GSM TCH Allocation Failure & Solutions

7/23/2019 GSM TCH Allocation Failure & Solutions

http://slidepdf.com/reader/full/gsm-tch-allocation-failure-solutions 1/19

Internal Use Only▲

GSM TCH Allocation Failure & Solutions

Special Subject Guidebook

Version: V 2.0

Released by:

GSM Network Planning & Optimiation !ept.

"ngineering Ser#i$e !i#ision

%" 'orporation

This document contains proprietary information of ZTE Corporation and is not to be disclosed or used

except in accordance with applicable agreements.




GSM Network Planning & Optii!ation "ept#

Special Subject Guidebook

$ersion introduction%

Version Date Writer Assessor Amendment records

V.! "!!#$"$" %i %un &an Wen'un

V".! "!!($)$ %i %un &an Wen'un Counters rele*ant to TC+ allocation of V"

and V) were added,

The document template was renewed.






e' words%

TC+ allocation

Abstract%

This guideboo- mainly describes TC+ allocation failure and in*estigation of the

problem.

(e)erences%

《&/ Traffic tatistics》

《Z0&! 1（V".(#） 1C 2erformance Counter /anual》

《Z0&! i1C（V3."!） 1C 2erformance Counter /anual 456》





Contents1 TCH Allocation Failure & Investigation...........................................................................................1

. Definition of TC+ allocation failure.................................................................................................

." /ain causes of TC+ allocation failure............................................................................................."

.) 2roblem handling procedures7...........................................................................................................8

2 Relevant Counters& KPI....................................................................................................................6

". Z0&!$1（V".(#） counters......................................................................................................3

".. C3!( 9total number of attempts for TC+ 4excl. +:6.........................................................3

".." C3! 9total number of unsuccessful assignments due to TC+ o*erflow 4excl. +:6.........3

"..) C3;) 9total number of successful TC+ assignments 4excl.+:6........................................#

"..8 C3;8 9total number of TC+ allocation failures 4excl.+:6.................................................#

"..; C3;< 9total number of unsuccessful TC+ assignments.....................................................#

"." Z0&!$51C（V3."!） counters....................................................................................................#

".) Definition of TC+ assignment success rate......................................................................................#

Ty!ical Cases........................................................................................................................................"

). =ow TC+ allocation success rate in a cell........................................................................................(

)." Antenna connection problem............................................................................................................(

).) 5nterference within networ-............................................................................................................

).8 Discrepancy in time at 1C and :/C>........................................................................................"





Figure Contents?ig $ TC+ allocation flow...................................................................................................................."

?ig $" ?low for handling TC+ assignment failure.................................................................................;

?ig )$ ignaling trace.............................................................................................................................!

?ig )$" ignaling trace.............................................................................................................................





Ta#le ContentsTable "$ Definition of TC+ assignment success rate 4excl. +:6...........................................................#

Table )$ Contrast of TC+ allocation success rate indicator before and after problem handling............(

?ig )$" Contrast before and after antenna problem is sol*ed.................................................................

Table )$) Contrast of number of TC+ assignment failures...................................................................."






* TCH Allocation Failure & +n,estigation*#* "e)inition o) TCH allocation )ailure

When / fails to access into TC+ after it sends out re@uest for TC+ allocation to

1C and it must be /:C or /TC to the / excluding +: 4except for intra$cell

+:6 we can regard this situation as TC+ allocation failure.

TC+ allocation failure rate is calculated as follows7

Bumber of busy hour TC+ allocation failures 4excl. +:6 number of busy hour TC+call attempts 4excl. +:6!!

FBumber of busy hour TC+ call attempts 4excl. +:6G refers to the number of TC+

call attempts after / occupies DCC+. The statistics counted is number of

FAssignment >e@uestG sent from /C to 1C.

FBumber of busy hour TC+ allocation failures 4excl. +:6G refers to the number of

calls when / fails to occupy TC+ after it seiHes DCC+. The statistics gathered is

FAssignment ?ailure I Clear >e@uest4radio interface failure6”. Allocation failure may

be caused due to lac- of TC+s or problems on lower layers.

TC+ allocation flow7


except in accordance with applicable agreements. 1




?ig $ TC+ allocation flow

Assignment ?ailure and Assignment >e@uest correspond with each other they reflect

one time of TC+ allocation failure. As for situation 46 shown in ?ig $ TC+

assignment flow the reason in most cases is the lac- of idle TC+s, as for situation 4"6

TC+ assignment failure is caused by 1T problems, as for situation 4)6 it mainly

refers to TC+ assignment failure at air interface which can be caused by co*erage and

interference etc

*#- Main causes o) TCH allocation )ailure

Common causes of TC+ allocation failure are as follows7

Traffic congestion in cells

5f traffic congestion rate is high in cells when / applies for TC+ the system

will find there are no TC+s a*ailable for allocation which causes allocation

failure.

+ardware problem






Transcei*er problem7 TC+ allocation failure rate is usually high when something

wrong with T>0 so does incoming +: failure rate because 1C also assigns

TC+ to / during incoming +:. 5f TC+ allocation failure rate exceeds ! in a

cell the most probably cause will be T>0 problem. As for cells li-e this in order

to locate the T>0 with problem we can record the signaling at Abis interface and

through signaling analysis find out the specific T>0 which causes allocation

failure.

Combiner problem7 li-e forward power output problem.

Co$channel or ad'acent channel interference

+igh error rate caused by interference will hinder / from establishing the

second layer lin- with 1T which results in assignment failure.

Antenna feeder problem

1ro-en or eroded feeders can cause high VW> which will affect transcei*er

performance.

Antenna is bloc-ed or co*erage is not e*en$distributed. When antenna carrying

TC+ only is bloc-ed or its co*erage is different from that carrying 1CC+ or

DCC+ / may fail to occupy the TC+.

Jnreasonable setting of parameters

5f ?+ is adopted while +B or /A5: isnKt set correct serious interference may

be resulted such as interference within a cell or co$channel or ad'acent channel

interference between cells with the same /A list which can cause serious

assignment failure rate.

5f T)!# is set with a too small number the networ- will release the channel

before it recei*es FAssignment completeG due to T)!# expiration.

Transmission problem at A interface or Abis interface

5f error rate is high at A interface or Abis interface signaling exchange between

/ and networ- can not be completed normally which will result in assignment

failure as well.

5nfluence of repeater






>adio transmission is usually adopted in external repeater. When the repeater

amplifies the J=D= signals it amplifies interference signals as well hence signal

@uality will deteriorate and call drop will be resulted and TC+ assignment failure

rate will ine*itably increase.

*#. Proble /andling procedures%

Chec-ing radio parameters and hardware is recommended for locating problems.

2rocedures for handling TC+ assignment failure7

46 Chec- if traffic congestion exists in the cell. 5f it does sol*e the problem through

methods li-e expanding 1T balancing traffic etc..

4"6 Chec- if cell radio parameters are correctly set such as ?+ parameter fre@uency

data etc.. :ptimiHe those which are set unreasonable.

4)6 Chec- indicators li-e 1E> idle interference band class reduce or eliminate radio

interference.

486 Chec- cell hardware li-e transcei*er combiner di*ider >? cables between

boards change those with problems.

4;6 Chec- antenna system li-e VW> whether antennas in one cell radiate in the

same direction antenna feeders are connected falsely re*ersely or not.

?low for handling TC+ assignment failure7






?ig $" ?low for handling TC+ assignment failure






- (ele,ant Counters& P+-#* 01G*234SS $-#56 counters

-#*#* C**725 8total nuber o) attepts )or TCH 9e:cl# HO;

Definition7 5t represents the total number of attempts to ac@uire TC+s 4seiHure

attempts6 after all DCC+s in cells are seiHed. 5t includes /:C and /TC and

*ery early assignment of TC+ as DCC+ but hando*ers are excluded. TC+

falls into TC+? and TC++.

Calculation formula7 C3!( 4total number of attempts for TC+ 4excl. +:66 L

C!)! 4TC+? seiHure attempts4signaling64for assignment66 I C!)"! 4TC+?

seiHure attempts 4speech64for assignment66 I C!); 4TC+? seiHure attempts

4data64for assignment66 I C!8! TC++ seiHure attempts4signaling64for

assignment66 I C!8"! 4TC++ seiHure attempts 4speech64for assignment66 I

C!8; 4TC++ seiHure attempts4data64for assignment66

-#*#- C**7*2 8total nuber o) unsuccess)ul assignents due to TCHo,er)low 9e:cl# HO;

Definition7 5t represents the total number of o*erflows in calls to TC+s after all

DCC+s in cells are seiHed. 5t includes /:C and /TC and the situation of

/TC failing in setting up TC+ after itKs seiHed DCC+. 5t also includes *ery

early assignment of TC+ as DCC+ but hando*ers are excluded. TC+ falls into

TC+? and TC++.

Calculation formula7 C3! 4Total number of unsuccessful assignments due to

TC+ o*erflow 4excl. +:66 L C!)!) 4TC+? seiHure failures 4signaling64for

assignment66 I C!)"" 4TC+? seiHure failures 4speech64for assignment66 I

C!);) 4TC+? seiHure failures 4data64for assignment66 I C!8!) 4TC++

seiHure failures 4signaling64for assignment66 I C!8"" 4TC++ seiHure failures

4speech64for assignment66 I C!8;) 4TC++ seiHure failures 4data64for

assignment66






-#*#. C**7<. 8total nuber o) success)ul TCH assignents 9e:cl#HO;

Definition7 5t represents the number of recei*ed acti*ation response messages

after channels are acti*ated for TC+ assignment 4signaling speech and data6.

Calculation formula7 C3;) L C!)!< I C!)"# I C!))) I C!))(

IC!);<I C!8!< I C!8"# I C!8)) I C!8)( IC!8;<

-#*#= C**7<= 8total nuber o) TCH allocation )ailures 9e:cl#HO;

Definition7 5t represents the number of negati*e replies in response to TC+

acti*ation re@uest for assignment or the number of acti*ation ac-nowledge

which ha*enMt been recei*ed 4signaling *oice and data6.

Calculation formula7 C3;8 L C!)!( I C!)"< I C!))8 I C!)8!

IC!);(I C!8!( I C!8"< I C!8)8 I C!88! IC!8;(

-#*#< C**7<> 8total nuber o) unsuccess)ul TCH assignents

Definition7 5t represents the number of negati*e replies in response to TC+

assignment command 4signaling speech and data6 or the number of unrecei*ed

FEstablish 5ndicationG or FAssignment CompleteG. >efer to chapter ).; TC+?

/easurement for explanations of C!);、C!)83、C!)3; and chapter ).3.TC++ /easurement for explanations of C!8;、C!883、C!83;.

Calculation formula7C3;< L C!); ＋ C!)83 ＋ C!)3;I C!8; I

C!883 IC!83;

-#- 01G*23+4SC $7#-2 counters

>efer to chapter ".) for explanations of the counters.

-#. "e)inition o) TCH assignent success rate

Table "$ Definition of TC+ assignment success rate 4excl. +:6

N25 name TC+ allocation success rate464exclude hando*er6

Definition Bumber of TC+ assignment success times 4excl. +:6number of TC+ assignment

attempts 4excl. +:6

Counter formula V" 4$（C3;8IC3!IC3;<） C3!(6!!






V) V3."

4C(!!!3!!#IC(!!!3!!"<IC(!!!3!!)3IC(!!!3!"); I

C(!!!3!((IC(!!!3!"!6!!4C(!!!3!!!IC(!!!3!!(

IC(!!!3!!)!IC(!!!3!!)<IC(!!!3!!8"IC(!!!3!!836

ZTE TC+ assignment falls into three phases7 seiHure 4congestion6 allocation 4channel

acti*ation6 and assignment.

eiHure 4congestion6 of TC+7 1C searches for idle TC+s in database after it recei*es

A5&B >EOJET. 5f TC+s are a*ailable seiHure success is mar-ed, if not and

@ueuing directed retry and forced release do not exist one time of TC+ congestion is

recorded, if @ueuing directed retry and forced release exist start the corresponding

timer and wait for usable TC+s, if usable TC+s are obtained before the timer expires

seiHure success can be mar-ed, if not one time of TC+ congestion will be recorded.

TC+ allocation 4channel acti*ation67 after succeeding in applying for TC+ from

database 1C sends out Channel Acti*ation to 1T which is TC+ allocation attempt.

When 1C recei*es ChannelActi*ationAc- from 1T allocation is a success. 5f it

recei*es ChannelActi*ationBac- from 1T or Channel Acti*ation Ac- isnMt recei*ed

when the timer expires the allocation is regarded a failure.

TC+ assignment7 after ha*ing recei*ed ChannelActi*ationAc- during TC+ allocation

1C sends out A5&B C://ABD on D= DCC+ and TC+ assignment attempt isrecorded once. When 1C recei*es A5&B C:/2=ETE from 1T the assignment

is regarded a success, if not a failure.






. T'pical Cases.#* ?ow TCH allocation success rate in a cell

【2roblem description】

5ndicator statistics at :/C> showed that TC+ allocation success rate was low

in Cell" under 1T DA.

【2roblem analysis】

Chec-ed radio parameters but no problem was found. ?rom the dynamic

management at :/C> we found that one T>0 in the cell couldnKt be seiHed.

We confirmed the T>0 has problem and deli*ered a report as-ing for chec-ing

of hardware.

【2roblem handling】

After the T>0 board with problem was replaced TC+ allocation success rate

increased greatly and problem was sol*ed.

Table )$" Contrast of TC+ allocation success rate indicator before and after problem handling

Date 1T nameCell location area 4=AC$

C56

Total of

TC+

allocation

re@uests

Total of

TC+

allocation

success

times

TC+

allocation

success

rate46

"!!;$!#$3 <7!!$(7!! DA =AC<83$C5";!# <"; 3;! #<.#(

"!!;$!#$# <7!!$(7!! DA =AC<83$C5";!# <8# 383 #3."#

"!!;$!#$< <7!!$(7!! DA =AC<83$C5";!# <;) 33 ##.8(

After problem was sol*ed

"!!;$!#$"! <7!!$(7!! DA =AC<83$C5";!# <3 <! (<.3

"!!;$!3$" <7!!$(7!! DA =AC<83$C5";!# <"( <# (<.;;

.#- Antenna connection proble







TC+ allocation success rate was low in Cell in an area which was below

(8.!.


?rom analysis of :/C> performance indicators we found the problem was

caused by the low TC+ assignment success rate at air interface. ?rom further

analysis we found the assignment failure rate of T>0) was high 4/aioL"6 and

D= signal was wea- 4wea-er than D= signal of 1CC+ T>0 by about "!d1.6

?ig )$) ignaling trace

?ig )$8 ignaling trace







+ardware engineers went to the site and chec-ed finding that antenna system

was connected wrong 4the T>0) of Cell was connected to the antenna of

Cell".6

?ig )$) Contrast before and after antenna problem is sol*ed

Bame Cell location area 4=AC$C56 Date

Total of TC+

allocation

re@uests

4excl.+:6

Total of TC+

allocation success

times4excl.+:6

TC+ allocation

success rate46

4excl.+:6

;"< =AC)#3<)$C5;"< "!!#$$"" <8 #38 ().<3

;"< =AC)#3<)$C5;"< "!!#$$") <"# #;" (.!;

;"< =AC)#3<)$C5;"< "!!#$$"8 #3 # ().;3

;"< =AC)#3<)$C5;"< "!!#$$"; #8" 3(# (8.!#

After problem was sol*ed

;"< =AC)#3<)$C5;"< "!!#$$"# 3#! 33) (<.(3

;"< =AC)#3<)$C5;"< "!!#$$"< 3!# ;(3 (<.);

.#. +nter)erence wit/in network


2erformance statistics showed that TC+ assignment failure rate was high in

Cell" and Cell) which led to low TC+ allocation success rate.


As for this -ind of problem we usually chec- if co$channel or ad'acent channel

interference exists. ?rom the planning software we could see the two cells were

right opposite each other and their o*erlapped co*erage area was large and co$

channel existed 4fre@uency ( in Cell" shared fre@uency with Cell)6 thus

serious interference was ine*itable and TC+ assignment failure rate was high

as well.


After we changed fre@uency point in Cell) from ( to ! number of TC+

assignment failures in the two cells reduced greatly.






Table )$8 Contrast of number of TC+ assignment failures

Cell name 1T 5DCell location area

4=AC$C56Date

Total of

TC+

allocation

re@uests

4excl.+:6

Bumber of

TC+

assignment

failures

Cell" 1sc)$ite);$1ts" =AC38(<$C5;)(" "!!3$"$< (( #

Cell" 1sc)$ite);$1ts" =AC38(<$C5;)(" "!!3$"$( !" #

Cell" 1sc)$ite);$1ts" =AC38(<$C5;)(" "!!3$"$! (# ;

Cell" 1sc)$ite);$1ts" =AC38(<$C5;)(" "!!3$"$ <) "

Cell" 1sc)$ite);$1ts" =AC38(<$C5;)(" "!!3$"$" #) !

Cell" 1sc)$ite);$1ts" =AC38(<$C5;)(" "!!3$"$) 3( !

Cell) 1sc)$ite)!3$1ts) =AC38(<$C5;)!<) "!!3$"$< )8; (

Cell) 1sc)$ite)!3$1ts) =AC38(<$C5;)!<) "!!3$"$( 8!< 8

Cell) 1sc)$ite)!3$1ts) =AC38(<$C5;)!<) "!!3$"$! )<3 (

Cell) 1sc)$ite)!3$1ts) =AC38(<$C5;)!<) "!!3$"$ )"(

Cell) 1sc)$ite)!3$1ts) =AC38(<$C5;)!<) "!!3$"$" )8(

Cell) 1sc)$ite)!3$1ts) =AC38(<$C5;)!<) "!!3$"$) )33

.#= "iscrepanc' in tie at 4SC and OMC(


A customer reported that call completion rate exceeded !! almost

reached;.


Analysis of basic measurement data7D"83LC3!($

4C3!IC3;8IC3;<6 ＝ 8(;;times ＝ sum of Call2roceeding and

CallConfirm measured at /C which was 8(;;. um of Call2roceeding and

CallConfirm measured by the customer was (8"<. While the number of TC+

assignment success 4excl. +:6 4D"836 shown in the performance report was

)<8# and #8"#. After chec-ing statistics we found that )<8# and #8"# were

counted in only three granularities while usually statistics of four granularities

should be gathered 4;mins per granularity6. We also found that time at 1C

was ;mins )!secs behind the time at :/C>. o when renewing data ;mins






past integral hour only data in three granularities could be renewed since the

data of the fourth granularity hasnMt been created yet.

Through signaling at interface A we found that the sum of Call2roceeding and

CallConfirm was much larger than the number of TC+ assignment success

counted at :/C> of the same period. 1ecause in the formula of switching

system completion rate the denominator consists of the number of TC+

assignment success in this case it became smaller then increase in the

completion rate was ine*itable e*en exceeded !!.


We ad'usted time at :/C> and renewed data of :/C> with 2/D> and the

data in :/C> and signaling de*ice conformed to each other.



GSM TCH Allocation Failure & Solutions

Documents