------------------------------------- %sum(

nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_1_UL_RXQUAL_7,0+nvl(AMR_FR_MODE_2_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_2_UL_RXQUAL_7,0+nvl(AMR_FR_MODE_3_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_3_UL_RXQUAL_7,0+nvl(AMR_FR_MODE_4_UL_RXQUAL_0,0)+ + nvl(AMR_FR_MODE_4_UL_RXQUAL_7,0)


Figure 703. UL cumulative quality % in class X, FR AMR, S10 (ulq_5)



UL cumulative quality % in class X, non-AMR S10 (ulq_6)

Use: This PI gives a cumulative percentage of all call samples(AMR and non AMR) in classes 0 to Z. Z=5 is normally usedas quality indicator. If Z=5 and this figure is 100 %, then theMS users obviously have not perceived any quality problems.

Known problems: Investigations late 1997 showed that UL DTX makes ULquality to show worse. The impact was about one 1% unit (1%of samples more in classes 6 and 7). When investigated withfield tests no real degradiation of quality could be found.

sum(freq_ul_qual0 + ... + freq_ul_qualZ)-sum(

nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_1_UL_RXQUAL_Z,0)+ nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_2_UL_RXQUAL_Z,0)+ nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_3_UL_RXQUAL_Z,0)+ nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_4_UL_RXQUAL_Z,0)

)-sum(

nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_1_UL_RXQUAL_Z,0)+ nvl(AMR_FR_MODE_2_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_2_UL_RXQUAL_Z,0)+ nvl(AMR_FR_MODE_3_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_3_UL_RXQUAL_Z,0)+ nvl(AMR_FR_MODE_4_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_4_UL_RXQUAL_Z,0)

)100 * ------------------------------------------------------------------------- %

sum(freq_ul_qual0 + ... + freq_ul_qual7)-sum(

nvl(AMR_HR_MODE_1_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_1_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_2_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_3_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_3_UL_RXQUAL_7,0)+ nvl(AMR_HR_MODE_4_UL_RXQUAL_0,0) +...+ nvl(AMR_HR_MODE_4_UL_RXQUAL_7,0)

)-sum(

nvl(AMR_FR_MODE_1_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_1_UL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_2_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_2_UL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_3_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_3_UL_RXQUAL_7,0)+ nvl(AMR_FR_MODE_4_UL_RXQUAL_0,0) +...+ nvl(AMR_FR_MODE_4_UL_RXQUAL_7,0)

)


Figure 704. UL cumulative quality % in class X, non-AMR S10 (ulq_6)

UL quality 0-5 %, HR, FER based, S10 (ulq_7)

Use: FER based quality benchmark for HR calls.

sum(NBR_OF_UL_FER_CL_0+ +NBR_OF_UL_FER_CL_5)100 * --------------------------------------------- %

sum(NBR_OF_UL_FER_CL_0+ +NBR_OF_UL_FER_CL_7)


Figure 705. UL quality 0-5 %, HR, FER based, S10 (ulq_7)



UL quality 0-5 %, FR, FER based, S10 (ulq_8)





Figure 706. UL quality 0-5 %, FR, FER based, S10 (ulq_8)

UL quality 0-5 % EFR, FER based, S10 (ulq_9)

Use: FER based quality benchmark for EFR calls.




Figure 707. UL quality 0-5 % EFR, FER based, S10 (ulq_9)

UL quality 0-5 % AMR HR, FER based, S10 (ulq_10)

Use: FER based quality benchmark for AMR HR calls.




Figure 708. UL quality 0-5 % AMR HR, FER based, S10 (ulq_10)

UL quality 0-5 % AMR FR, FER based, S10 (ulq_11)

Use: FER based quality benchmark for AMR FR calls.



codec_type = 10..17




Figure 709. UL quality 0-5 % AMR FR, FER based, S10 (ulq_11)

2.40 Downlink and uplink level

2.40.1 Downlink level (dll)

Share % per range, S4 (dll_1)

sum over a range (class_upper_range)(freq_dl_qual0+freq_dl_qual1+freq_dl_qual2+freq_dl_qual3+freq_dl_qual4+freq_dl_qual5+freq_dl_qual6+freq_dl_qual7)

100 * ----------------------------------------------------------- %sum over all ranges

(freq_dl_qual0+freq_dl_qual1+freq_dl_qual2+freq_dl_qual3+freq_dl_qual4+freq_dl_qual5+freq_dl_qual6+freq_dl_qual7)


Figure 710. Share % per range, S4 (dll_1)

Sorting factor for undefined adjacent cell, S4 (dll_2)

Use: Helps to sort the list of undefined adjacent cells.

sum(ave_dl_sig_str)/1000

Counters from table(s):p_nbsc_undef_adj_cell

Figure 711. Sorting factor for undefined adjacent cell, S4 (dll_2)

2.40.2 Uplink level (ull)

Share % per range, S4 (ull_1)

sum over a range (class_upper_range)(freq_ul_qual0+freq_ul_qual1+freq_ul_qual2+freq_ul_qual3+freq_ul_qual4+freq_ul_qual5+freq_ul_qual6+freq_ul_qual7)

100 * ----------------------------------------------------------- %sum over all ranges

(freq_ul_qual0+freq_ul_qual1+freq_ul_qual2+freq_ul_qual3+freq_ul_qual4+freq_ul_qual5+freq_ul_qual6+freq_ul_qual7)




Figure 712. Share % per range, S4 (ull_1)

2.41 Power (pwr)

Average MS power, S2 (pwr_1)

max_power-2*sum(ave_ms_power)/sum(power_denom1)

max_power = 43 (GSM900) or max_power = 30 (GSM1800, GSM1900)


Figure 713. Average MS power, S2 (pwr_1)

Average MS power, S2 (pwr_1b)

Note: max_power = 43 (GSM900) or 30 (GSM1800, GSM1900)

decode(objects.frequency_band_in_use,0,43,30)-2*sum(ave_ms_power)/sum(power_denom1)


Figure 714. Average MS power, S2 (pwr_1b)

Average BS power, S2 (pwr_2)

max_power - 2*sum(ave_BS_power)/sum(power_denom2)

max_power depends on the TRX used.


Figure 715. Average BS power, S2 (pwr_2)

2.42 Level (lev)

Average DL signal strength, S2 (lev_1)

-110+sum(ave_dl_sig_str)/sum(power_denom3)




Figure 716. Average DL signal strength, S2 (lev_1)

Average DL signal strength, S2 (lev_1a)

decode(ave_dl_sig_str/power_denom3,0,’< -110’,63,’> -48’,(-110+(round(ave_dl_sig_str/power_denom3)-1))||’..’||(-110+round(ave_dl_sig_str/power_denom3)))


Figure 717. Average DL signal strength, S2 (lev_1a)

Average UL signal strength, S2 (lev_2)

-110+sum(ave_ul_sig_str)/sum(power_denom4)


Figure 718. Average UL signal strength, S2 (lev_2)

Average UL signal strength, S2 (lev_2a)

decode(ave_ul_sig_str/power_denom4,0,’< -110’,63,’> -48’,(-110+(round(ave_ul_sig_str/power_denom4)-1))||’..’||(-110+round(ave_ul_sig_str/power_denom4)))


Figure 719. Average UL signal strength, S2 (lev_2a)



2.43 Distance (dis)

Average MS-BS distance (dis_1)

avg(ave_ms_bs_dist)*550 meter


Figure 720. Average MS-BS distance (dis_1)

The condition below should be applied in order to filter out the hours that do nothave traffic and for that reason show 0:

peak_ms_bs_dist+ave_dl_sig_str +ave_ul_sig_str > 0

Average MS-BS distance (dis_1a)

Counted per BTS/trx-type

decode(trx_type,0, ave_ms_bs_dist*550/1000,1, ave_ms_bs_dist*550/1000 + c_bts.radius_extension)

Unit: km


Figure 721. Average MS-BS distance (dis_1a)

If counted using average, the condition below should be applied in order to filterout the hours that do not have traffic and for that reason show 0:

peak_ms_bs_dist+ave_dl_sig_str +ave_ul_sig_str > 0

MS-BS distance class upper range (dis_3a)

decode(trx_type,0, class_upper_range *550/1000,1, class_upper_range *550/1000 + c_bts.radius_extension)

Unit: km




Figure 722. MS-BS distance class upper range (dis_3a)

2.44 Link balance, power, level (lb)

Link balance, S1 (lb_1)

Known problems: Inaccurate.

avg(ave_dl_sig_str/power_denom3) - avg(ave_ul_sig_str/power_denom4)


Figure 723. Link balance, S1 (lb_1)

Share in acceptance range, S4 (lb_2)

Known problems: The usefulness of link balance measurement is questionable.

sum(normal+ms_limited+bs_limited+max_power){where class_sig_level <= upper threshold

and class_sig_level >= lower threshold }100 * ------------------------------------------------- %

sum(normal+ms_limited+bs_limited+max_power)

Counters from table(s):p_nbsc_link_balance

Figure 724. Share in acceptance range, S4 (lb_2)

Share in normal, S4 (lb_3)


sum(normal)100 * ------------------------------------------------- %



Figure 725. Share in normal, S4 (lb_3)

Share in MS limited, S4 (lb_4)




sum(ms_limited)100 * ------------------------------------------------- %



Figure 726. Share in MS limited, S4 (lb_4)

Share in BS limited, S4 (lb_5)


sum(bs_limited)100 * ------------------------------------------------- %



Figure 727. Share in BS limited, S4 (lb_5)

Share in maximum power, S4 (lb_6)


sum(max_power)100 * ------------------------------------------------- %



Figure 728. Share in maximum power, S4 (lb_6)

Average MS power attenuation, S2 (lb_7)

2*sum(ave_MS_power)/sum(power_denom1)

Counters from table(s):p_nbsc_powerUnit = dB

Figure 729. Average MS power attenuation, S2 (lb_7)

Average MS power, S2 (lb_7b)

avg(decode(o_bts.freq_band_in_use,0,43,1,30)-2*ave_MS_power/power_denom1




p_nbsc_powerUnit= dBm

Figure 730. Average MS power, S2 (lb_7b)

Average UL signal strength, S2 (lb_9)

Use: Values as defined by GSM 5.08. Values 0 to 63:0 = less than -110 dBm1 = -110 to -109 dBm3 = -109 to -108 dBm62 = -49 to -4863 = greater than -48

sum(ave_ul_sig_str)/sum(power_denom4)


Figure 731. Average UL signal strength, S2 (lb_9)

Average DL signal strength, S2 (lb_10)

Use: Values as defined by GSM 5.08. Values 0 to 63:0 = less than -110 dBm1 = -110 to -109 dBm3 = -109 to -108 dBm62 = -49 to -4863 = greater than -48

sum(ave_dl_sig_str)/sum(power_denom3)


Figure 732. Average DL signal strength, S2 (lb_10)

Average MS power attenuation, S2 (lb_11)

2*sum(ave_MS_power)/sum(power_denom1)

Counters from table(s):p_nbsc_powerUnit = dB

Figure 733. Average MS power attenuation, S2 (lb_11)



Average BS power attenuation, S2 (lb_12)

2*sum(ave_BS_power)/sum(power_denom2)

Counters from table(s): p_nbsc_powerUnit = dB

Figure 734. Average BS power attenuation, S2 (lb_12)

Average link imbalance, S2 (lb_13)

2*sum(ave_BS_power)/sum(power_denom2)+sum(ave_dl_sig_str)/sum(power_denom3)-2*sum(ave_MS_power)/sum(power_denom1)-sum(ave_ul_sig_str)/sum(power_denom4)


Unit = dB

Figure 735. Average link imbalance, S2 (lb_13)

2.45 Call success (csf)

SDCCH access probability, before FCS (csf_1)

Use: Gives the probability to access SDCCH without the effect ofFCS. Applicable for area and BTS level.

Known problems: 1) The momentary SDCCH blocking phenomenon is met insome networks.2) sdcch_busy_att triggered also in the case of HO attemptif there are no free SDCCH.

sdcch_busy_att100*(1- ---------------) %

sdcch_seiz_att


Figure 736. SDCCH access probability, before FCS (csf_1)

SDCCH access probability (csf_1a)

Use: Gives the probability to access SDCCH. Applicable for areaand BTS level. A low value means high traffic on SDCCH andlack of SDCCH resources, that is SDCCH blocking.



Experiences on use: The value should be kept very close to 100% in a networkwith traffic. -After S7 the Dynamic SDCCH Allocation can be used toprevent SDCCH congestion. -Before S7 the FACCH call setup could already be used toimprove this KPI.

Known problems: 1) The momentary SDCCH blocking phenomenon is met insome networks and if the traffic is low, this KPI can showlower values that, however, do not mean bad access to thenetwork perceived by the user.2) SDCCH_busy_att triggered also in the case of HO attemptif there are no free SDCCH.3) The formula does not separate calls from LU and other use.In some cases this would be needed (e.g. a train crossing LAboundary creates a high LU peak).

100-blck_5a =

sdcch_busy_att- tch_seiz_due_sdcch_con100-(100* ----------------------------------------) %

sdcch_seiz_att


Figure 737. SDCCH access probability (csf_1a)

SDCCH success ratio (csf_2a)

Experiences on use: The best values seen are around 98%.Known problems: The formula does not separate the SDCCH call seizures from

other seizures (such as LU). The failure rate in the case of acall or LU can greatly differ from one another, wherefore youcannot use this formula for SDCCH call success ratiocalculation.It is not exactly known how large a share ofsum(sdcch_abis_fail_call +sdcch_abis_fail_old) really are setup failures.

100 - non abis SDCCH drop ratio =

sum(sdcch_radio_fail+ sdcch_rf_old_ho+ sdcch_user_act+ sdcch_bcsu_reset+ sdcch_netw_act+ sdcch_bts_fail+ sdcch_lapd_fail+ sdcch_a_if_fail_call+ sdcch_a_if_fail_old)

100 - 100 * ------------------------------------------------------------ %sum(sdcch_assign + sdcch_ho_seiz)

- sum(sdcch_abis_fail_call + sdcch_abis_fail_old) ; phantoms




Figure 738. SDCCH success ratio (csf_2a)

SDCCH success ratio, area (csf_2e)

Use: Indicates how well the SDCCH phase is completed.Experiences on use: The best values seen are around 98%.Known problems: The formula does not separate the SDCCH call seizures from

other seizures (such as LU and SS). The failure rate in the caseof a call or for example LU can greatly differ from oneanother, wherefore you cannot use this formula for SDCCHcall success ratio calculation.

100 - SDCCH drop ratio =

sum(a.sdcch_radio_fail+ a.sdcch_rf_old_ho+ a.sdcch_user_act+ a.sdcch_bcsu_reset+ sdcch_netw_act+ a.sdcch_bts_fail+ a.sdcch_lapd_fail+ a.sdcch_a_if_fail_call+ a.sdcch_a_if_fail_old+ a.sdcch_abis_fail_old+(a.sdcch_abis_fail_call - C))

100 – 100 * ----------------------------------- %sum(b.succ_seiz_term

+ b.succ_seiz_orig+ b.sdcch_call_re_est+ b.sdcch_loc_upd+ b.imsi_detach_sdcch+ b.sdcch_emerg_call

C= part of sdcch_abis_fail_call that occurs before establishment indication =a.sdcch_assign - (b.succ_seiz_term

+ b.succ_seiz_orig+ b.sdcch_call_re_est+ b.sdcch_loc_upd+ b.imsi_detach_sdcch+ b.sdcch_emerg_call)


Figure 739. SDCCH success ratio, area (csf_2e)

SDCCH success ratio, BTS, S6 (csf_2g)

Use: BTS level.Experiences on use: Includes A interface blocking!



Note

Known problems: As consistency is a critical property in measurements, thecombining of three tables can lead into problems.

sum(a.tch_norm_seiz) ;(all TCH seiz.for new call)=100* -------------------------------------------------------------------- %

sum(b.succ_seiz_term+b.succ_seiz_orig+b.sdcch_call_re_est+b.sdcch_emerg_call);(calls,sms, ss reqs)

- sum(b.succ_sdcch_sms_est+ b.unsucc_sdcch_sms_est) ;(sms attempts)

+ sum(c.msc_i_sdcch + c.bsc_i_sdcch ;(net SDCCH HO in)-c.msc_o_sdcch - c.bsc_o_sdcch) ;(unknown how big part calls

- sum(a.tch_call_req-a.tch_norm_seiz) ;(DR and air itf blocking)

- supplem.serv. requests ;(unknown factor)- call clears before TCH ;(unknown factor)- supplem.serv. requests ;(unknown factor)


Figure 740. SDCCH success ratio, BTS, S6 (csf_2g)

This formula includes also A interface blocking. If call re-establishment occursalready on SDCCH, the formula is not correct, but if it occurs on TCH, it iscorrect.

SDCCH success ratio, BTS (csf_2i)

Use: BTS level.Experiences on use: Includes A interface blocking!Known problems: 1)As consistency is a critical property in measurements, the

combining of three tables can lead into problems. Unknownfactors in the denominator make the values seem pessimistic2)The formula does not separate the SDCCH call seizuresfrom other seizures (such as LU and SS). The failure rate inthe case of a call or for example LU can greatly differ fromone another. For this reason you cannot use this formula forSDCCH call success ratio calculation.


sum(b.succ_seiz_term+b.succ_seiz_orig+b.sdcch_call_re_est+b.sdcch_emerg_call);(calls,sms, ss reqs)



- sum(c.cell_sdcch_tch)+ sum(a.tch_succ_seiz_for_dir_acc);direct access related correction- sum(a.tch_call_req-a.tch_norm_seiz) ;(DR and air itf blocking)



Note

- supplem.serv. requests ;(unknown factor)- call clears before TCH ;(unknown factor)- supplem.serv. requests ;(unknown factor)


Figure 741. SDCCH success ratio, BTS (csf_2i)

This formula includes also A interface blocking. If call re-establishment occursalready on SDCCH, the formula is not correct, but if it occurs on TCH, it iscorrect.

SDCCH success ratio, area, S10.5 (csf_2m)

Use: Used on the area level.Experiences on use: The best values seen are around 95%. Includes A interface

blocking!Known problems: As consistency is a critical property in measurements, the

combining of three tables can lead into problems. Unknownfactors in the divisor make the values seem pessimistic.1) The calls are cleared before TCH can vary betweennetworks depending on the call setup time which, again, maydepend on the use of DR or queuing features. Other reasonscan be authentication fails, identity check fails and MOC callshaving wrong dialling, for example.2) This formula does not count correctly the situation whenthe first call or call re-establishment fails on SDCCH (MSnever comes to TCH).3) For the BTS area there is no way of knowing how muchSDCCH-SDCCH handovers take place across the area border.The net incoming amount of SDCCH handovers ends up in asuccessful case in TCH seizures but they are not seen in thenominator.4) For the BTS area there is no way of knowing how muchSDCCH-TCH (DR) handovers take place across the areaborder. The net incoming amount of SDCCH-TCH (DR)handovers ends up in a successful case in TCH seizures butthey are not seen in the nominator.5) LCS requests that are made on SDCCH (MS is idle)increment the denominator but not the numerator, thusmaking the ratio seem too pessimistic. A new counter isneeded to fix the formula for this.

(succ tch seiz) - (call re-establ.)



Note

100 * ------------------------------------------------------- %(sdcch seizures for new calls) - (blocked calls)

sum(a.tch_norm_seiz) ;(all TCH seiz. for new call)- call re-establ. (unknown factor)

= 100 * ------------------------------------------------------------------- %sum(b.succ_seiz_term

+ b.succ_seiz_orig+ b.sdcch_emerg_call+ b.sdcch_call_re_est+ b.call_assign_after_sms) ;(calls,sms, ss reqs)

- sum(b.succ_sdcch_sms_est + b.unsucc_sdcch_sms_est);(sms attempts)- sum(c.bsc_o_sdcch_tch - c.msc_o_sdcch_tch - c.cell_sdcch_tch) ; (DR)+ sum(a.tch_succ_seiz_for_dir_acc ) ; direct access- sum(b.succ_seiz_supplem_serv) ;(supplementary service requests, S9)- call clears before TCH (unknown factor)+ net impact of SDCCH-SDCCH HO on area(unknown factor)+ net incoming DR to area (unknown factor)


Figure 742. SDCCH success ratio, area (csf_2m)

This formula includes also A interface blocking. It works for call re-establishmentif the drop occurs on TCH. If the drop occurs on SDCCH and the call is re-established, there is double count in the divisor.

SDCCH success ratio, BTS, S10.5 (csf_2n)

Use: BTS level.Experiences on use: Includes A interface blocking!Known problems: As consistency is a critical property in measurements, the

combining of three tables can lead into problems. Unknownfactors in the divisor make the values seem pessimistic.1) The calls are cleared before TCH can vary betweennetworks depending on the call setup time which, again, maydepend on the use of DR or queuing features. Other possiblereasons are authentication fails, identity check fails and MOCcalls having wrong dialling, for example.2) This formula does not count correctly the situation wherethe first call or call re-establishment fails on SDCCH (MSnever comes to TCH).3) It is not known how much of SDCCH-SDCCH handoversare calls.4) LCS requests that are made on SDCCH (MS is idle)increment the denominator but not the numerator, thusmaking the ratio seem too pessimistic. A new counter isneeded to fix the formula for this.



Note


sum(b.succ_seiz_term+b.succ_seiz_orig+b.sdcch_call_re_est+b.sdcch_emerg_call+b.call_assign_after_sms) ;(calls,sms, ss reqs)



- sum(c.cell_sdcch_tch)+ sum(a.tch_succ_seiz_for_dir_acc); direct access related correction- sum(a.tch_call_req-a.tch_norm_seiz) ;(DR and air itf blocking)- sum(b.succ_seiz_supplem_serv) ;supplem.serv. requests (S9)- call clears before TCH ;(unknown factor)- supplem.serv. requests ;(unknown factor)


Figure 743. SDCCH success ratio, BTS (csf_2n)

Includes also A interface blocking. If call re-establishment occurs already onSDCCH, the frmula is not correct, but if it occurs on TCH, it is correct.

SDCCH success ratio, area, S10.5 (csf_2o)

Use: Area level.Experiences on use: The best values seen are around 95%. Includes A interface

blocking!Known problems: As consistency is a critical property in measurements, the

binding of three tables can lead into problems. Unknownfactors in the divisor make the values seem pessimistic.1) The calls cleared before TCH can vary between networksdepending on the call setup time which, again, may depend onthe use of DR or queuing features. Other reasons can beauthentication fails, identity check fails and MOC callshaving wrong dialling, for example.2) This formula does not count correctly the situation whenthe first call of call re-establishment fails on SDCCH (MSnever comes to TCH).



Note

4) For the BTS area there is no way knowing how muchSDCCH-TCH (DR) handovers take place across the areaborder. The net incoming amount of SDCCH-TCH (DR)handovers ends up in a successful case in TCH seizures butthey are not seen in the nominator.5) LCS requests that are made on SDCCH (MS is idle)increment the denominator but not the numerator, thusmaking the ratio seem too pessimistic. A new counter isneeded to fix the formula for this.

(successful tch seizures) - (call re-establ.)100 * ------------------------------------------------- %

(sdcch seizures for new calls) - (blocked calls)

sum(a.tch_norm_seiz) ;(all TCH seiz. for new call)- call re-establ. ;(unknown factor)

= 100 * --------------------------------------------------------------------- %sum(b.succ_seiz_term

+ b.succ_seiz_orig+ b.sdcch_emerg_call+ b.sdcch_call_re_est+ b.call_assign_after_sms) ;(calls,sms, ss reqs)

- sum(b.succ_sdcch_sms_est + b.unsucc_sdcch_sms_est);(sms attempts)- sum(c.bsc_o_sdcch_tch + c.msc_o_sdcch_tch) ;(DR)+ sum(a.tch_succ_seiz_for_dir_acc ) ; direct access- sum(b.succ_seiz_supplem_serv) ;(supplementary service requests, S9)- call clears before TCH ;(unknown factor)+ net impact of SDCCH-SDCCH HO on area ;(unknown factor)+ net incoming DR to area ;(unknown factor)


Figure 744. SDCCH success ratio, area, S10.5 (csf_2o)

Includes also A interface blocking. Works for call re-establishment if the drop ison TCH. If the drop is on SDCCH and the call is re-established, then there isdouble count in the divisor.

TCH access probability without DR (csf_3a)

Use: This PI indicates what would be the blocking if DR was notused. When compared to csf_3, you can see, assuming thatDR is in use, the improvement that the DR has caused.

100-blck_8 =

sum(tch_call_req - tch_norm_seiz)100-100* -------------------------------- %

sum(tch_call_req)




Figure 745. TCH access probability without DR (csf_3a)

TCH access probability without DR and Q (csf_3b)

Use: This PI indicates what would be the TCH blocking if DR andqueuing were not used. When compared to csf_3a, you cansee, assuming that DR is in use, the improvement that the DRhas caused.

Known problems: See XX1.

sum(tch_call_req - tch_norm_seiz)+ sum(tch_qd_call_att - XX1 - unsrv_qd_call_att)

; calls succeeded via queuing100 - 100 * ------------------------------------------------- %

sum(tch_call_req)

Counters from table(s):p_nbsc_trafficXX1 = attempts taken from queue to DR (unknown)

Figure 746. TCH access probability without DR and Q (csf_3b)

TCH access probability without Q (csf_3c)

Use: This PI indicates what would be the blocking if queuing wasnot used (but DR is used).

sum(a.tch_call_req- a.tch_norm_seiz- b.msc_o_sdcch_tch- b.bsc_o_sdcch_tch)

+ sum(a.tch_qd_call_att- a.unsrv_qd_call_att) ;calls that succeeded via queuing

100 - 100 * ------------------------------------------------------------- %sum(a.tch_call_req)


Figure 747. TCH access probability without Q (csf_3c)

TCH access probability, real (csf_3d)

Use: This KPI is affected by the congestion on TCH.100 - blck_8b =

sum(a.tch_call_req - a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch

+ b.bsc_o_sdcch_tch+ b.cell_sdcch_tch); DR calls



100 - 100 * ------------------------------------------ %sum(a.tch_call_req)


Figure 748. TCH access probability, real (csf_3d)

TCH access probability without DR (csf_3i)

Use: This PI indicates what would be the TCH blocking if DR wasnot used. When compared to csf_3a, you can see, assumingthat DR is in use, the improvement that the DR has caused.Does not contain the congestion of the A interface circuitpool.

sum(a.tch_call_req - a.tch_norm_seiz)- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool HO failures+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type))

100 - 100 * ------------------------------------- %sum(a.tch_call_req)

- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool HO failures+ b.msc_controlled_in_ho+ b.ho_unsucc_a_int_circ_type))


Figure 749. TCH access probability without DR (csf_3i)

TCH access probability without DR and Q (csf_3j)

Use: This PI indicates what would be the TCH blocking if DR andqueuing were not used. When compared to csf_3a, you cansee, assuming that queuing is in use, the improvement that thequeuing has caused. Does not contain the congestion of the Ainterface circuit pool.

Known problems: See XX1.

sum(tch_call_req - tch_norm_seiz)+ sum(tch_qd_call_att-XX1-unsrv_qd_call_att) ;succ. calls via queuing- sum(tch_rej_due_req_ch_a_if_crc) ; Aif pool rejections

100-100*-------------------------------------------------------- %sum(tch_call_req)

- sum(tch_rej_due_req_ch_a_if_crc) ; Aif pool rejections




p_nbsc_trafficXX1 = attempts taken from queue to DR (unknown)

Figure 750. TCH access probability without DR and Q (csf_3j)

TCH access probability, real (csf_3k)

Use: This KPI is affected by the blocking on TCH.

100-blck_8c =

sum(a.tch_call_req-a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch+ b.bsc_o_sdcch_tch+b.cell_sdcch_tch); DR calls- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool HO failures+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type))

100-100* ----------------------------------------------------------- %sum(a.tch_call_req)

- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool HO failures+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type))


Figure 751. TCH access probability, real (csf_3k)

TCH access probability, real (csf_3l)

Use: See blck_8d.Known problems: On cell level the formula is inaccurate in case of inter cell

direct access (BSS7057).

100-blck_8d =

sum(a.tch_call_req-a.tch_norm_seiz)- sum(b.msc_o_sdcch_tch+ b.bsc_o_sdcch_tch+b.cell_sdcch_tch); DR calls+ sum(a.tch_succ_seiz_for_dir_acc) ;ref.2- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion

-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool HO failures+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type))

100-100* ----------------------------------------------------------- %sum(a.tch_call_req)

- sum(a.tch_rej_due_req_ch_a_if_crc ; Aif type mismatch or congestion-(b.bsc_i_unsucc_a_int_circ_type ; Aif circuit pool HO failures+b.msc_controlled_in_ho+b.ho_unsucc_a_int_circ_type))


Figure 752. TCH access probability, real (csf_3l)



Ref.2. Compensation needed since in case of Direct Access to super reuse TRXthe tch_norm_seiz is triggered in parallel with the cell_sdcch_tch.

TCH access probability without DR and Q (csf_3m)

Use: This PI indicates what would be the TCH blocking if DR andqueuing were not used. When compared to csf_3i, you cansee, assuming that queuing is in use, the improvement that thequeuing has caused. Does not contain the congestion of the Ainterface circuit pool.

Known problems: Inaccurate when the feature 'TCH assignment to super-reuseTRX in IUO' is applied. In this case tch_call_req andremoval_from_que_due_to_dr are triggered multipletimes if the target cell is congested and queuing is started butthe call is removed to normal DR.

sum(tch_call_req - tch_norm_seiz)+ sum(tch_qd_call_att

-removal_from_que_due_to_dr-unsrv_qd_call_att) ;succ. calls via queuing

- sum(tch_rej_due_req_ch_a_if_crc); Aif pool rejections100-100 * -------------------------------------------------------- %

sum(tch_call_req)- sum(tch_rej_due_req_ch_a_if_crc); Aif pool rejections


Figure 753. TCH access probability without DR and Q (csf_3m)

TCH success ratio, area, before call re-establisment (csf_4o)

Use: Used on the area level. The impact of call re-establishment isnot yet taken into account.



100 - 100 * ------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)


+ sum(a.tch_seiz_due_sdcch_con) ; FACCH call setup calls





Figure 754. TCH success ratio, area, before call re-establisment (csf_4o)

TCH success ratio, area, after call re-establishment, S6 (csf_4p)

Use: Used on the area level.Known problems: 1) See dcr_3g.

2) It is assumed that call re-establishments happen on TCH. Infact they may happen also on SDCCH.3) The counters used to compensate re-establishments are theones that indicate re-establishment attempts, not thesuccessful re-establishments. In S7/T11 re-establishments canbe considered accurately (see csf_4v).4) On cell level it can happen that the call is re-established ina different cell than where it was dropped, which results ininaccuracy.

100 - dcr_3f =


- sum(b.sdcch_call_re_est + b.tch_call_re_est) ;call re-establ.100 - 100 * ------------------------------------------------------------------- %



+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls- sum(b.sdcch_call_re_est+b.tch_call_re_est) ;call re-establ.


Figure 755. TCH success ratio, area, after call re-establishment, S6 (csf_4p)

TCH success ratio, BTS, before call re-establisment (csf_4q)

Use: Used on the BTS level.Known problems: See dcr_4c.



100 - dcr_4b =


100 - 100 * -------------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)


+ sum(a.tch_seiz_due_sdcch_con) ; FACCH call setup calls+ sum(c.msc_i_tch_tch+c.bsc_i_tch_tch) ;(TCH-TCH HO in)


Figure 756. TCH success ratio, BTS, before call re-establisment (csf_4q)

TCH success ratio, BTS, after call re-establishment (csf_4r)

Use: Used on the BTS level.Known problems: See dcr_3g.


- sum(b.sdcch_call_re_est + b.tch_call_re_est) ;call re-establ.100 - 100 * -------------------------------------------------------------------- %


+ c.bsc_i_sdcch_tch+ c.cell sdcch_tch) ;(DR calls)

+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls+ sum(c.msc_i_tch_tch

+ c.bsc_i_tch_tch) ;(TCH-TCH HO in)- sum(b.sdcch_call_re_est

+ b.tch_call_re_est) ;call re-establishments




b = p_nbsc_res_accessc = p_nbsc_ho

Figure 757. TCH success ratio, BTS, after call re-establishment (csf_4r)

TCH success ratio, BTS, after call re-establishment (csf_4t)

Use: Used on the BTS level.Known problems: See dcr_3d.


- sum(b.tch_re_est_assign) ;call re-establishments100 - 100 * ----------------------------------------------------------- %


+ c.bsc_i_sdcch_tch+ c.cell sdcch_tch) ;(DR calls)

+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch) ;(TCH-TCH HO in)- sum(b.tch_re_est_assign) ;call re-establ.


Figure 758. TCH success ratio, BTS, after call re-establishment (csf_4t)

TCH success ratio, area, before call re-establishment, S7(csf_4u)

Use: See dcr_3i.Known problems: See dcr_3g. The impact of call re-establishment is not yet

taken into account.

100 - dcr_3i =

sum(a.tch_radio_fail+ a.tch_rf_old_ho+ a.tch_abis_fail_call++ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail



+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

100 - 100 * ---------------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)


- sum(a.tch_succ_seiz_for_dir_acc ) ;ref.2+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls


Figure 759. TCH success ratio, area, before call re-establishment, S7(csf_4u)


TCH success ratio, area, after call re-establishment, S7 (csf_4v)

Use: On the area level. See dcr_3j.Known problems: 1) It is assumed that call re-establishments happen on TCH. In

fact they may happen also on SDCCH.2) On cell level it can happen that the call is re-established ina different cell than it was dropped and this causes inaccuracy.

100 - dcr_3j=


- sum(b.tch_re_est_assign) ;call re-establishments100 - 100 * ----------------------------------------------------------- %



- sum(a.tch_succ_seiz_for_dir_acc) ;ref.1+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls- sum(b.tch_re_est_assign) ;call re-establishments




b = p_nbsc_servicec = p_nbsc_ho

Figure 760. TCH success ratio, area, after call re-establishment, S7 (csf_4v)


TCH success ratio, BTS, after call re-establishment (csf_4x)



- sum(b.tch_re_est_assign) ;call re-establishments100 - 100 * ---------------------------------------------------------------- %



- sum(a.tch_succ_seiz_for_dir_acc ) ;ref.2+ sum(a.tch_seiz_due_sdcch_con) ;FACCH call setup calls+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch) ;(TCH-TCH HO in)- sum(b.tch_re_est_assign) ;call re-establishments


Figure 761. TCH success ratio, BTS, after call re-establishment (csf_4x)


TCH success ratio, BTS, before call re-establishment (csf_4y)


100 - dcr_4e=

sum(a.tch_radio_fail+ a.tch_rf_old_ho



+ a.tch_abis_fail_call+ a.tch_abis_fail_old+ a.tch_a_if_fail_call+ a.tch_a_if_fail_old+ a.tch_tr_fail+ a.tch_tr_fail_old+ a.tch_lapd_fail+ a.tch_bts_fail+ a.tch_user_act+ a.tch_bcsu_reset+ a.tch_netw_act+ a.tch_act_fail_call)

100 - 100 * --------------------------------------------------- %sum(a.tch_norm_seiz) ;(normal calls)

+ sum(c.msc_i_sdcch_tch+ c.bsc_i_sdcch_tch+ c.cell sdcch_tch) ;(DR calls)

- sum(a.tch_succ_seiz_for_dir_acc ) ;ref.2+ sum(a.tch_seiz_due_sdcch_con) ; FACCH call setup calls+ sum(c.msc_i_tch_tch + c.bsc_i_tch_tch) ;(TCH-TCH HO in)


Figure 762. TCH success ratio, BTS, before call re-establishment (csf_4y)


Activation related SDCCH access probability, S7, (csf_12)


sum(sdcch_assign + t3101_expired)100 * --------------------------------- %

sum(served_sdcch_req)


Figure 763. Activation related SDCCH access probability, S7, (csf_12)

SDCCH call success probability, S10.5 (csf_13a)


sum(a.tch_call_req)100 * ------------------------------ %

sum(c.call_assign_after_sms+ a.sdcch_new_call_assign+ b.sdcch_ho_call_assign+ a.sdcch_re_est_assign- a.sdcch_re_est_release- a.sdcch_sms_assign- a.sdcch_ho_rel.)




p_nbsc_service, a = source, b = targetc = p_nbsc_res_access

Figure 764. SDCCH call success probability, S10.5 (csf_13a)

2.46 Configuration (cnf)

Reuse pattern (cnf_1)

Experiences on use: For example, 30/3 = 10 means that the frequency can berepeated with 10 cells!The smaller the figure, the better the planning.

Known problems: This indicator can be counted from the Nokia NetAct only forthe latest moment (no history).

number of used frequencies------------------------average TRXs per cell

Figure 765. Reuse pattern (cnf_1)

The used frequency means that the TRX and its parents (BTS and BCF) areunlocked.

Reuse pattern, S1 (cnf_2)

Decode(Avg(trx_type),0,’normal’,1,’extended’,’mixed’)

Figure 766. Reuse pattern, S1 (cnf_2)




Missing Counters

3 Missing CountersThis chapter handles counters that are found to be needed in the formulas but aremissing or scheduled for later BSC software releases.

3.1 XX1

Meaning: How many calls were taken from the queue to DR.Related problem: If a call attempt is in queue, it is taken to DR as soon as the

DR target list is ready. In this case counters show as if thequeuing took place: tch_qd_call_att is triggered butunsrv_qd_call_att is not.

Planned schedule: S8 (BSC name: REM_FROM_QUEUE_DUE_DR, counter ID1173)

3.2 XX2

Meaning: Clear by MS user during HO procedure.Related problem: Affects the counting of HO failure ratios. The ratio cannot be

counted accurately (see hfr_2).Planned schedule: Open.

3.3 XX3

Meaning: Clear by another procedure during the HO procedure, forexample assignment.

Related problem: Affects the counting of HO failure ratios. The ratio cannot becounted accurately (see hfr_2).

Planned schedule: Open.



3.4 XX4

Meaning: The number of available timeslots for TCH (half or full rate).Related problem: Without this counter it is not possible to count the TCH

availability in case HTCH is used.Planned schedule: S9.


Index

DN98619493 ©Nokia Corporation 323 (324)Issue 2-5 en Nokia Proprietary and Confidential



BSS Network Doctor Formulas

Documents

rlc data blocks ul cs1 20 rlc data blocks ul cs2 30 rlc data blocks dl cs1 20 rlc data blocks dl cs2 30

req 1 tsl ul 2 req 2 tsl ul 3 req 3 tsl ul 4 req 4 tsl ul

alloc 1 tsl dl 2 alloc 2 tsl dl 3 alloc 3 tsl dl 4 alloc 4 tsl dl

req 1 tsl dl 2 req 2 tsl dl 3 req 3 tsl dl 4 req 4 tsl dl

alloc 1 tsl ul 2 alloc 2 tsl ul 3 alloc 3 tsl ul 4 alloc 4 tsl ul

req 1 tsl dl req 2 tsl dl req 3 tsl dl req 4 tsl dl req 5 8 tsl dl

req 1 tsl ul req 2 tsl ul req 3 tsl ul req 4 tsl ul req 5 8 tsl ul

ho fail to reg due ret ho fail to reg ms lost ho fail to reg freq