CDMA Basic

Pan American Wir eless Infrastructure DivisionFt. Lauderdale Center for Excellence

CDMA Network OperationsPrepared & Presented by Antonio Shappley

Motorola Confidential Proprietary 3 of 259

PWID Training: CDMAForeword

PWID Center for Excellence Revision 0, 4/24/97 Foreword

4 of 259 Motorola Confidential Proprietary

Foreword

• The objective for this course is to present the Super-Cell CDMA system network operations in detail, inorder to prepare Motorola Personnel for placing aCDMA system in operation.

• It is recommended that students take the followingcourses (available from TED) in preparation for thiscourse:

- GNL180: IS-95 Implementation & Operation

- GNL190: CDMA Call Processing

- CNL070: SC9600 Overview (CDMA)

• Most of this manual presentsMotorola ConfidentialProprietary information, so care must be taken bythe student to comply with POPI regulations at alltimes.

• This course was prepared by Antonio Shappley ofthe Training & Certifications department, based outof Ft. Lauderdale, Florida.

• Some material for this course has been taken fromother groups’ documentation- please see theAcknowledgements section for the names of thesegroups and/or individuals.


Table of Contents:Foreword 4Introduction to CDMA Theory & IS-95 Standard 7An Introduction to the SuperCell System 37The Tandem MM Platform & UNIX OS 73The OMC-R and UNO Platforms 111The Transcoder Platform 131The SS7 Sub-System: MM and MSC 163The SC9600 BTS 181Output from Work Sessions 193CDMA Call Processing 207Tricks, Tips & Tools for Unix and CDMA 237

PWID Center for Excellence Revision 0, 4/24/97


Page Intentionally left blank.


PWID Training: CDMAIntroduction to CDMA Theory & IS-95 Standard


Objectives:

• To present a basic introduction of the CDMA technol-ogy and the IS-95 standard, as it pertains to Motor-ola’s SuperCell product.


Table of Contents:

Fundamental Statements................................................................................................ 11Why CDMA?................................................................................................................... 13Fundamental Comparisons............................................................................................ 15CDMA Basic Concepts ................................................................................................... 17Digital Voice Coding Concepts....................................................................................... 19Basic Digital Signaling Concepts- Isomorphism & Spreading Sequences .................. 21Multiple Conversations on One Channel? .................................................................... 23Forward CDMA channels .............................................................................................. 25Reverse Link Channels................................................................................................... 27IS-95 Forward Link Encoder ......................................................................................... 29IS-95 Reverse Link Encoder .......................................................................................... 31CDMA Call Processing Overview .................................................................................. 33CDMA Handoffs............................................................................................................. 35

PWID Center for Excellence Revision 0, 4/24/97 Introduction to CDMA Theory & IS-95 Standard


“Code Division Multiple Access”:• Digital technology for secure transmission of data.

• CDMA is used in a Cellular Radio environment:

• Many users may occupy a physical CDMA channelat one time.

• Mobile Stations are dual or tri-mode (CDMA,AMPS, NAMPS).

What is CDMA?

BTS

MS

Forward Link

Reverse Link

CDMA Channel Bandwidth (Either Forward or Reverse Channel):

1.23 Mhz

MS

MS

.27 Mhz .27 Mhz



NOTES:

Fundamental Statements

CDMA- Code Division Multiple Access:

• CDMA technology was developed originally by the government in order to provide asecure communications path which was also resistant to the effects of jamming by enemysignals.

• CDMA “spreads” the digital data which represents the signal to be delivered over a widefrequency spectrum, thus lessening the negative effect of a narrow bandwidth jammingsignal at a particular frequency.

• The CDMA technology was first proposed for use in a cellular environment by acompany called QualComm, based in San Diego, California. QualComm owns manypatents on the CDMA technology for cellular- Motorola has made licensing agreementswith QualComm in order to build CDMA equipment.

• CDMA is defined by an Interim Standard of the CTIA: IS-95.

• The first physical CDMA channel will require 1.77 Mhz (1.23 Mhz at 3dB point, 270 Khzguard band on each end). This is the equivalent of 59 AMPS channels. Note that ifCDMA is being deployed at 800 MHz (the AMPS band), then 59 AMPS channels mustbe cleared from use in the area in which CDMA is being deployed.

• Each additional carrier requires a separation from other CDMA carriers of 200 kHz.Recommended center frequency AMPS channels for the A-Band’s first CDMA carrierare 283 (Primary CDMA channel), and for the B band it’s 384 (Primary).

• Multiple “logical” channels reside in a single physical channel.

• A Forward CDMA channel carries traffic from the Base Site to the Mobile Station.

• A Reverse CDMA channel carries traffic from the Mobile Station to the Base Site.

• A CDMA channel (either Forward, or Reverse) will be utilized by multiple MobileStations. Special digital codes are used to keep the conversations apart.

• CDMA Mobiles are being built by Oki, QualComm, Motorola and other companies.Most of these mobiles are either dual mode (CDMA/AMPS), or tri-mode (CDMA/AMPS/NAMPS).



Why CDMA?

• Provides more capacity than any other system, digi-tal or analog.

• Secure system.

• Make-before-break handoffs.

• Adapts to varying load conditions.

• Mobiles utilize less transmit power:- Raising battery life

- Lowering health risks

• Utilizes additive multipath.

??????????



NOTES:

Why CDMA?

Before embarking on a study of CDMA technology and Motorola’s CDMA prod-uct, it is convenient to study the reasons why CDMA is receiving so much atten-tion in the cellular world today.

• Current analog system (TACS, AMPS, NAMPS, etc.) are rapidly reaching theirsaturation point. Although in theory analog systems could expand without bounds byintroducing microcells and more powerful computers, the cost of this expansion isprohibitive. A new system is required which will provide more capacity than theseanalog systems- CDMA (in theory) provides up to 10 times the capacity of AMPSsystems.

• CDMA provides the highest level of security of any digital or analog system currently onthe market. By encoding the signal to be transmitted with digital sequences which areextremely hard (if not impossible) to reproduce by eavesdroppers, a fraud-proof systemcan be achieved. This is very attractive to cellular operators, since fraud is very commonin current analog cellular technology.

• CDMA is currently the only system to provide make-before-break handoffs. This meansthat the signal is not interrupted while the mobile station starts functioning with anotherbase station. The perceived voice quality is improved in CDMA (over any other system)due to this feature.

• A CDMA cell site can automatically change its output power, and thus change the cell’scoverage area. This change in output power is inversely related to the amount of trafficcovered in a cell- thus, a cell which is handling much traffic will shrink its coverage area,and other cells around it will pick up some of the traffic in the remaining area.

• It is critical that all CDMA mobiles use only the power necessary to achieve a goodsignal at the base site. As a result, CDMA mobiles utilize very little power on average-thus, CDMA mobiles will have longer batter life than their analog or digitalcounterparts. This “feature” of CDMA also results in lowered health risks.

• The CDMA technology utilizes a concept called “additive multipath”, in which time-delayed versions of the same signal are combined in a way such that the final signal isalways stronger than any of the separate paths. Thus, CDMA will provide a betterquality of service due to this feature.



Comparison of Cellular Technologies:

• FDMA: physical channel = 30 kHz, 1 user at onetime.

• TDMA: physical channel = 30 kHz, 3 users (3 logicalchannels), each at a given time-slot.

• CDMA: physical channel = 1.23 MHz, N users, eachwith a given code which keeps the conversations sep-arate.

• CDMA can utilize FDMA concepts by employingmultiple CDMA channels.

1.23



NOTES:

Fundamental Comparisons

FDMA (AMPS, N-AMPS, TACS):

• An FDMA AMPS channel separation (separation between center frequencies) is 30 Khz,while NAMPS channel separation is 10 Khz, and TACS is 25 Khz.

• One channel (i.e. one frequency) may be utilized by only one user at a time.

• Channels are assigned to conversations on-demand; there is a “hard” limit on thenumber of simultaneous conversations allowed.

TDMA (GSM, PDC, IS-54):

• Each conversation is assigned a given amount of time (a time slot) on a single FDMAchannel.

• In GSM, up to 8 users may utilize a single FDMA channel (200 Khz channel separation)simultaneously, while PDC and IS-54 both place up to 3 users per AMPS channel (30Khz).

• Channels are assigned on-demand; thus the “hard” limit still exists.

CDMA:

• Physical channels consists of a relatively large bandwidth (1.23 Mhz).

• Multiple conversations (logical channels) are supported on the same physical channel.Special digital codes called Walsh codes and special pseudo-random noise sequencescalled Pn codes are utilized to keep conversations apart on the same physical channel.

• A “hard” limit does not exist, since the system is limited by the noise level on a CDMAchannel. To enable more conversations, the overall quality of all the conversations maybe slightly diminished for a short period of time.

• FDMA concepts may be utilized in a CDMA system by providing multiple CDMAcarriers, each occupying 1.23 Mhz of bandwidth. Frequency planning must be employedin order to take advantage of the multiple carriers.



CDMA Concept: “The International Party”:

• Each conversation shares the same spectrum.

• Every pair speaks a different language.

• If noise level increases, all guests must raise theirvolume.

• Guests may go to another room if the noise is toohigh.

• One guest speaking too loudly can ruin all other con-versations.



NOTES:

CDMA Basic Concepts

CDMA- Similar to an “International Cocktail Party”:

• Couples speak different languages (French, German, Spanish, English, etc.). Thus, wesee that linguistic diversity is used to enable different conversations at once in the sameairspace.

• If the total volume is too high in the room, some couples may opt for going to anotherroom and continuing their conversation there. Thus, we see that spacial diversity mayalso be utilized to enable more conversations.

• A single person who raises his or her voice level will cause the other couples in the roomto have more difficulty in understanding each other. These other couples may be forcedto increase their volume as well. This process quickly results in having an unacceptablyhigh level of noise in the room.

• A load rock-and-roll band (or any kind of music, for that matter), would cause the totalnoise in the room to increase as well. This will again cause couples to raise their volumein order to be heard, and could result in the same situation as above.

• Couples who are in close proximity do not need to use a high volume level in order to beheard, thus these couples do not contribute as much to the noise level in the room as doother couples who are farther apart.

Conclusions from the “International Party” which apply to CDMA:

• Existing noise floor must be low (i.e. rock-and-roll band must play with low volume).This means that the existing radio spectrum must be clear from interference (microwaveoperators at 1.9 GHz, and regular AMPS spectrum at 800 MHz are two examples ofinterference contributing to the noise floor).

• Power control is crucial in CDMA. Mobiles that are close to the BTS must use a smallertransmit power than mobiles that are farther away. It is crucial that all mobiles betransmitting the lowest possible power at all times, since each mobile contributes to thegeneral noise floor perceived by all other mobiles that are operating at the same CDMAfrequency.

• The concept of using different languages at the party is parallel to using digital codes inCDMA. By multiplying the digital stream of data (which represents the conversation) bya much longer sequence of digital data which has been specially selected, the originaldata can be combined non-destructively with other conversations on the same physicalchannel.



Digital Voice Coding Concepts

• Digital voice coding concepts called “vocoding” usedin CDMA.

• Encoder takes digital samples of analog voice wave-form as input, outputs stream of digital codes.

• Decoder does the inverse.

• Objectives:- Minimize amount of data transmitted

- Maintain acceptable voice quality

- Keep equipment “affordable”

• QCELP: QualComm Code-Excited Linear Predic-tive Coding.

• Vocoder is variable-rate: .8, 2, 4, 8.6 and 13 Kbps.

QCELP

Encoder

QCELP

DecoderA/D D/A



NOTES:

Digital Voice Coding Concepts

The human voice produces analog waveforms. These waveforms are picked upby a microphone, and transmitted as electrical impulses to a circuit which per-forms analog to digital conversion. The manipulations performed on the result-ing digital stream are performed by a device called the “vocoder”. Some pointsregarding the device are as follows:

• Conversion from a digitally sampled waveform (call Pulsed Code Modulation or PCM)to a compressed digital format is called encoding.

• Conversion from a compressed digital format back to PCM is called decoding.

• The CDMA vocoder functions on the basis of modeling the human voice tract as a linearpredictive filter. The filter uses a pitch generator and a white noise generator as inputs,and determines the set of parameters (called filter coefficients) for the filter in order tobest represent the 20 ms of speech being encoded. The result is data which representsparameters to be fed to the decoder in order to reproduce the speech.

• The number of bits transmitted to represent the parameters can be modified- thusmodifying the resolution of the filter parameters. It is logical to deduce that as highervoice quality is desired, more bits will be needed to reproduce the filter parameters morefaithfully at the decoder.

• The CDMA SuperCell vocoder is a variable rate vocoder, supporting rates of .8, 2, 4, 8.6and 13 Kbps. The 13 Kbps rate is not part of the IS-95 standard, rather it is an effort (byMotorola) to provide an option for high quality voice. In addition, several companies(including Motorola) have been working on an “enhanced” 8 Kbps vocoder whichreportedly provides very good audio quality.

• The vocoder switches the encoding rate based on the perceived energy of the signal beingencoded. That is, if the person is listening (not speaking at all), a slower vocoding ratewill be utilized in order to send fewer bits over the transmission medium. One of theside-effects of this is that background noise is diminished during a conversation with alow voice activity factor (many silences in the conversation, or one party speaks most ofthe time).

• The technology for the CDMA vocoder is called QCELP, for QualComm Code ExcitedLinear Prediction.



Digital Signaling Concepts- Work Sheet• Fill-in first two rows with zeros and ones:

• X-OR the first two rows, place result in third.

• Copy third row above into first row below, secondrow above into the second row below. X-OR the twosequences.

• Increase the ratio of bits in the scrambling code tothe bits in the data sequence up to 4:1:

• Introduce errors, de-code as above.

• When scrambling code has more bits than the databeing scrambled, resulting sequence has wider BW.



NOTES:

Basic Digital Signaling Concepts- Isomorphism & Spreading Sequences

The work sheet on the previous page will help in understanding two basic princi-ples: isomorphism and signal spreading.

• An X-OR of two binary sequences will yield another binary sequence, which when X-ORed by either of the first will result in the remaining sequence. This property is calledisomorphism. For binary sequences, both modula-2 multiplication and X-OR operationsexhibit this behavior.

• If one of the X-ORed sequences is made longer (i.e. one sequence is multiplied by severalbits of the other sequence), then the resulting sequence will have the same amount of bitsas the longer sequence.

• When a binary sequence which is running at a slow speed is X-ORed by another binarysequence which is running at a higher speed, the result is a binary sequence which runsat the higher speed. Translating this into the frequency domain, the slower binarysequence occupies a smaller bandwidth, while the faster binary sequence occupies alarger bandwidth.

• By X-ORing a slow sequence by a fast sequence, the slow sequence is “spread” to alarger bandwidth.

• In order to recuperate the original (slower) sequence, the resulting fast sequence is X-ORed by the same fast sequence used to spread the slower sequence, then the result isintegrated (averaged) across every <N> bits to produce the original (slower) sequence.Note that we must use the exact same spreading sequence!

• In CDMA, the slower sequence is the data which we want to transmit (which has beendigitally manipulated to insure more resistance to fading and interference), while thefaster (spreading) sequence is called the Pseudo-random noise code (Pn code). Theresulting fast sequence is used to modulate the RF carrier.

• Note that the Pn code must be carefully selected, since we want a good balance of onesand zeros (imagine how bad the spreading would be if the Pn code were all zeros!).

• Also note that, since there is a processing gain (in the work sheet, the gain is 4:1), theresulting signal is more resistant to fading.



Most definitely!• Physical RF carrier of 1.23 Mhz transmits multiple

conversations.

• Multiple CDMA carriers may exist, each with many“virtual” channels on it.

How?• Spreading sequence runs at 1.2288 Mcps, while data

runs at 9600 bps... 128 chips per bit.

• Up to 63 bits can be damaged, and original bit canstill be recovered!

• Digital properties of orthogonality and auto-correla-tion used to determine Pn sequences and Walshcodes.

• Different Pn code offsets used between BTSs, to keepconversations from interfering.

Multiple Conversations on One Channel?



NOTES:

Multiple Conversations on One Channel?

The concept of having multiple conversations multiplexed on one physical chan-nel at the same timeis one of the main differences between CDMA and any othersystem. Understanding how this is achieved is beyond this section (more to followin another section). However, the following is presented to give the reader someclues:

• Special digital codes called Walsh codes multiply the data to be sent in order to increasethe probability of de-coding the conversation after all other conversations have beenmultiplexed on the same channel.

• Walsh codes are orthogonal, which implies that the multiplication of a Walsh code withanother Walsh code will not produce another Walsh code.

• Pn codes are specially selected so that a time-displaced version of the code multiplied bythe same Pn code with another time-displacement will result in zero.

• “Pn code offset” is a term that is utilized often in CDMA, since each CDMA BTS willtransmit with a different time-offset of the same Pn code. Since multiplication of thecodes will result in zero, CDMA cell sites can use the same CDMA carriereven thoughthe sites are next to each other!

• The CDMA system changes the Pn code being used very rapidly in order to provideresistance to fraud. Due to the time-orthogonal nature of the forward link, all cell sitestransmitting the same CDMA carrier must change their Pn code at exactly the sametime. In order to do this, each cell site needs an absolute time reference, which is takenfrom Global Positioning System satellites orbiting the earth.



CDMA Forward Link Channels:

• Walsh codes used to separate conversations on For-ward link.

• 64 Walsh codes, each assigned as illustrated above.

• Pilot Channel used by MS to acquire CDMA system.

• Sync Channel used by MS to acquire synchroniza-tion and system parameters.

• Paging Channels used to broadcast pages and ordermessages to MS.

• Traffic Channels carry CDMA STRAU frames,along with Power Control Bits.

Forward CDMA Channel (1.23 MHz)

Pilot

Channel

Paging

Channel1

Paging

Channel7

* * *Traffic

Channel1

Sync

Channel

Traffic

Channel23

Traffic

Channel24

Traffic

Channel55

* * * * * *

W0 W1 W63W33W32W31W8W7

Wn = Walsh code channel.

TrafficChannel

N

CDMA Strau Data PowerControl



NOTES:

Forward CDMA channels

• Logical channels inside the forward CDMA link are identified through the use of time-orthogonal codes called Walsh codes. Walsh codes are binary sequences of 64 bits inlength. The fact that they are orthogonal implies that one code multiplied by anotherwill not result in another of the 64 codes. This property of orthogonality only holds truewhen the Walsh codes are not “displaced” with respect to one another- this is why Walshcodes are used to separate conversations only on the Forward link.

• Assignment of the 64 possible Walsh codes is as follows:

• Walsh 0 is the Pilot Channel.• Walsh 1-7 are the Paging Channels.• Walsh 32 is the Synch Channel.• All other Walsh codes (55 in all) identify traffic channels.

• The Pilot Channel allows the mobile stations to identify the cell site or sector which themobile stations are listening to, and it provides a coherent phase reference so that themobile station can de-modulate the other channels correctly. The Pilot Channel does nottransmit any data, it only transmits the carrier modulated by Walsh code zero (which ismade up of all zeros).

• The Synch Channel constantly transmits a single message containing informationneeded by the MS in order to synchronize with the CDMA infrastructure, and to benotified of certain data such as system and network identification numbers.

• The traffic channels carry the user’s encoded data bits, which are in data packets calledSTRAU frames, along with power control information (one bit every 1.25 ms)- a “1”means lower power by 1 dB, a “0” means raise the power by 1 dB).



CDMA Reverse Link Channels:

• Reverse Link modulation is different- channels areseparated by different offsets of a long digital codesequence.

• Certain offsets set aside for use as Access Channels.

• IS-95 provides for maximum of 30 Access Channelsper Paging Channel.

• Access Channel used by MS to communicate withthe BTS when MS is NOT on a traffic channel.

Reverse CDMA Channel (1.23 MHz)

Access

Channel

1

Access

Channel

N

* * *

Traffic

Channel

1

Traffic

Channel

M* * * * * * * * * * * *

Addressed by Long Pn Code Offsets



NOTES:

Reverse Link Channels

CDMA Reverse Link Channel Structure

• Modulation on the reverse link is different than for the forward link- time offsets of thePn long code are used to keep conversations separate, and Walsh codes are used torepresent the data being sent.

• The Pn Long Code is a 42 bit code, so there are 242-1 possible time shifts of the Pn LongCode. Thus there are 242-1 codes available which may identify separate conversations.

• Access Channels are used whenever the mobile is not on the Traffic Channel and mustcommunicate with the Base Site. Access Channels are used for the following:

• Call Originations• Responses to Page• Responses to Order messages from BTS• Mobile Registrations

• There are a maximum of 30 Pn Long Code offsets per paging channel (for a maximum of210 offsets) which are reserved for Access Channels.

• Those Pn Long Code offsets which are not used for access channels may be used astraffic channels.



IS-95 Forward Link Encoder



NOTES:

IS-95 Forward Link Encoder

• At the top of the encoder diagram is the Pilot channel. Note that Walsh code 0 (which iscomposed of all zeros) is used to “spread” a digital sequence which is all zeros, and thenthe result (which will be all zeros) is sent to the quadrature spreader and then used tomodulate the carrier.

• A convolutional encoder will “scramble” the bits to be sent, and at the same time willproduce more bits at the output than those presented at the input. The encoder “loads”<length> bits into registers, and then performs operations (such as ORs and ANDs) onthe bits. The results of the operations are then sent as output. The rate of theconvolutional encoder in IS-95 can be either 1/2 (meaning that two bits are output forevery bit of input), or 1/3 (meaning that three bits are output for every bit of input).

• Symbol repetition basically repeats each symbol a given amount of times. This is done inorder to insure proper delivery at the other end. The Block Interleaver will shuffle largeblocks of bits around such that spurious errors can be recovered through CRC errorcorrection.

• The Sync channel data “runs” at 1200 bps, always repeating the same message (whichcontains information on the system time, long code state, etc.). This information is sentto a 1/2 rate convolutional encoder, and then to a 2x block interleaver. This informationstream is “spread” by Walsh code 32 and then sent to the quadrature spreader.

• The Paging channel data is run through a convolutional encoder, symbol repeater, blockinterleaver, and then the information is scrambled through the use of a Pn long code.The receiving side knows the long code and the mask used, so it can de-scramble theinformation. The Paging channel may transmit at any of several rates: Motorola iscurrently using 4800 bps.

• The Traffic channel will transmit data from the vocoder at a variable rate (which will bedetermined by the vocoder). This data is sent through a convolutional encoder, symbolrepeater, and block interleaver. The rate of symbol repetition will depend on the rate ofthe data coming from the vocoder- the end result is always a data stream of 19.2 ksps(symbols per second). This stream is scrambled with a long code, using a long code maskwhich is based on the user’s mobile id number.

• The Traffic channel also carries power control bits- these bits are “pushed” onto thestream of data bits (called blank-and-burst).



IS-95 Reverse Link Encoder

• Access channels and Traffic channels on reverselink.

• Convolutional encoder is rate 1/3 (as opposed to rate1/2 on fwd link).

• Modulation is through a Orthogonal Modulator.

• Interference is minimized at lower bit rates by DataBurst Randomizer.

• Signal is scrambled with long code.



NOTES:

IS-95 Reverse Link Encoder

• The Reverse Link is composed of two types of channels: Access channels, and Trafficchannels. Access channels use a data rate of 4800 bps, while Traffic channels havevariable rate.

• Data is output by the IS-95 variable-rate vocoder at either .8, 2.0, 4.0, or 8.6 kbps. Thisdata stream is fed to a Cyclic Redundancy Code generator, which will add data for re-constructing erroneous bits. This is done only for the 4 and 8.6 kbps rates, since theother rates will be repeated so often that data integrity can be provided throughrepetition.

• After CRC, 8 more bits are added to each frame for encoding purposes- this brings thespeed up to 1.2, 2.4, 4.8, or 9.6 kbps. Then the data stream is fed to a convolutionalencoder of length 9 bits, and rate 1/3 (meaning that 3 bits will be output for each bit ofinput). The resulting data stream is fed to a block interleaver, and then to an orthogonalmodulator.

• Orthogonal modulation takes 6 bits of input (which in binary can specify 64 differentpossibilities), and looks up the corresponding Walsh code. For example, if 6 bits of inputare ‘101100’ (56 decimal), then Walsh code 56 will be sent in place of the 6 bits. SinceWalsh codes are time orthogonal, it’s very easy to detect when one Walsh code ends, andanother begins by simply comparing each 64 bits with all of the Walsh codes until aperfect match is found.

• The data burst randomizer will blank out data from the stream to be sent, based on thedata rate of the information bits. More bits are blanked out for the lower informationdata rates, since at lower rates the data is repeated more often. This has the effect oflowering interference on the reverse link when lower data rates are used.

• The resulting data stream is then scrambled with a long code (which has been modifiedwith a mask based on the mobile id), and then sent to a quadrature spreader.



CDMA Call Processing Overview

• Call Processing specified by “states” and “events”.- MS may be in any particular state.

- Event will cause transition to another state.

• Event will be signaled by arrival of message fromdevice in CDMA network.

• State transitions usually involve performing anaction.

• MS may be in any of the following CP states:- Initialization State

- Idle State

- System Access State

- Control on the Traffic Channel State

PowerUp

MSInitiali-zationState

MSIdleState

MSAccessState

MSControlon TchState

MS

MS has sele-cted andacquired theSystem.

MS moni-tors thePaging chnl.

MS isdirected toTraffic chnl.



NOTES:

CDMA Call Processing Overview

The process of setting up and tearing down CDMA calls is called “Call Process-ing”. Several definitions are presented in order to facilitate the discussion of callprocessing (CP):

• A device can be in any of several call processing states. The more general states are asfollows:

• Initialization State. This is the first main state, to which an MS will always revertupon critical problems in any of the other states, and to which an MS will transitionupon power-up. In this state, first the system to use is determined (whether CDMAor another technology such as AMPS). If CDMA is to be used, the MS then acquiresthe Pilot channel, and the Sync Channel. From the Sync channel, the MS is able tosynchronize its internal timing mechanisms to that of the CDMA infrastructure.

• Idle State. From within this state, the MS can initiate registration messages, initiateor receive a call, receive informational messages and orders, and initiate a messagetransmission. The Pilot channels from different cell sites are constantly beingscanned to determine the strongest cell site, and an Idle State handoff may also beperformed so that the MS is always “listening” to the strongest Pilot channel.

• System Access State. In this state, the MS sends messages to the BTS over any of theReverse Channel Access channels (of which there are a maximum of 210), andreceives messages from the BTS over the Paging channel(s).

• Control on the Traffic Channel State. The MS is on the traffic channel in this state(presumably transmitting digital data which represents either voice or telemetrybeing sent from the MS). Power control data is sent to and from the mobile in thisstate, hand-off orders and traffic data are also sent in this state.

• An “event” may cause a transition from one CP state to another. Events will usually bemessages sent over communication links to devices which “see” the physical event takingplace (such as a perceived signal quality drop at a transceiver).

• A state transition will usually involve performing some pre-defined action (such asturning on a busy signal at the MS speaker, or sending an acknowledgment message overthe access channel).



CDMA Handoffs

• Soft Handoff.- One CDMA BTS to another CDMA BTS.

- Both BTSs under same CBSC.

- Same CDMA carrier.

• Softer Handoff- One CDMA sector to another.

- Both sectors on same BTS.

- Both sectors using same CDMA carrier.

• Hard Handoff- CDMA-CDMA, BTSs under different CBSCs.

- CDMA to analog system.

• Lists of pilot channels are kept at the MS:- Active Pilots

- Candidate Pilots

- Neighbor Pilots

MMXC

BTS

MSC

BTS BTS

CBSC

MS



NOTES:

CDMA Handoffs

Handoffs are the process of getting a mobile handset to “talk” to another BTS,or to another sector of the same BTS. The following is information on the differ-ent types of handoffs available in CDMA:

• Soft Handoff. This is an attractive feature of CDMA, where the conversation or dataflow is not interrupted due to the hand-off. Soft handoff can currently be performedwhen an MS goes from one CDMA BTS to another, where both are under the control ofthe same centralized controller. Also, the source and target BTSs must both be utilizingthe same CDMA carrier frequency. In the near future, soft handoff will be extended sothat it applies when the BTSs are not under the same centralized controller.

• Softer Handoff. This is the same as Soft Handoff, except that it applies only when aCDMA handset goes from one sector to another under the same CDMA BTS, at thesame CDMA carrier frequency.

• Hard Handoff. This occurs under two different circumstances- when a CDMA handsethands-off from a CDMA BTS to an Analog BTS, or when a CDMA handset hands-offbetween CDMA BTSs which are utilizing different CDMA carrier frequencies.Conversation is interrupted for a brief instance in hard hand-off.

• The MS is constantly scanning all possible offsets of the Pn code (it takes 26.6 ms perscan cycle), and keeps three lists of Pilot Channels:

• The Active Set. This is a list of Pilot Channels which are actively used by the MS totransmit digital data. CDMA (IS-95) provides for up to three members in this activeset at once. That is, a CDMA handset may be “talking” over at most three differentsectors or cell sites at once.

• The Candidate Set. This is a list of Pilot channels which are not strong enough to bein the active set, but are strong enough to be candidates for hand-off.

• Neighbor Set. These are candidates which are very weak, but still visible at the MS.• Remaining Set. These are all other Pilot Channels (from the 512 channels that are

available in a single CDMA carrier) that are not members of any of the above threesets.

36 of 259 Motorola Confidential Propretary

Conclusions:• “CDMA” stands for Code Division Multiple Access.

• Physical channels: 1.23 MHz in BW, support manylogical channels simultaneously.

• Forward link transmits from BTS to MS, whilereverse link transmits from MS to BTS.

• CDMA is like an International Party, each pair ofguests are conversing in a different language.

• QCELP is the digital variable rate vocoder used.

• Pseudo-random noise codes carefully selected, makingit very difficult to “break” CDMA.

• Walsh codes separate conversations on forward link.There are 64 Walsh codes in all.

• Short Pn code time offsets identify sectors and cells.

• Long Pn code time offsets identify conversations onreverse link.

• Soft handoffs are from BTS to BTS (same controller,same carrier). Softer handoffs are from sector to sec-tor (same BTS), while hard HO are from CDMA toanalog or from one CDMA carrier to another.


PWID Training: CDMAAn Introduction to the SuperCell System


Objectives:

• To present an introduction to Motorola’s SuperCellproduct for CDMA technology in order to provide thestudent with a basic understanding of the SuperCellproduct.


Table of Contents:

CDMA Network Architecture......................................................................................... 41CDMA Functional Partitioning..................................................................................... 43CDMA System Cardinality ............................................................................................. 45Basic System Inter-Connections .................................................................................... 47CDMA CBSC Overview.................................................................................................. 49Logical Paths through the CBSC................................................................................... 51The CDMA Tandem Platform........................................................................................ 53The CDMA Transcoder .................................................................................................. 55The CDMA Transcoder Cards ....................................................................................... 57Transcoder Connectivity................................................................................................. 59The CDMA BTS ............................................................................................................. 61The CDMA SC9600 BTS Devices.................................................................................. 63CDMA Software Overview ............................................................................................. 65CDMA System Database Creation................................................................................. 67CDMA Software Installation & Initialization ............................................................... 69CDMA Customer Interfaces........................................................................................... 71

PWID Center for Excellence Revision 0, 4/24/97 An Introduction to the SuperCell System


CDMA Network Architecture

• PSTN- provides path to & from land subscribers.

• MSC- interfaces PSTN to Supercell.

• CBSC- provides management of real-time cellularradio functions.

• BTS- provides air interface to MS. LMF used forcalibration, on-site maintenance.

• OMC-R/UNO- provide customer presentation.

MMXC

BTS

MSC

PSTN Public Switched Telephone Network

MSC Mobile Switching Center

OMC-R Operations & Maintenance Center - Radio

XC TranscoderBTS Base Transceiver Station

LMF Local Maintenance FacilityMM Mobility Manager

voice & control

control / signalling

BTS BTS

LMF

LMF

LMF

OMC-R

PSTN

CBSC UNO (AP)

MS MS

UNO/AP UNO Platform (Application Processor)CBSC Centralized Base Station Controller

The Supercell System

Base Site System (BSS)



NOTES:

CDMA Network Architecture

This diagram is simplified, so as to present only the major “black boxes” whichimplement CDMA. Here are short descriptions for each of the boxes in order toprovide a basis for more in-depth discussion:

• PSTN - Public Switched Telephone Network, responsible for providing paths to andfrom land-based telephone subscribers.

• MSC- Mobile Switching Center, responsible for interfacing the SuperCell CDMAsystem to the PSTN. The MSC may be provided by Motorola (an EMX2500), or byanother vendor. The MSC routes calls to/from the PSTN to the CBSC, and maintainssubscriber records and billing. The MSC is NOT considered a part of the SuperCellarchitecture, although an MSC is necessary to provide cellular service.

• CBSC- Centralized Base Station Controller. The CBSC is the combination of the MMwith the XC and a “Router Frame” which contains routers, dsu’s and optionally a XC-PC.

• MM- Mobility Manager. This is a Tandem Computers Integrity-UX box, which isresponsible for specialized radio channel management features, managing channelallocation, certain types of handoffs, and is the managing agent for all the manageddevices in the SC network (BTS, XC).

• XC- Transcoder. This box performs transcoding of digital data from PCM to a CDMAformat, and it concentrates span lines from many BTSs to the MSC.

• OMC-R- Operations & Maintenance Center, Radio. This is a Tandem ComputersIntegrity-UX box, responsible for statistics & alarm collection, and configurationmanagement of one or more CBSCs.

• UNO- Unified Network Operations Processor. This is a SUN SPARCStation 20, whichprovides a graphical user interface for managing and providing status of the SCnetwork equipment.

• BTS- Base Transceiver Station. This entity provides the RF interface to the MobileStations.

• LMF- Local Maintenance Facility. This is a portable PC which is used to perform on-sitemaintenance of several SC devices, including calibration of the BTS.



Functional Partitioning

A+ Interface to MSCHandoff ArbitrationBTS Fault Mngmnt.Channel AllocationPage DistributionBTS Init. & Download

Traffic ChannelsPower ControlSignal Quality MeasurementsSignalling on page & access channelsSofter Handoff

MobilityManager

XC

CBSC

MCC

BTS

GLI

OMC-R

Centralized BSS O&MCentralized Configura-tion ManagementCode and Data StorageCommand Line Inter-faceEvent Log/CDL LogConfiguration Database

Message RoutingSite O&M

QCELP TranscodingSwitching FunctionMobile Event Detect.Signalling on the TCHPaging DistributionSoft handoff han-dling

UNO

Alarm ManagementPerformance DataPlanning DataStorage & ReportsGraphical User Inter-face

MSC

Billing RecordsNumber TranslationSubscriber Data-baseHard HandoffA+ Interface toCBSCPSTN, IS-41



NOTES:

CDMA Functional Partitioning

• The MSC has a limited role in Supercell (as compared to analog). It is limited tomaintaining billing records, the subscriber database, providing the A+ interface, anddialed digit translations, as well as some functionality for hard handoffs.

• The OMC-R is the operations & maintenance center for all CBSCs under its control.The OMC-R is the central repository for code and data in the system. Software forSuperCell releases is stored on the OMC-R, and propagated to each CBSC when asoftware upgrade is performed. In addition, the OMC-R contains the database whichspecifies how the CBSCs are configured. The OMC-R also collects Call Detail Logs andstatistics for later analysis.

• The UNO (Unified Network Operations Platform) is the customer presentationplatform. This platform presents a user-friendly graphical interface through which thecustomer may obtain status of the network elements, and the customer may also managethe network through the use of alarms, reports, and CLI (command-line interface)commands.

• The CBSC is comprised of the MM, XC, DSU’s and Routers.

• The MM is one of the “active brains” of the system. The MM performs handoffarbitration, device fault management, channel allocation, page distribution, BTSinitialization and code download, as well as the collection of alarms and statistics forlater relaying to the OMC-R.

• The XC (Transcoder) performs switching of timeslots and conversion of PCM (Pulse-Code Modulation) to QCELP (QualComm Code Excited Linear Prediction). Inaddition, the XC takes care of signaling information while the mobile is on a trafficchannel. The XC handles soft handoffs for mobiles under base sites controlled by theXC. The XC also distributes pages to mobiles, and routes control data from the MM tothe base sites. The XC may also provide echo cancelling.

• The BTS (Base Transceiver Site) is composed (among others) of the GLI (Group LineInterface) and the MCC (Multiple Channel Card). The GLI routes messages to/from theCBSC, and it performs site operations & maintenance. The MCC cards handle trafficchannels, power control, signaling, quality measurements and softer handoff.



CDMA System Cardinality:

• CDMA system can be grown to cover large areas.

• MSC may have multiple CBSCs connected to it.

• OMC-R will typically control from 1 to 5 CBSCs.

• CBSC will typically control from 1 to 90 BTSs.

• Numbers depend on load distribution, softwarerelease, and hardware release.

MMXC

CBSC

MMXC

BTS

MSC

BTS BTS

OMC-R

PSTN

CBSC

UNO (AP)

MSC

MMXC

CBSC

Other CellularNetwork (Analog,CDMA, etc.).

BTS BTS BTS

BTS BTS BTS



NOTES:

CDMA System Cardinality

This diagram gives a sense of how devices are inter-connected to cover a largearea with CDMA.

• The maximum number of sub-tending devices that can be connected to a parent devicewill depend upon several factors- including the following:

• The call processing load seen by the sub-tending devices.

• The release of software and hardware in the devices.

• The amount of equipment in each of the sub-tending devices (MCC cards, etc.).

• One MSC may have multiple CBSCs connected to it. The Supercell PDC project inJapan had at one point a single MSC with 7 CBSCs connected to it.

• One OMC-R may have several CBSCs connected to it. Again, the maximum will dependon the load that the OMC-R is capable of handling. Typical for the Supercell PDCproject was to have 3 to 5 CBSCs on one OMC-R.

• One MM may have several transcoder frames connected to it. The transcoder isexpanded with more frames when traffic conditions require the expansion.

• Each transcoder can handle several BTSs- maximum number will depend on theconditions mentioned above. The Release 4 XC system limits are 50 KBHCA, 200 BTSlinks, 350 E1/T1 Span lines, 2000 PCM Voice Circuits.

• The MM in Release 4 will handle 35 KBHCA.



Basic System Inter-Connections

• Co-located MM and MSC.

• Ethernet carries data for OMC-R to AP, MM toOMC-R and MM to XC connections.

• E1 or T1 span lines carry traffic and control data forthe BTS<->XC link.

• Redundancy for most links is available (for extra$$).

Con

trolSwitching

Matrix

MSC OMC-R

BTSCluster

TCP/IPoverEthernet

Network Interface

Network Interface

XCDR

Transcoder

MobilityManager

Bridge

BTS BTS BTS

TCP/IP overToken Ring

CBSC

T1 or E1

RS-232

UNOPCM overT1 or E1

A+ over SS7/C7.

SCAP over LAP-D@ 64 Kbps

Traffic Channels. Each T1 Link4 TCH. Traffic is QCELPcarries

STRAU frames.

Legend:Control & DataTraffic



NOTES:

Basic System Inter-Connections

This diagram illustrates the basic connections between the network elementswhich make up a CDMA SuperCell system.

• The diagram represents a situation where the MSC is co-located with the CBSC, sincethe SS7 links are directly connected between the CBSC and the MSC. When the twoentities are not co-located, it is necessary to transport the SS7 links through the XC andover a T1 or E1 span line to the MSC. In order to do this, two additional DSU’s areneeded for the V.35 lines from the MM to be placed on T1 links to the XC. All CDMAsystems so far have co-located MSC and CBSCs.

• Redundancy is present in the system, but not represented in the diagram. There are atleast two SS7 links to the MSC, exactly two TCP/IP links between the MM and the XC,and two TCP/IP links between the MM and the OMC-R. Redundancy for the AP wasnot deemed necessary, so there is only one TCP/IP link over Ethernet between an APand the OMC-R.

• A+ over SS7 is used to communicate control data from the MM to the MSC. A+ is anOpen Interface, meaning that the information necessary to comply with the interface isfreely available, so that any switch manufacturer may build a switch that can “talk” A+.

• SCAP (SuperCell Application Protocol) is a set of messages and message formatsdesigned by Motorola for communication with the BTSs. This is a Motorola Proprietaryinterface, so companies outside of Motorola cannot build Base Sites compatible with thisstandard. SCAP is sent over a LAPD transport, using one channel of a T1 or E1 carrier.

• Traffic Channels to the Base Sites are multiplexed 4 to 1 on the T1 time slots. EachTraffic Channel uses 16 kbps of QCELP (QualComm Code Excited Linear Prediction)data. This data represents the speech over the traffic channel- two of these channels(forward and backward) make up a two-way conversation.



CDMA CBSC Overview

• CBSC controls all BTSs under it. Code & datadownload, initialization, fault management, call-setup, call-teardown, soft/hard handoff, statisticscollection, concentration of span lines.

• Routers convert TCP/IP/Ethernet to TCP/IP/Token-Ring. DSU’s convert V.35 to DS0 for SS7 links toremote MSC. XC-PC used to download code & datato XC.

MMFrame Transcoder Frame

Router

Router

DSU’s

XC-PC:

Router Frame

MM

612 Kg 100 Kg 250 Kg

DSU/ Transcod-AB

C

D

min. 457 mm Frame Clearance

min. 600 mm Frame Clear-

E

F

A - 600 mmB - 762 mmC - 457 mmD - 600 mmE - 2,162 mmF - 1,819 mm



NOTES:

CDMA CBSC Overview

The CBSC (Centralized Base Site Controller) is a component of the SC9600 sys-tem responsible for controlling all the Base Sites that have been assigned to it, inaddition to concentrating conversations from several BTSs over T1 or E1 spanlines to an MSC.

• A CBSC is responsible for downloading and initializing the devices in the BTS whichmust have code and/or data on them.

• The CBSC is responsible for performing Fault Management of the BTSs- meaning thatif a device in a BTS goes OOS (Out of Service), then the CBSC will attempt to takecorrective action.

• The OMC-R contains all the code and data required by the CBSC, but the OMC-R isnot crucial to the CBSC’s functionality. In fact, should the OMC-R not be present, theCBSC will continue to perform correctly in order to implement the CDMA callprocessing features.

• The XC (Transcoder) is a child of the MM, meaning that the MM performs somemanagement of the XC, and performs some Fault Management on the XC in addition todevice downloading. However, downloading and management by the MM are currentlylimited to the FEP cards- all other cards in the XC are managed by the OMP GPROC.

• Unfortunately, at the present time not all of the devices in the XC can be downloaded bythe MM- so, a special PC called the “XC-PC” must be used to download the XC. TheXC-PC “talks” to the XC’s OMP GRPOC through the fiber-optic token-ring LAN. Notethat the service is interrupted to download the XC, since the MM is taken off of the LANat download time.

• The Routers (or Bridges) serve to convert from TCP/IP over Ethernet on the MM side toTCP/IP over Token-Ring on the XC side. This is because neither the MM nor the XCcurrently support each other’s formats.

• The DSU’s are necessary in order to convert the V.35 data paths from the MM to T1spans to the XC when the MM and the MSC are NOT co-located.



Logical Paths through CBSC

• MSC, OMC-R both at remote sites.

• TCP/IP WAN used for OMC-R<->MM connection,SS7 links nailed through XC over E1 span to MSC.

• MM instructs XC to set-up connections.

Con

trol

BTS BTS BTS. . .

MSC

XC

CBSCDSU MM

TandemI/OProcessor

Router

OMC-R

Cage

XCDR

XCDR

XCDR

SCAP linkTCH

terrestrialcircuit

56 Kb PCM

16 KbQCELP 56 Kb digital

MM-OMCusing TCP/IP

‘A+’ interface over E1

WAN

TCP/IP VirtualCircuits



NOTES:

Logical Paths through the CBSC

This diagram presents the logical paths for control and traffic data through theCBSC.

• Note that the case presented is for a remotely located CBSC. Both the MSC and theOMC-R are removed from the CBSC. This case has not occurred yet, but there is thepossibility that future customers will choose this option.

• A WAN (Wide-Area-Network) is established between the MM and the OMC-R whenthey are not co-located. This WAN is composed of DSU’s and dedicated T1 lines whichtransport TCP/IP data.

• The A+ data for communicating between the MSC and the MM is transported throughnailed circuits in the XC, such that a T1 or E1 to the MSC from the Transcoder willcarry both channel traffic (conversations in PCM format) and control data (A+messages over a 56 kbps or 64 kbps channel).

• One of the functions of the Transcoder is to concentrate conversations from many T1 orE1 lines from several BTSs to a few T1 or E1 lines to the MSC. When the MSC is not co-located, this cuts down considerably on the operator’s cost.

• Note that XC stands for Transcoder (which is the whole frame), while XCDR stands for“Transcoder Board”, which is the device responsible for converting between PCM on theland side (to/from the MSC) to the CDMA Air Interface side (which uses QCELPSTRAU data frames).

• Logical paths are established in the XC at the request of the MM. The XC manages itsown resources in order to establish the appropriate paths as requested by the MM.

• In release 4, when handling a mobile-to-mobile call, the XC transcodes bothconversations from QCELP to PCM, and then from PCM to QCELP again. However, ina future release, mobile-to-mobile calls under the same CBSC will by-pass thetransocding step- thus providing better voice quality for M-M calls.



The CDMA Tandem Platform

• Fault Tolerant Tandem Integrity UX (Unix).

• Doubly-redundant hardware. RISC-based technol-ogy (MM uses MIPS R4400 CPU’s at 175 MHz, with128 Mbytes of fast local memory, 16 Mbytes of glo-bal memory).

• Mirrored disk storage system (running Veritas soft-ware).

I/O ControllerCard Cage

ProcessorCard Cage

Air intake (front only)

Air intake

System Disks orTape Modules (upto 9)

Air Exhaust

Fan Modules(3)

Fan Modules(3)

Bulk PowerSupply Module(4)

MemoryModules (2)CPU Mod-ules (3)

SP Modules

IOP Mod-

DCC(DCConverter)Modules (6)

SDCC(Sin-gle DC Con-verter)

1829 mm

762 mm

IO Cards (willvary from OMC-Rto MM)



NOTES:

The CDMA Tandem Platform

Both the OMC-R and the MM are built on a computing platform provided byTandem Computers, Inc.

• The Tandem box utilizes redundant hardware such that a single failure in the box willnot cause functionality to be interrupted. For example, should a FAN go bad, the otherfans will speed up to compensate for the failure. The Tandem platform will post alarmswhen devices fail such that operator action may be performed. Only one CPU out ofthree may fail at any time (two CPU’s must be sane for the system to operate correctly).All other systems are doubly redundant.

• The box runs a specialized version of UNIX called “Non-Stop UX”, which has beencustomized by both Tandem and Motorola in order to provide more robustness in theOperating System.

• The disk storage system is also doubly redundant and mirrored, such that any failure ona drive will not cause data to be lost. A software suite called “Veritas” implements thisfeature.

• The MM and OMC-R are different in that the OMC-R does not need V.35 interfacehardware for the SS7 links, nor does it need as much processing power as the MM. Onthe other hand, the MM does not need as much disk space as the OMC-R, since theOMC-R is the central repository for the code and data in the SuperCell system.

• The platform runs on -48 Vdc power.



The CDMA Transcoder Frame

• Concentrates & converts data formats betweenMSC, MM, and BTSs.

• Optional Echo Cancelling.

• Gets involved in soft and softer handoffs.

• Can be expanded to multiple cages for more capac-ity if necessary.

Full Size

Daughter

Fan

Full Size

Daughter

Fan

Power Supplies (3)

Power Supplies (3)

Full Size Cards (29 max.)

Fan Modules

Fan Modules

Full Size Cards (29 max.)

Daughter Cards (29 max.)

Circuit Breaker Housing

Power Supply Bds. (2)

Alarm Bds. (2)

Top Interconnect Panel

800 mm

1800 mm

Daughter Cards (29 max.)

LANX, KSWX, CLKX

GPROC, GCLK, XCDR, KSWMSI



NOTES:

The CDMA Transcoder

The CDMA Transcoder frame is presented in the diagram.

• The CDMA XC design was carried over from the GSM project- much of the hardwareand software is “core” between the two technologies.

• The XC’s main functions are:

• To concentrate traffic from spans to many BTSs to few spans to the MSC. Thus cut-ting down on the cost to the customer for rented spans.

• To convert from PCM (pulse-code modulation) to QCELP (QualComm ExcitedLinear Prediction), which is the format that represents voice traffic in CDMA.

• Optional echo cancelling.

• Shares responsibility in hand-off arbitration, along with the MM.

• The XC has two main shelves, the upper and the lower. Each main shelf has a bay forhalf-size cards, and a bay for full-size cards.

• The XC initially may be configured with only one shelf, and can be configured to have upto 13 shelves (this is the LAN constraint- other constraints would limit this numberfurther).



The CDMA Transcoder Cards

• GPROC- generic processors. Function depends onsoftware downloaded. Functions are: FEP, OMP,CPP, Shelf.

• GCLK- Global Clock. Provides timing for XC. MSI-Multiple Serial Interface. Terminates two T1 or E1span lines. XCDR- Performs transcoding of digitalsignal, optional echo cancelling.

• KSW- Kilo-Switch card. Provides switching of TDMchannels.

• Half-size cards: KSWX- extends TDM bus to othershelves. LANX- extends token-ring fiber optic net-work. CLKX- distributes clock signal.

BT

C 0

BT

C 1

XC

DR

1

KS

W A

GC

LK

B

GC

LK A

KS

W B

XC

DR

0

RM

TK

SW

X A

4

RM

TK

SW

X A

3

RM

TK

SW

X A

2

RM

TK

SW

X A

1

RM

TK

SW

X A

0

KS

WX

A2

KS

WX

A1

KS

WX

A0

EX

P

EX

P

EX

P LAN

X A

LAN

X B K

SW

X B

0

KS

WX

B1

EX

P

EX

P

KS

WX

B2

EX

P

RM

TKS

WX

B0

RM

TKS

WX

B1

RM

TKS

WX

B2

GP

RO

C 1

GP

RO

C 0

MS

I 0

MS

I 1

MS

I 2

MS

I 3

XC

DR

2

XC

DR

3

MS

I 4

XC

DR

4

XC

DR

5

XC

DR

6

XC

DR

7

XC

DR

8

XC

DR

9

XC

DR

10

XC

DR

11

CLK

X B

0

CLK

X B

1

CLK

X A

0

CLK

X A

1

CLK

X A

2

CLK

X B

2

RM

TKS

WX

B3

RM

TKS

WX

B4

LOC

AL

KS

WX

A

LOC

AL

KS

WX

B

U0

U1

U2

U3

U4

U5

U6

U7

U8

U9

U10

U11

U12

U13

U14

U15

U16

U17

U18

U19

U20

U21

U22

U23

U24

U25

U26

U27

U28

L0L1L3L5L6L7L8L9L10

L11

L12

L13

L14

L15

L16

L17

L18

L19

L20

L21

L22

L23

L24

L25

L26

L27

L28

XC

DR

12

XC

DR

13

XC

DR

14

XC

DR

15

XC

DR

16

XC

DR

17

XC

DR

18

MS

I 5

MS

I 6

MS

I 7

MS

I 8

MS

I 9

MS

I 10

MS

I 11

GP

RO

C 2

GP

RO

C 3

GP

RO

C 4

GP

RO

C 5

GP

RO

C 6

GP

RO

C 7



NOTES:

The CDMA Transcoder Cards

• The full-size cards are:

• GPROCs (generic processors). Based on 25 MHz MC68030 CPU, 16 Mbytes RAM,1Mbyte EPROM, Dual redundant 16 Mbps Token Ring LAN, dual redundant TDMhighways (32 channels @ 64Kbps).

• GCLK (clock reference card), used as a single clock reference for the entire XC(there is a redundant GCLK card in case the primary fails).

• MSI (Multiple Serial Interface). This card terminates two E1 or T1 span lines.

• The XCDR cards- perform transcoding and optional echo cancelling.

• The BTC (bus terminator card).

• KSW (Kiloport SWitch) cards. This is a TDM switching matrix which supports4096 input ports and 1024 output ports (each port at 64 kbps or 56 kbps).

• Half size cards extend the clock circuits and KSW bus between shelves, and interface theXC’s internal bus with the token-ring LAN. They are the LANX (extends token ring lanto multiple cages), the KSWX (extends TDM highway to multiple cages and multipleKSW cards), and the CLKX (extends clock reference to multiple cages).

• The GPROC cards may take on three different roles:

• Front-End Processor (FEP GPROC). This processor routes control traffic betweenthe XC devices, BTSs, and the MM. It converts between LAPD and TCP/IP so theMM can talk to the BTSs.

• Operations & Management Processor (OMP GPROC). This processor is the masterGPROC for the XC subsystem- it performs operations and maintenance (O&M),fault management, and device state management for all the devices in the XC. TheOMP also has an RS-232 interface for management and debugging of the XC.

• Call Processing Processor (CPP GPROC). This device carries out orders dictatedby the MM for channel allocations and call processing functionality.



Transcoder Connectivity

• Connections from MM to XC’s CPP and OMPGPROCs are called “mxldl”.

• Connections from MM to BTS are called “csldl”.

• Voice paths to BTSs are called “traffic”, while voicepaths to MSC are called “circuit”.

• A+ path through XC is called “c7sl”- C7 SignallingLink.

• The KSW card provides switching of all TDM trafficbetween the XC’s cards- including nailed circuitssuch as csldl, mxldl, and c7sl.

BTS BTS

MSI MSI

MSI

FEP

MSIGCLK

E1 E1

E1To MSC

KSWXCDR XCDR

To MM

csldlmxldl

c7sl

TokenRingLAN

ToOMC-R

Bridge

To MM

QCELP

PCM

circuit

LANX

A+

OMP

CPP

traffic



NOTES:

Transcoder Connectivity

This diagram illustrates connections made through the XC in order to providethe XC’s functionality.

• This diagram is for a remotely located MSC, since the A+ links are being nailed throughthe XC.

• E1 (or T1) links are always terminated on MSI cards- each card can terminate two links.

• The MM establishes TCP/IP Virtual Circuits in order to communicate with the CPPGPROC, the OMP GPROC, the FEP GPROC, and to each of the BTSs.

• Inside the XC, the FEP converts the TCP/IP packets to LAPD packets, and forwardsthose packets to the appropriate device.

• In the XC’s internal configuration database, the nailed connections made for MM toBTS control paths are called “CSLDL” for “Cell Site LapD Link”, while the connectionsmade for MM to CPP and OMP GPROCs are called “MXLDL” for “MM to XC LapDLink”.

• The nailed connections made for MM to MSC (A+) paths are called “C7SL” for C7Signalling Link.

• The KSW card provides switching of TDM slots over a dual (in/out) TDM highway towhich the XCDR, MSI, FEP, CPP, and OMP GPROCs have access. Thus any port on theIN path may be connected to any port on the OUT path by the KSW.

• The OMP GPROC also provides an RS-232 interface which is used to input XC MMIcommands to perform Recent Change commands in order to change the XC’sconfiguration, or to perform debugging and/or status commands on the XC. Thisinterface is also available at the OMC-R via a dial-up modem (when the OMC-R is notco-located), or through a dedicated RS-232 line to the OMC-R.



The CDMA SC9600 BTS

• SC9600 BTS is composed of three Frames:- RF Modem Frame. Converts between a terrestial

interface (CDMA Strau Data over LAPD link) toCDMA Air Interface (layer 1).

- Site Interface Frame. Interfaces the site’s anten-nas with the LPA frame(s), and the RFMF(s).

- Linear Power Amplifier Frame. Contains up tothree linear power amplifiers, each capable ofamplifying multiple CDMA carriers.

- Redundancy is provided for critical elements.

• All cabling is through the top of the frames.

RF Modem Frame Site Interface Frame

Linear Power AmplifierFrame

LPAFrame

A

CB

A - 2400 mmB - 600 mmC - 1350 mm

RFMFrame

400 Kg 360 Kg

SIFFrame

320 Kg

min. 150 mm Air Flow

min. 600 mm Front Access



NOTES:

The CDMA BTS

The CDMA BTS is composed of three frames:

• The Site Interface Frame (SIF)

• The RF Modem Frame (RFMF or sometimes just MF)

• The Linear Power Amplifier Frame (LPA)

The BTS provides the interface between the CDMA infrastructure equipmentand the air interface.

• A SC9600 CDMA system must have at least one of each of the three shelves mentionedabove. Other configurations (such as SuperCell 4850) may not necessarily be under thatconstraint. An SC9600 BTS may have multiple RFMF and LPA frames, but only oneSIF is needed at a BTS.

• Antenna configurations supported are Omni-Receive, Omni-Transmit (Omni-Omni),and 3 sector transmit/receive.

• The frames work with +27 Vdc power.

• Maintenance may be performed locally at the BTS through the use of an LMF (LocalMaintenance Facility), which is a customized portable PC.

• Direct dial-up modems are usually connected to the site in order to enable the OMC-Rto interface with the BTS when specialized site Operations & Maintenance must beperformed. This step is usually only necessary when the site has lost its T1 or E1 link tothe CBSC.

• Redundancy is provided for critical devices in the BTS, such that failure of a singledevice will not cause a service-affecting outage at the site. However, some device failures(such as a Multiple Channel Card outage) will result in lower capacity at the site.

• Physical interfaces for the BTS are presented at the top of the frames, thus allowing foreasy access to the cabling (this is a general SuperCell design philosophy).



The CDMA SC9600 BTS Devices

• RFMF contains:- GLI cards: terminate T1|E1, provide site O&M.

- GPS & Loran cards (CSM & LFR): system clock.

- BBX cards handle the RF/Baseband conversion.

- BDC cards handle Pilot Channel and SignalSpreading on the Tx path, antenna diversity onthe Rx path.

- Preselectors, AMRs: RF and alarm functions.

• SIF contains filters for the Rx path, multicouplers toallow multiple RFMFs at one site, and directionalcouplers to patch in an RFDS test station.

• Dir.Couplers• Multicouplers• Frame Power• RFDS• Filters• RGLI

• LPA Modules• Frame Power• LCIs

LINEAR POWERAMPLIFIER

FRAME

Digital interface

RF interface

Ethernet

RS232LMF

Span Lines

To/From

BTS Control

SCAPTCP/IP

XCVRRF Outputs

RF Rx Channels

RF Tx Channels

Tx/Rx Antenna Systems

modem

(opt)

GPS and Loran AntennasRF MODEM

FRAME

SITEINTERFACE

FRAME

Other interface

AMR Bus for CDMA

CBSC

• Preselectors• AMRs• BBXs• GLIs• CSMs & LFR• MCCs

LMF



NOTES:

The CDMA SC9600 BTS Devices

The RF Modem Frame-

• Contains the GLI card (Group-Line Interface), which terminates the T1 or E1 spansfrom the CBSC, and also performs site O&M (Operations & Maintenance) for the BTS.It also houses equipment for GPS (Global Positioning System) signal acquisition, inaddition to a Loran system in case GPS fails.

• In addition, the frame houses the MCC (Multiple Channel Cards), which perform thecomplicated Layer 1 and Layer 2 functions of the IS-95 standard. These functionsinclude forward power management, signaling on the traffic channel, and traffic channelmanagement.

• The RFMF houses the BDC card. The BDC card combines the forward baseband signalsfrom each of the MCC cards and performs the direct-sequence spreading of the signal.In addition, the RFMF also houses the BBX card. The BBX card receives signal from theBDC card, and is responsible for the CDMA RF functions (up-conversion, down-conversion, etc.).

The Site Interface Frame-

• Provides the RF interface to the propagation medium (air).

• Consists of RF filters, directional couplers, an RFDS (Radio Frequency DiagnosticSubsystem), multicouplers (to support more than one RFMF), and antennas.

The Linear Power Amplifier Frame-

• Provides power amplification for transmit signals from the BTS.

• Houses up to three Linear Power Amplifiers (LPA’s), each capable of amplifyingmultiple CDMA carriers from multiple RF Modem Frames.



CDMA Software Overview

• CDMA uses UNIX on MM, OMC-R, AP, and LMF.

• Most interfacing will occur through CLI environ-ment, although emergencies will require UNIXknowledge, and knowledge of debuggers available ineach platform.

OMC-R

Transcoder

OMC-R

MM

OMC-R

Runs Unix, Informix, SC9600.Contains the MIB for SuperCell.Provides a CLI environment.Unix K-shell programming.Provides a debugger, tracer, and

Same as OMC-R, but no MIB

DB in the form of CDFs.

system kernel analyzer.Also provides X-Windows.

or X-Windows.

Runs VRTX OS.

which is downloaded from a PC.

Provides a CLI-ish interface through

Has its own DB,

RS232 cable.Provides a monitor and message filter utility

LMF

Runs Solaris UNIX.OS is pre-installed.LMF application isloaded by operator.



NOTES:

CDMA Software Overview

Properly configured and running software is crucial in acheiving a functionalCDMA system.

• CDMA utilizes UNIX on many of its platforms (MM, OMC-R, LMF, AP/UNO).Unfortunately, a careless operator with root access can cause serious damage to thesoftware on any of the platforms, rendering the platform unusable. Care must be takento only allow knowledgeable and responsible operators to have root privileges. Evennon-root users have the potential of affecting the performance of a platform- by startingup too many processes at once, for example.

• Note that the database is stored on the OMC-R as an Informix MIB (ManagedInformation Base), while on the MM the database is in the form of ASCII files calledCDFs, and on the XC the database is a separate entity. These three databases must be inagreement on those items which are shared among them.

• The LMF (Local Maintenance Facility) is a portable PC which runs Sun’s Solaris forPortables Operating System. This PC is pre-loaded with the Sun OS from the factory,but must be loaded with the proper LMF Application Software (distributed with theSuperCell software releases) by the operator (usually an RF Cellular Field Engineer). Inaddition, this PC must have the proper CDF (configuration data files) for the cell-siteswhich are to be optimized with the LMF.



CDMA System Database Creation

• Customer’s requirements (# of subscribers, area tobe covered, etc) must be determined.

• Run the CDMA Static System Simulator (CSSS) todetermine the cell site placement and configurationto cover the required area properly.

• Run the CDMA Configurator to determine the num-ber of resources needed for this CDMA system(#MSI cards, #XCDR cards, etc.).

• Run the CDMA Configuration Manager tool to gen-erate CLI and MMI scripts. Scripts used to generatea MIB on the OMC-R and a XC database object onthe XC.

Establish Requirements

Run CSSS and Configurator

Run CDMA Configuration Manager

OMC-R

Transcoder FrameOMC-R

Run CLI scripts produced inStep 3 on the OMC-R, runMMI scripts produced inStep 3 on Transcoder.

Export the MIB to a Tape,Download the XC databaseobject to a diskette.



NOTES:

CDMA System Database Creation

In order to get the CDMA equipment to function properly, the equipment data-bases must be populated correctly. These databases specify how the CDMA sys-tem has been equipped. The following steps are given to provide an overview ofthe system database generation (or “sysgen”, for short) process:

• First the Motorola personnel must dialog with the Customer in order to determine therequirements for the CDMA system. Among those requirements are: coverage area,estimated erlangs (subscriber loading), and the total number of subscribers to be served.

• Then Motorola personnel (usually Systems Engineers) must input that data into asoftware package called the CDMA Static System Simulator (or CSSS). The CSSS is acollection of programs which will help determine an adequate placement for BTSs inorder to cover the area specified. NOTE: The CSSS algorithms are being incorporatedinto Motorola’s NetPlan software. The CSSS will soon be replaced by NetPlan! ContactNetPlan project management for a release date.

• Another software package called the CDMA Configurator is run to determine theamount of hardware which will be necessary to comply with the customer requirements.

• Any additional parameters that are needed are filled in by the Systems Engineer, usuallywith the help of the CDMA Motorola Recommended Parameters document.

• A set of ASCII files are then created which are the minimal amount of informationnecessary to specify the CDMA network elements. These files are fed to a programcalled the CDMA Configuration Manager. This program produces CLI script files andMMI script files.

• The CLI script files are run on an OMC-R (or “equivalent”- such as a developmentmachine which sort of emulates the OMC-R) in order to produce an Informix MIB. TheMIB is exported to a cartridge tape, which is then ready to ship to the site.

• The MMI script files can be run on-site on the XC, or in a lab on the XC, in order toproduce an XC database object. If a lab was used, a diskette containing the databaseobject is shipped to the site.



CDMA Software Installation & Initialization

• 1st: Obtain installation tapes. Install software onOMC-R, MM, AP, and XC platforms.

• 2nd: Obtain System Database Tape & Floppies,install MIB on OMC-R, XC Database on XC.

• 3rd: Generate CDF files on OMC-R, propagatethem to the MM.

• 4rd: Perform a software INIT command on MM andOMC-R.

• 5th: MM will automatically download those BTSswhich need new code & data downloads.

• 6th: BTS will start providing service as soon as pos-sible.

OMC-R

MM

OMC-R

Transcoder Frame

OMC-R

BTS

#1 #2

#2

#3#4

#4

#5



NOTES:

CDMA Software Installation & Initialization

The previous diagram illustrates the abbreviated steps necessary to load a newsoftware release onto the Supercell system.

• This is not intended to replace the Software Installation Procedures (which should bepresent with the new load to be installed), it’s just presented as an overview of the typicalinstallation process.

• The Release Notes that accompany a release, along with the Installation Procedures willspecify which network elements, and which software packages will need to be installedas part of the release. There are major releases, and “point-fixes”. Major releases maycause all the software in the system to be upgraded, along with some hardwarecomponents as well. Minor releases usually require only a subset of the software to beupgraded.

• The software release media is usually shipped to the site. At times the software will beFTP’ed to the Area Office, where tapes are made and then physically transported to thesite. The same can be said about the site-specific database (both the MIB and the XCdatabase image).

• Once the software is loaded on the OMC-R, it can be installed from the OMC-R directlyto the sub-tending MM’s through the ethernet interface. Another option is to install thesoftware directly onto the MM’s. Yet another option (which is usually done oncommercial systems) is to perform a “Brain Transplant” on the MMs, which involvesswapping out half of the MM’s hard drives for new drives with the new release pre-installed on them.

• A process called “CDF Gen” is then performed on the OMC-R in order to create newCDF files for the MM. The CDF files are transferred to the MM via the ethernet LAN.

• The system software is then re-initialized (by performing a CLI INIT command). Thiscauses all the devices under an MM to be re-loaded if the load identifier in the CDF filefor the device has changed.

• The BTSs will start providing service as soon as possible (as soon as the minimal amountof devices required have been downloaded and initialized by the MM).



Customer Interfaces to CDMA

• Main customer interface through UNO platform.

• UNO provides GSD (Geographical System Display),an Alarm Manager, and a Status Display.

• The UNO can be remote, or local to the OMC-R.

• Other devices have more rustic interfaces (CLI,MMI)

Bridges andRoutingEquipment

OMC-R

ExternalNetwork

Ethernet

X - Terminals

Remote MMI Terminals

MM

RedundantEthernets

UNO

OMC-R

Ethernet

X - TerminalsLocal MMI Terminals

UNO

Transcoder Frame

MM

OMC-R

Transcoder Frame

OMC-R

Console Terminal

RS-232

RS-232

RS-232

Ethernet

Console Terminal

XC-PC



NOTES:

CDMA Customer Interfaces

Several interfaces are in place so that the customer may interface with theCDMA system.

• MMI stands for “Man-Machine-Interface”.

• Note that MSC interfaces are not presented, since the MSC is not considered part ofSC9600. The MSC would have its “normal” User Interfaces- in the case of an EMXswitch, these would be the Switch Terminals and (optionally) the CAT terminals whichare connected to a SwitchMate box.

• The main interface is provided through the UNO boxes, which are server platforms forone or more X-Terminals. At either the UNO or the X-Terminals the customer will beable to bring up graphical user interfaces (GUI’s) which will detail the state of the RadioNetwork Devices. The customer may also view statistics reports and perform systemmaintenance from any of the X-Terminals or AP boxes.

• The UNO provides software packages which allow the operator to manage the SC9600network, such as the Status Display Tool, the Geographical Systems Display, and theAlarm Manager.

• Other interfaces are less user-friendly, such as the console terminals of the OMC-R, andthe MMs, as well as the XC-PC interface to the Transcoder.

• SC9600 commands may be performed at the console terminals, or through UNIX telnetsessions connected to the MM or the OMC-R platforms. The software which parses andstarts execution of SC9600 commands is called the “CLI”- for “Command-LineInterface”.

• The XC has a CLI-like environment which may be accessed over an RS-232 serial linewhich is connected to the OMP GPROC. This is a non-user-friendly environment, and isusually used only by XC specialists.

• An UNO cluster may be remotely located, by using bridges and routers and to carryhigh speed serial lines over dedicated T1 or E1 spans. However, this option has not yetbeen used by any CDMA customer to date.


Conclusions:• SuperCell CDMA product is a distributed environ-

ment for providing cellular telecomunications.

• SuperCell system consists of CBSC, BTS, OMC-R,and UNO platforms.

• CBSC handles switching and control functions in theBSS network, along with transcoding of QCELP intoPCM (and vice versa).

• CBSC is composed of MM and XC, along with otherequipment for O&M.

• BTS handles air interface for CDMA.

• MSC is responsible for PSTN interface to SuperCell,billing and subscriber database functions.

• Several CBSCs may be connected to a single MSC,while several BTSs may be connected to a CBSC.

• UNIX is widely used in the SuperCell network.

• MM, OMC-R are Tandem platforms, running Unix.

• XC provides switching matrix, transcoding functions.

• Proper software installation and configuration is cru-cial to SuperCell.


PWID Training: CDMAThe Tandem MM Platform & UNIX OS


Objectives:

• Present the MM platform in detail- its specializedhardware as well as the SuperCell software runningon the platform, in order to equip the student with thenecessary skills to operate the MM.

• Present the basics on the UNIX operating system, inorder to give the student a working understanding ofUNIX.


Table of Contents:

The MM Architecture and Interfaces............................................................................ 77MM Hardware Architecture........................................................................................... 79MM Physical Configuration........................................................................................... 81MM Functions ................................................................................................................ 83The UNIX OS- Theory & Definitions ............................................................................ 85Supercell File System Structure..................................................................................... 87MM File System Theory & Operation ........................................................................... 89Unix: Basic Communications ........................................................................................ 91The Unix Korn Shell....................................................................................................... 93Supercell MM Software Overview ................................................................................. 95Supercell MM Software Architecture ............................................................................ 97MM Platform Initialization............................................................................................ 99Supercell Initialization: SCSIIM................................................................................. 101The Supercell Process Set ............................................................................................ 103Supercell Process Set- Continued ................................................................................ 105MM Software Installation ............................................................................................ 107MM Operations- Tools & Tricks .................................................................................. 109

PWID Center for Excellence Revision 0, 4/24/97 The Tandem MM Platform & UNIX OS


MM Architecture and Interfaces

• MM is a fault-tolerant UNIX box built by TandemComputers.

• At least two instances of every critical device.

• The Operating System is UNIX SVR4.

• Supercell application software interfaces with UNIXto perform Supercell functions.

• MM interfaces with MSC, BTSs, OMC-R, XC, andits local console terminal.

• MSC interface is through SS7 stack running onMM’s SYNC cards.

• BTS/XC, OMC-R interfaces are through TCP/IPstacks running over MM’s ethernet cards.

• Local console terminal is through an RS-232 serialline to the MM’s I/O processor.

Sync.Cntrls

DiskCntrls

EnetCntrls

EnetCntrls

TCP/IP

MM

SCAP overTCP/IP (Ethernet &Token Ring)

Supercell Application SW

SS7

UNIX SVR4OperatingSystem

SS7

UtilitiesUNIX

To MSC (through DSUs)



NOTES:

The MM Architecture and Interfaces

The Mobility Manager (MM) is a computer platform which Motorola purchasesfrom Tandem Computers, Inc. The following points concern this platform:

• The platform is fully hardware fault tolerant, meaning that any single device failure willnot cause a system-wide (or catastrophic) failure.

• All devices except the CPUs are doubly redundant, while the CPUs are triple-modular-redundant (which means that 2 out of 3 CPUs must be in agreement).

• The ethernet cards and Synch cards are active/standby, while the disk (SCSI) controllersare active/active. An active/standby redundancy scheme means that one card will alwaysperform all the work, while the other will only come into service when the first card fails.An active/active scheme requires both devices to perform the same work at the sametime- An external entity will “watch” both cards, in order to detect failures. Thedefective card will be removed, and all work will continue to be performed on theremaining good card. A software/hardware system called “midas” is responsible for the“watching”.

• The operating system is UNIX SVR4 (POSIX compliant). The OS is the “core” softwarecomponent in the MM- it interfaces with the system hardware (such as ethernetcontrollers, sync cards, and SCSI disk controllers), and provides an environment inwhich the Motorola software can interface with the rest of the “real world”.

• The software for receiving and sending information over SS7 and ethernet interfacesresides in the hardware (rom/ram) for the interface (Async cards, ethernet cards), in theOperating System (OS) itself, and in the Motorola Software as well. The MotorolaSoftware makes software calls to the OS in order to send and receive information fromthe communications “stack” (such as a TCP/IP stack, or an SS7 stack).

• A standard VT100 terminal at 19200 baud (8 bits, no parity) may be connected to theMM in order to function as a “console terminal”. From this terminal, an operator willbe able to access the MM when all other access methods (TCP/IP) are not available.From this terminal, the MM may also be reset should all other recovery attempts fail.



MM Hardware Architecture

• 3x MIPS R4400 CPUs. 175 Mhz internal, 75 Mhzexternal clock speeds. 128 Mb memory.

• 16 Mb external (shared) memory.

• 2x I/O highways (I/O Processors control traffic).

• 2x SCSI busses, and disk drives.

• 2x SYNC controllers (for SS7 links).

• 2x ethernet controllers (for FEP/OMC links).

CPU A CPU B CPU C

MEM A MEM B

SYNC (2)Controllers

LAN (4)Controllers

VME bus VME bus

MSC ControlChannels

Ethernet LAN

MMIPort

MIRRORED DISKS

DiskController

DiskController

IOP 1IOP 0ServiceProcessors

4

2 pair



NOTES:

MM Hardware Architecture

The MM platform was especially built for Motorola by Tandem Computers forhandling the demanding real-time environment of cellular call processing, whileproviding the reliability necessary for the cellular environment. The followingare points concerning the MM platform:

• There are three CPUs, two of which must be in agreement at all times. Each CPU has128 Mb of on-board memory (fast NVRAM). The CPUs are MIPS R4400 series, runningat 150 Mhz internal clock speed, 75 Mhz external bus speed. Each CPU resides on itsown card. A CPU may be pulled out, but upon re-inserting the CPU, the Tandem boxwill automatically start a process called “re-integration”. This will bring the CPU “up tospeed” with the other two CPUs, and will be done very slowly so as to not affect the othertwo CPUs. Watch the file /var/adm/messages for indications as to when the CPU re-integration is finished.

• There are two global memory cards, each with 16 Mb of memory. This global memorymay be accessed by all three CPUs, while the CPU local memory can be accessed only bythe local CPU. Information which must be “seen” by all three CPUs is placed in thismemory. Access to global memory is slower than access to local memory (as is to beexpected).

• I/O processors provide a bus for communication to the I/O devices in the computer. Thusbus is called “VME bus”, and it allows access to the Ethernet, Sync, and SCSI buscontrollers by the CPUs.

• The disk controllers each control a SCSI chain of devices, which include the hard diskdrives, cartridge tape drive, and an optional DAT tape drive. A software suite called“Veritas” provides low-level formatting and access to the disk volumes, in order tohandle active/active redundancy.

• The Sync controllers are also called V.35 controllers (V.35 is the name of thecommunications standard provided by the controllers). The Sync controllers providefour ports each, labelled 0 through 3. These ports are used for connections to the MSC(through DSUs).

• The ethernet controllers provide one ethernet port each. There are four ethernet cardsin the MM- two for active/standby connections to the XC (FEP GPROCs), and two foractive/standby connections to the OMC-R.



MM Physical Configuration

• Each member of 2x set of cards on different powersupply.

• Most cards hot-swappable by using the cfdown andcfonline commands.

• Optional DAT tape drive may be used.

• LEDs on the cards indicate card status (red isBAD!).

• “cfstatus” command may be used to show card sta-tus.


Air intake

Air Exhaust(bottom, front and rear)

Fan Modules(3)

Fan Modules(3)

Bulk PowerSupply Mod-ule

151 152 161 162 163 164 165 166 167 168

SPC SPC Ether-net

SCSI SCSI Ether-net

SS7 Ether-net

SS7 Ether-netor

SPCBypass

}

c0d0

c1d1

c1d6

c1d0c0d1

c0d2

1 gig Disk

1 gig Disk

1 gig Disk1 gig Disk (tape)c0d4

201

202

203

211

212

213

221

222

223

c0d3

c1d5

DC PWR (75

Mem

ory

A

Mem

ory

B

CP

U-A

CP

U-B

CP

U-C

I/O P

roce

ssor

0

I/O P

roce

ssor

1

}I/O Controller Card Shelf

Processor Card Shelf

Storage Device Bay

0 1 2 30 10 1



NOTES:

MM Physical Configuration

The diagram illustrates the physical position of the cards on the MM. Thesepositions are dictated by the MM Hardware Installation document. The follow-ing points concern the MM’s hardware:

• All removable devices may be queried for status by using the following command(partial output is shown as an example):

# cfstatusComponent State Cabinet IOP/Controller DCC CRU______________ _______________________ __________ ____________________ _______enet1 online system iop0 DCC3scsi1 online system iop0 DCC3

• All removable devices may be placed out of service with the “cfdown” command.Beware of using this command! A careless operator can bring the MM down throughcareless use of this command. See the cfdown manual page for more information.

• All removable devices may be placed into service through the “cfonline” command. Thecommand will need the name of the controller for the device to be specified with a “-c”argument. Example:

# cfonline -c scsi0 disk0

• Notice that the slot numbers on the top of the I/O controller card shelf correspond to thenumbers on the back of the frame.

• The cfinstall command must be used if the device is shown as “Absent” through acfstatus command. The cfinstall is used as follows (in the example, we are installingethernet card 0, which is in slot 161). See the man page for cfinstall for more details.

# cfinstall -i iop0 -s ioc0 -c enet0

• Notice that the disk drive identifiers are given in the diagram for the storage device bay.These devices are then accessed at the low level by addressing the files in the directory /dev/dsk/. For example, the 1 gig drive in the upper left hand corner would be addressedas /dev/dsk/sc0d0.

• All cards have LED indicators on them- red is not a good color! (this means the card isoffline).

• An important point to make about the SS7 cards: port 0 of the SS7 card is actually thebottom-most port on the card! (Not the top, as one would suppose).



MM Functions

• Call Control and Channel Allocation.

• Control point for handoff operations.

• Builds partial call detail records.

• Performs Operations & Maintenance actions (underthe instruction of OMC-R).

• Provides initialization, fault management and eventprocessing for itself and devices it controls.

• Runs Tests, Audits, and Diagnostics (TAD) pro-grams periodically.

• Provides control of software releases and data ver-sions on its sub-tending devices (except most XCboards).

• Collects statistics for later transmission to OMC.

• Provides Command Line Interface (CLI) for inter-action with operator.

MSC

CBSC

LAPD and TCH’s

Transcoder

OMC-R

A+

TCP/IPEthernet

BTS BTS BTS BTSCluster

DSU

BridgToken

A+ MobilityManager

Ethernet

DSU

RS-



NOTES:

MM Functions

The MM (Mobility Manager) is the “brains” of the SuperCell Real-Time opera-tions. Functions which require access to database information, or complex func-tions which must be performed quickly are delegated to the MM. The followingfunctions are delegated to the MM:

• Basic call control and channel allocation functions. This includes allocating trafficchannel elements in the BTS which is servicing an MS, as well as assigning Walsh codesfor the conversation.

• Control point for handoff operations. For handoffs between two channels that are bothunder control of the same MM, the MM can control the entire handoff. For inter-CBSChandoffs, the source MM will provide overall control, but the process will also involvethe MSC and the MM that controls the destination channel.

• Collect the cellular data needed for call detail records. This data is forwarded to theMSC, where it will be combined with the land-network information. Producing thecompleted call records is then a function of the MSC.

• Acts as an agent for operations and maintenance actions. It is important to note that theMM does not need the OMC-R in order to perform call processing tasks. However, theMM acts as a slave to the OMC-R when provisioning or changes are done to the cellularnetwork.

• Provides a system configuration database that is consistent with the data stored at theOMC-R. The MM’s version can support the real-time demands of call processing. TheMM’s version of the database is stored in files called “cdfs”.

• Provides high level fault management and event processing for itself and subtendingdevices. This is done through a software sub-system called “scdevom”.

• Provides system level diagnostics and test control in cooperation with the OMC-R.

• Provides control for distribution of software and data base downloads.

• Controls initialization of MM and all subtending subsystems.

• Collects performance measurement information for later processing and display at theOMC-R and UNO platforms.

• Provides local user interface through a Command Line Interface (CLI).



The UNIX OS- Theory & Definitions

• Operating System: set of computer instructionswhich provide software interfaces to hardware.

• UNIX: powerful multi-tasking OS.

• UNIX allocates slices of time to each stream of com-mands (process).

• SVR4 allows several scheduling classes:- System.

- Real-Time.

- Time Scheduling.

Sync.Cntrls

DiskCntrls

EnetCntrls

EnetCntrls

TCP/IP

MM



SS7


SS7

UtilitiesUNIX




NOTES:

The UNIX OS- Theory & Definitions

It is important to understand some of the theory behind the UNIX OS in order tobetter understand the operation of the MM. The following statements are pro-vided to this purpose:

• An operating system (OS) is a set of software instructions which run on a computer’sCPU in order to provide interfaces to the computer’s hardware. The interfaces are usedby application programs. The OS will generally provide an environment in which auser’s program can access data residing in the computer’s devices, as well as data whichis arriving on a telecommunications link.

• The UNIX OS was invented by AT&T in 1969- it is a powerful OS, and very popular forsuch applications as telephony, scientific studies, modeling, simulation, and many others.

• The combination of instructions for a user program, along with memory in which toperform operations is called a Unix “Process”. The function of assigning time to aprocess in an ordered manner is called “scheduling”.

• The Unix OS provides a process called the “scheduler” which has the capacity to removea process from the CPU and put another process in its place. This is called “pre-empting”. A process of higher priority may pre-empt another process of lower priority.

• Unix SVR4 provides different scheduling classes and priorities as follows:

• System Class: Processes in this class are guaranteed to run any time they need to.This is the highest priority class, any process in this class may pre-empt any otherprocess at any time. The Unix system scheduler process is an example of this type ofprocess.

• Real-Time Class: Processes in this class run until they give up the CPU of their ownaccord, or until an optional time quantum is exceeded (the time quantum isassigned by the process itself at the beginning of its execution). This class is usuallyonly used by time-critical applications. Inside this class, a process may be of prior-ity from 0 to 59, 59 being the highest priority.

• Time Scheduling Class: Processes are run in a non-deterministic manner (i.e. theprocess does not know when its going to be pre-empted). The time quantum isassigned by the unix scheduler based on a scheduling algorithm for this class. Thealgorithm takes into account several factors, including recent CPU consumption,and process priority (which may be from -20 to +20).



Unix & Supercell File System Structure

• MM utilizes Unix’s hierarchical file system.

• MM has following file systems: / (root), /var, /home, /usr, and /screl.

• Code & data files for Supercell stored in /screl.

• Software Load Management is used in Supercell.

• MM may have several releases of software in /screl.

• The README file for a release will have informa-tion on software release (including size data).

SC-1.2.1.0.0

loadable

bts fep mm omc xc

(configuration data)(cdf Files)

(slm scripts)

/screl

(platform code and data files)

(additional releases) install

(slm scripts)

bin

release

1.2.1.0.0-Release.lst1.2.1.0.0-Release.rds1.2.1.0.0-README

active

(MM/OMCexecutables)

[soft links]

bindata

(MM/OMCdata files)

loadable



NOTES:

Supercell File System Structure

The MM utilizes the Unix hierarchical file system structure. “Hierarchical”means that the structure is similar to that of the management in most compa-nies, with one entity as the “top” or “head”, several entities report to the “head”,and so on. In Unix, a directory is a place-holder for other directories and/or files.The following points refer to the file systems used by the MM:

• The MM has several file systems (which are independent of each other), each of these filesystems has its own space assigned to it. The following are file systems on the MM:

• / (root). This is the top directory, under which all other file systems are “mounted”.Mounting a file system is the process of making that file system available for use as asub-directory of the root file system. (More on this later).

• /usr. This filesystem contains several files and sub-directories which contain soft-ware libraries and binary files for the proper operation of the MM platform. Pro-grams such as “cat”, and “crash” reside in this file system.

• /var. This filesystem contains files and directories which contain system messagefiles, and other files which specify the behavior of certain Unix programs runningon the MM. The /var/adm/messages file is a very important log file to which manyprocesses will output messages.

• /screl. All the Supercell software for the MM and its sub-tending devices resides inthis file system, as well as log files and data files for Supercell. The file system isorganized as seen in the previous slide. A Supercell software sub-system called“SLM” manages the information in this file system.

• /home. All the user accounts reside in this file system. Default users are “scadm”,“lmf”, and “a7adm”- these are created during a software installation of the MM.

• An MM may have several Supercell releases on it at the same time. The actual numberwill vary, depending on the release size, and the capacity of the /screl filesystem. At presstime, the MM could hold a maximum of six full releases of Supercell software at once.

• The README file for a release will have information such as release dependencies(version of A7 software required, whether or not a full XC reset is required, etc.), andrelease size information.



MM File System Theory & Operation

• File systems must be “mounted” for use.

• Veritas file system manages mirrored disks so allinformation resides on two disks.

• Basic unit in Veritas is “sub-disk”, which resides ona physical disk.

• One or more sub-disks make up a “plex”.

• One or more plexes make up a “volume”.

• “volprint” command prints status of veritas.

• “volplex” command re-integrates a bad plex.

Disk 0:

Disk 1:

sub-disk sc0d0-s1sub-disk sc0d0-s2

sub-disk sd0d1-s1sub-disk sd0d1-s2

}

}

Plex: test-0

Plex: test-1

Volume: test}Veritas System



NOTES:

MM File System Theory & Operation

• All file systems other than root are attached to the root file system through a processcalled “mounting”. This process allows the file system to be accessed as if it were adirectory under root. The command “mount” is used as follows to do this:

mount -F <type> [-o <mount options>] <device to mount> <mount_point>

The file system type will be “vxfs” for most file systems on the MM. Mount options(the -o option) are usually not used. The device to mount will be the full pathnameto the volume to be mounted, this will usually be /dev/vol/<filesystem name>. Themount point is the full path to a directory under which the file system will beaccessed. It is important to note that if the directory contains any information in itbefore the mount, the information will be “hidden” after the mount command.

• The “mount” command (with no options) will print information on the file systemsmounted, as in the following example:

/ on /dev/vol/root read/write/setuid on Mon Apr 29 13:37:27 1996/usr on /dev/vol/usr read/write/log/setuid on Mon Apr 29 13:37:43 1996/var on /dev/vol/var read/write/log/setuid on Mon Apr 29 13:37:45 1996/screl on /dev/vol/screl read/write/log/setuid on Mon Apr 29 13:40:16 1996/home on /dev/vol/home read/write/log/setuid on Mon Apr 29 13:40:19 1996

• The Veritas system uses the concept of a “sub-disk” as the lowest unit of file storage. Asub-disk will be directly associated to a particular area on a physical disk. The next unitis the “plex”, which is composed of one or more sub-disks. A plex is usually composed ofsub-disks which are all on the same physical disk. The top-most unit is the volume,which is composed of one or more plexes.

• The veritas command “volprint” will print information such as the following:TYPE NAME ASSOC KSTATE LENGTH COMMENTsd sc0d1s3-m3 screl-0 - 1375176sd sc1d0s3-m3 screl-1 - 1375176plex screl-0 screl ENABLED 1375176plex screl-1 screl ENABLED 1375176vol screl fsgen ENABLED 1375176

• The veritas command “volplex” can be used to re-integrated a FAILED volume,although usually this command is performed automatically by the MM when a failedvolume is detected. The following is an example volplex command (see the man page forvolplex for more details):

volplex -U gen -o iosize=16k -o slow=200 -- att screl screl-1



Supercell: Inter-Process Communications

• SC processes receive communications through:- Shared Memory.

- Messages.

- Pipes.

- Signals.

• Shared Memory segments are set up by a processcalled “scsiim”.

• Use the “ipcs” command to check IPC status.

IPC Msg.“Tickle

TCP/IPMessages

SS7 / J7Messages

OperatingSystemServices

Poll(2) Interface

APISharedMemorySegment

IPCMessageQueue

IPC MessageQueues for OtherProcess

MsgSend

Application

TCP/IPEthernetProtocolStack

SS7 / J7DriverProtocolStack

NamedPipe(FIFO)

NecaComm/AplusComm

ScapCommAPIIPC Msg.Handler

Application

Timer

Shared Mem.Allocator

API Shared Memory Segment

Other ApplicationShared Memory Segments



NOTES:

Unix: Basic Communications

Unix processes often need to be “told” of events that have occurred in the system,or they need to “talk” to other processes in the system. The OS provides severalways for a process to receive indications from the outside world- following aresome points concerning Inter-Process Communications (IPC):

• A process may receive or send a signal, which causes a pre-defined routine to beexecuted upon reception of that signal. All possible signals are defined in the file /usr/include/sys/signal.h. The key sequence “cntrl-C” produces a signal 2 (SIGINT), whileother key combinations can produce SIGTERM (signal 15). These signals generally havethe effect of causing a process to stop execution and exit. Care must be taken when usingthe “kill” command (this command causes signals to be delivered to processes).

• A process may receive or send a message from/to another process. The receiving processmay pick the message up from a queue (which is provided by the OS), and perform someoperation based on the data received.

• A process may receive or send data through a “pipe”, which is a special type of file towhich other processes may write data. Reception of data on that pipe is usually detectedby performing a poll() system call to the OS. The poll() system call will “wake up” only ifdata is received. Thus, the process will go to “sleep” until data is received over the pipe.

• A process may receive or send data through shared memory. Shared memory is usedextensively in Supercell to hold data which must be accessed by more than one process ata time. Shared memory is created by a process called SCSIIM (more on this processlater). However, in order for a process to be notified that a change has occurred inshared memory, another method of IPC is used (such as a pipe, signal, or message).

• The Supercell processes will sit in a poll() system call until “something” happens. Thepoll() system call is “listening” to several sources, among which may be pipes for EventMessages (special data which is placed in shared memory queues), pipes for A+messages arriving over SS7 links, and pipes for SCAP messages arriving over TCP/IPlinks to the XC FEPs.

• The “ipcs” command may be used to check the status of the shared memory segments.



The Unix Korn Shell

• A shell is an environment presented to user uponlogin into platform.

• Korn Shell (or Ksh) is one of the most popular shells,and is used by the Supercell product.

• Knowledge of Ksh is important!

• Example Constructs: “for” and “if-then-else”.

• Ksh also provides file name expansion, testing, sub-stitution, etc.

• From inside a Ksh session, a CLI session is started.

• The CLI session is really a Ksh session with morecommands available.

• Good reference material exists for Ksh- highly rec-ommended!

OMC-R

Wow! Ksh is great!



NOTES:

The Unix Korn Shell

In order to interact properly with unix-based computers, an operator needs toknow a unix shell. There are several shells available (such as C Shell, B Shell,and Korn Shell), but the MM uses only Korn Shell (or Ksh). The following pointsare given concerning the Ksh:

• When a user logs into the MM, a Ksh session is started. This session allows the user tointeract with the computer, offering the ability to start programs, status programs, andeven build complex programming language constructs to get jobs done.

• An example construct is the “for” loop. For example, if a user wants to create severaldirectories called cdfdir1, cdfdir2, cdfdir3, and cdfdir4 the user can write the followingat the ksh prompt:

# for name in 1 2 3 4# do> mkdir cdfdir${name}> done

• Another example construct is the “if-then-else” test clause. This can be used to performsome action based on a condition which must evaluate to true. The following is anexample where if argument #2 to a function is not writable, the file is made writable, /tmp/newFile is written over it, and then the file is made not writable. Otherwise, the file /tmp/newFile is simply written over it.

if [ ! -w $2 ] then chmod +w $2 cp /tmp/newFile $2 chmod -w $2 else cp /tmp/newFile $2 fi

• Once the operator has logged in, a CLI session is started simply by starting a programcalled “CLI” from inside the ksh session. A CLI session is really nothing more than aksh session which has been customized so that the prompts are presented with sequencenumbers, and the PATH environment variable has been modified so that access to Unixcommands is restricted (note: simply type “export PATH=$UPATH to get around this).

• Excellent reference books on Ksh abound- one of them (highly recommended) is “TheKornshell Command and Programming Language” by Bolsky and Korn, published byPrentice Hall.



Supercell MM Software Overview

• Supercell software running on MM:- Supercell System Initializer- SCSIIM.

- SS7 process set.

- Software Load Management- SLM.

- Supercell Device Operations- SCDEVOM.

- Volume Data Transfer- VDT.

- Performance Measurement- PM.

- Call Processing- CP, CLCP, CPMO.

- Event Management- SCEVMGR.

- Command-Line Interface- CLI.

- Gateway- MM_GW.

Sync.Cntrls

DiskCntrls

EnetCntrls

EnetCntrls

TCP/IP

MM



SS7


SS7

UtilitiesUNIX




NOTES:

Supercell MM Software Overview

The following are the main software sub-systems involved in Supercell, alongwith brief explanations of the functions covered by the sub-systems:

• The SCSIIM sub-system is responsible for establishing the Supercell application processset. The process is started during the MM platform’s normal startup sequence (as partof a Unix re-boot or platform cold start).

• The SS7 software sub-system is responsible for establishing SS7 connections over theMM’s V.35 cards to the MSC. This sub-system consists of several processes whichperform different functions related to the SS7 links (monitoring, sending/receiving dataover the links, routing, etc.).

• The Software Load Management sub-system (SLM) manages the software releases forthe MM and its sub-tending devices. Commands such as INSTALL, DEINSTALL, andACTIVATE are handled by SLM.

• The Supercell Fault Management sub-system (SCDEVOM) manages the MM’s sub-tending devices, taking the appropriate actions should a device go out of service. It isalso responsible for loading and initializing the devices.

• The Volume Data Transfer sub-system (VDT) is responsible for establishing TCP/IPconnections to devices and transferring data to and/or from the devices. Code and datafor devices needing download is sent through the VDT sub-system.

• The Performance Management sub-system (PM) is responsible for collecting statisticspegs and counts, and forwarding this information to the OMC-R. Information such asMCC element usage and call completion ratios is collected by the PM sub-system.

• The Call Processing sub-system is responsible for the call processing activities on theMM. It consists of Connection-Oriented Call Processing (CP, for short), Connection-lessCall Processing (CLCP), and Call Processing Maintenance and Operations (CPMO).

• The Event Management sub-system is responsible for propagating alarms and eventreports to all the proper destinations.

• The Command-Line Interface sub-system (CLI) is responsible for the basic userinterface functions at the MM.

• The Gateway sub-system (MM_GW) is responsible for establishing and monitoring aTCP/IP connection with the OMC-R.

PW

ID C

enter for Excellence

Revision 0, 4/24/97

The T

andem M

M P

latform &

UN

IX O

S

96

of 2

59

Mo

toro

la C

on

fide

ntia

l Pro

prie

tary

Supercell M

M S

oftware A

rchitecture

sc_devom

VDT

RealTimeDatabase

VDTControl &Buffering clcp

cpmo

EventManager

mmgw

Command

UNIX Shell

TCP/IP

Data DnldFile

Disk CDFAccess

LogEvents

TCP/IP-w- OMC Code/Data

Files andUploads

TCP/IP

SW Load

DB Updates

DB

ResourceAllocation

DB Access

AccessDB Access

PagingCntl &Call Assign.

Trace Uploads

ResourceAllocation

API

UNIX Operating System

scsiim

SharedCDF

DGM&SJ7/SS7 Pkg.

scdb

L

o

g

C

D

F

s

dgen

(RTOs)

R/W

VDTCntl

VDTCntl

ApplicationSegment

CDFAccess

TraceCntl

APISegments

Info

mmslmLoads

EventReports

Spooler

EventReports

CLI Session

SWLoadControl

pm_collect

sec_mgmt

UNIXTrigger

SecurityAlarms

callproc1

J7NodeControl

J7/SS7



NOTES:

Supercell MM Software Architecture

The purpose of the previous page is to present the “whole picture” of the Super-cell MM software. The following points are presented on the overall architec-ture:

• The UNIX OS is bottom-most layer of software, all access to physical devices by theSupercell processes must be made through UNIX. The operator must also use UNIXwhen equipping new cards on the MM, for installation of the MM software, as well asfor general operations of the platform (cleaning up files, etc.).

• A software entity called the Application Programming Interface (API) “lives” betweenthe OS and the Supercell processes. In reality, the API consists of a software librarywhich is linked in to the Supercell processes, along with a lot of shared memory which isestablished by the SCSIIM process. The software applications use the API in order toperform common tasks (such as sending and receiving messages, publishing alarms andevent reports, setting/clearing timers, etc.).

• At the top left corner of the previous page (when viewed in landscape mode) is theSCSIIM process. This process is the first Supercell process brought up by Unix. In turn,this process sets up the shared memory environment and starts bringing up all the otherprocesses in the diagram. Once SCSIIM has finished bringing all the processes up, it just“listens” for processes that die, and performs the appropriate action. So, SCSIIM is the“mother” of all Supercell processes.

• These processes have a great deal of debugging information built into them- when aproblem is suspected in a particular area (such as Call Processing), the debuggingmessages for CP can be activated by using a program call dbapi.

• Some of the processes in the diagram show up for the first time; sec_mgmt, which standsfor “Security Management” sends an alarm when someone has exceeded the maximumamount of failed login attempts. The dgen process produces data files to be downloadedto the MM’s sub-tending devices (such as MCC cards). The pm_collect program is thePerformance Measurement process. Note that all the processes are ovals, while squaresare either shared memory, or “entities” such as UNIX or the API.



MM Platform Initialization:

• Power up, I/O Subsystem Initialization- I/O cardsare recognized and initialized.

• Unix loaded from the boot partition, all cards andstorage units are recognized, memory initialized.

• Start system daemons.

• Check, Mount File Systems..

• Run startup scripts. The /etc/inittab file is “master”,then /etc/rc<N>.d scripts are executed.

• Present login prompt at console. User may thenlogin, and supercell processes should be coming upat this time.

Reset I/O Init UNIX Load

Daemons File Systems Startup Scripts

Login Prompt



NOTES:

MM Platform Initialization

• The MM will go through its start-up sequence at these times: When the MM is poweredup (both switches on the backs of the Service Processors are in the “on” position), orwhen the MM is reset through a “hostreset host” command at the console, or when theMM is reset through a “cntrl-CORE” key sequence at the console, or when the MMreboots itself due to a PANIC condition.

• As a general note: The MM’s console is the ONLY location where information on theMM reboot can be seen in real-time. So, it is important that the console be monitoredduring a reboot in case any anomalies should occur.

• First the MM will do an I/O initialization, where all the card hardware is recognized.Messages such as the following will be seen at the console:

Cabinet configuration: Central OfficeIdentified Interphase Ethernet Controller in iop 0, slot 0Identified Interphase Ethernet Controller in iop 1, slot 0

• Then the Unix OS is loaded from the boot partition. In order to access the boot partition,the command “mkbootdisk” may be used. The boot partition is special- it can only beaccessed through the mkbootdisk command (it is not normally mounted). Messages suchas the following will be output when Unix is loaded:

loading file disk(1,0,5)unix2349616+390992+1116272text entry:0x80031000, start: 0x80031000, end:0x8026ea2fdata start:0x8026ea30, end:0x802ce17fbss start:0x802ce180, end:0x803de9ef

• System daemons are then started. A daemon is a process that runs in the background,usually fielding asynchronous requests. Examples are the rpcd (Remote Procedure CallDaemon), and the inetd (internet services deamon).

• The init process is started, which then reads the /etc/inittab file and starts processesspecified in that file for the run level in which the machine is coming up. The /etc/inittabfile has a value called inittdefault, which sets the run level of the machine to 4 forSupercell. At run level 4, a script called /sbin/bcheckrc checks several sub-systems in theMM- including the file systems. The init process then runs the script /sbin/rc4, which inturn will run all the scripts in /etc/rc4.d.

• It is useful to look at the scripts in /etc/rc4.d, since they perform such tasks asconfiguring the ethernet cards, mounting file systems, removing files in /tmp, andresetting the SS7 cards.

• Towards the end of the /etc/inittab file, entries for a process called “ttymon” will enable alogin prompt to be presented at the console and at any tty session.



Supercell Initialization: SCSIIM

• SCSIIM:- starts from the unix init program.

- establishes Supercell system on MM and OMC.

- reads specifications for Supercell system from /screl/active/data/scsiim_cdf.mm1.

- outputs indications to /sc/scsiim/scsiim.logfile.

- brings up Supercell processes, monitors them forprocess death.

• Several types of INITs (done at CLI):- INIT Level 1 (re-boots the Tandem platform)

- INIT Level 2 (re-starts only the Supercell Pro-cesses)

- Both above may specify Force mode to be eitherYes or No, for re-reading the MM database files.

Unix Init

/etc/inittab

Unix Stuff

SCSIIM

/screl/active/data/scsiim_cdf.mm1/sc/scsiim/scsiim.logfile

Supercell Stuff

CLI INIT



NOTES:

Supercell Initialization: SCSIIM

• SCSIIM is the first Supercell process to be started up by the MM platform, after a Unixreboot.

• The scsiim process reads from the file /screl/active/data/scsiim_cdf.mm1, which specifieshow the MM platform’s Supercell software is to be configured. The shared memoryconfiguration (size, name, type, etc.) is included in this file, along with information onenvironment variables which must be established for the processes, and information onthe processes themselves. The following is an excerpt from the cdf file:

PROC = { // monitorApi - monitor proc growth & shared memory usage ... ActionOnDeath = Restart, (NOTE: some output deleted due to space). Name = monitorApi, Path = “/screl/active/bin/monitorApi”, Priority = 20, Args = “/screl/active/data/scsiim_cdf.omc -dbg ERROUT /sc/processes/monitorApi.err -dbg ALL 0 -dbg OUTPUT /sc/processes/monitorApi.out”, },

• The scsiim process will place information on what’s going on in the file /sc/scsiim/scsiim.logfile. The following is an excerpt from the file:

Fri Mar 29 00:24:18 1996 omc PID: 1710 scsiim: Attempting to spawn process:SyslogdRefreshFri Mar 29 00:24:18 1996 omc PID: 1710 scsiim: Non-lockstep process:SyslogdRefresh instantiated. Pid: 1779

• The CLI command called “INIT” is used to re-initialize an MM. The command’s formatis as follows:

cli prompt> INIT MM-1 [Level=1|2] [Force=Y|N]

• The Level may be either 1 or 2. If the level is 1, the MM platform will be re-booted via aunix “fastshutdown” command. This will cause all processes (including unix processes)to be shutdown and re-started. If the level is 2, only the Supercell software will be re-started. In any case, a Force parameter may be specified to “Y” (or Yes) if the user wantsto re-read the entire MM database from the CDF files residing on the hard drive.Normally this step is done only if corruption is suspected in the data.

• After bringing up all the processes, scsiim will monitor the processes and perform theappropriate action should a process die. The action will be specified in thescsiim_cdf.mm1 file.

• The /sc/scsiim/scsiim.logfile file is very important, especially during a platforminitialization. A good operator will keep an eye on the file through a “tail -f” command.



Supercell Process Set

• SCSIIM sequences supercell process initialization.

• “popper” starts up SS7 process set. Similar toscsiim, but only controls SS7 process set.

• Most processes started by scsiim output indicationsto files in /sc/processes/<processName>.

• MonitorAPI monitors consumption of certainresources on MM (and OMC).

• scemlogd manages ring buffer and set of files con-taining last messages sent and/or received.

• mmgw establishes and manages link to OMC.

• scevmgr sends events to all interested parties.

• sec_mgmt posts alarm on excessive failed logins.

• vdt_1, vdt_2 establish virtual connections for datatransfers on demand.

SCSIIM

popper MonitorAPI scemlogdmmgw

scevmgr

sec_mgmt

vdt_1, 2



NOTES:

The Supercell Process Set

• The SCSIIM process will sequence the start-up of all the Supercell processes. It isimportant to watch the initialization sequence carefully, since several processes are oftype “Mandatory”, meaning that if the process cannot initialize properly, the rest of theinitialization will be terminated, and SCSIIM will attempt to re-initialize the wholeplatform again.

• SCSIIM has protection against infinite “roll-overs”, and will generally do only 2 re-initializations when a mandatory process fails to initialize. After the second attempt,scsiim will bring all remaining Supercell processes down, and will wait for the operatorto fix the situation. After fixing the situation, the operator must re-initialize scsiimmanually.

• The initialization sequence proceeds in a clockwise manner in the diagram on theprevious page. Several short-lived processes have been omitted for simplicity. The firstprocess is “popper”- this process will eventually start all the SS7 process set.Initialization messages for the SS7 processes will be output to the screen (more on thislater). The SS7 process set has its own “scsiim” equivalent which monitors and re-startsany process which should die.

• Most of the processes started by scsiim will output indications to files in /sc/processes/<processName>. Any core dump files (which are files produced by the OS when aprocess dies prematurely) are also kept in these directories.

• The MonitorAPI process monitors the consumption of platform resources such asfilesystem size, memory consumed by processes, and shared memory usage. Thisinformation is kept in /sc/processes/MonitorAPI.?, where ? is a number from 0-4.

• The scemlogd process keeps log files of the most recent message activity in the system.The files are /sc/processes/scemlogd.log.?, where ? is a number from 0-2.

• The mmgw process reads the /screl/active/data/aec/ENTITIES file in order to establish aTCP/IP connection with the OMC-R.

• The scevmgr process is next- it will allow other processes to send events to all interestedparties, including the OMC-R through the mmgw process. The sec_mgmt process willpost an alarm when a user exceeds a certain amount of login attempts.

• The vdt processes (vdt_1 and vdt_2) will establish TCP/IP virtual circuits to devicesneeding download. This will be done on an as-needed basis, usually through a requestfrom the sc_devom process (more on this later).



Supercell Process Set- Cont.

• mm_slm checks consistency of release directory,handles requests for release management.

• dgen generates data images for download ondemand.

• scdb parses CDF files, establishes MM’s DB, handlesDB-related requests.

• sc_devom manages devices under MM, includinginitialization and fault management.

• pm_collect collects statistics information fromdevices in network, transfers files to OMC-R.

• mmtadadm manages periodic tests, diagnostics.

• cpmo handles requests for call traces, call teardown,call profile records.

• callproc1 handles most call processing tasks.

• clcp handles paging, re-paging, class0 msgs.

SCSIIM

mm_slm dgen scdbsc_devom

pm_collect

mmtadadm

cpmo

callproc1clcpscfm_pokefepMon



NOTES:

Supercell Process Set- Continued

• The mm_slm process checks the /screl/release/*.lst and *.rds files for consistency withthe release directory structure. It is important to note that if there is an inconsistency inthe software release, mm_slm will cause the MM’s initialization sequence to be halted.

• The dgen process will generate download files for devices in the network in an on-demand basis. The files will be downloaded to the devices using the vdt processes. Thisdownload process is managed by sc_devom.

• The scdb process parses all the CDFs (Configuration Data Files), in order to create areal-time database in shared memory which contains data on all the elements in theSupercell network. The process then handles requests for recent change on the data(when network elements are moved, added, or modified). It is important to note that if aCDF file contains errors, the scdb process will fail to initialize and that will cause theMM’s initialization sequence to halt.

• The sc_devom process handles all the operations & management of the Supercell devicesunder control of the MM. All data downloads, initializations, code downloads, and faultrecovery actions are coordinated by sc_devom. This process “listens” to a particularTCP/IP connection to the FEPs over which all asynchronous state change eventnotifications are sent by the devices under the MM, and it takes appropriate action uponreceipt of those messages.

• The pm_collect process initiates data uploads from several Supercell devices forcollection of performance measurements. This data is then transferred to the OMC-Rthrough a unix ftp command. The mmtadadm process periodically spawns tests, audits,and diagnostics on the devices under the MM, and publishes alarms should a device faila test.

• The cpmo process establishes several data structures in shared memory for reference bythe other CP processes, and it is also responsible for forwarding requests for call traces,call teardowns, and call profile record upload to the callproc1 process.

• The callproc1 process handles call set up, handovers, teardowns, disconnects,registrations, and also generates call detail records and statistics information. The clcpprocess handles all connection-less oriented message traffic.

• The scfm_poke process is an optional process which will cause the CLI command“ENABLE CBSC UNC” to be performed at the end of the initialization. The fepMonprocess monitors the FEPs for connectivity to the MM.



MM Software Installation

• Installation document has the final word!

• First, boot platform from tape device.

• Next, run “sysinstall” program.

• Boot from disk, install required packages, optionalpackages.

• Install IPMs, if necessary.

• Run Motorola Installation script.- passwords for root, scadm, a7adm, lmf accounts.

- Timezone information

- Ethernet configuration (addresses, gateways, etc.)

- SC9600 release software

- SS7 installation, card setup, Point Code Setup

- Activate SC9600 software

• Re-boot platform, MM should initialize the Super-cell release software.

OMC-R

MM



NOTES:

MM Software Installation

An overview of the software installation process is presented here, along with“tricks & tips” for future reference. Please note that the Supercell SoftwareInstallation Document is the authority on software installations.

• A full software installation on the MM will typically take about 3 hours to perform (ifthe hard drives are not re-formatted). Please note that it should be necessary to do thisprocedure only when initially installing software (on a just-out-of-the-box platform). Ascratch installation should not be performed as a “catch all” for problems that cannot besolved. The problem should be isolated first- a scratch install is the LAST RESORT fordebugging a problem (rarely will a scratch software installation solve the problemanyway).

• The platform is first booted from the boot tape (which should be supplied with theTandem box). After booting from tape, a “#” prompt will be presented, at which the usermust enter “sysinstall”. Note: the sysinstall command is actually a script which may bemodified if necessary.

• After the sysinstall, the machine is re-booted (this time from the newly loaded diskdrives), and the required and optional packages are loaded onto the platform. If anyIPMs (updates to the OS) are needed, then these are loaded at this time.

• The Motorola installation scripts are then run. These scripts allow an operator to installthe platform with a minimum of manual steps required. However, the scripts aresomewhat sensitive to values which are incorrectly entered, and to interruption of thescript flow. The scripts are all ksh scripts which reside in /usr/sc.install. The main scriptis called MakeSC9600, and it in turn invokes functions which are contained in the fileInstallFunctions.ksh. As the scripts are run, a file called “FunctionsDone” is populatedwith the names of the functions which have been performed successfully. Should it everbe necessary to perform one of those functions again, the function’s name must bedeleted from the FunctionsDone file. Thus, the next time MakeSC9600 is invoked, it willstart at the first function called in the MakeSC9600 script which does not have an entryin the FunctionsDone file.

• The MakeSC9600 script will prompt the operator for the values of several parameters,and will install the SS7 software, Supercell software, and configure the ethernet cardsand SS7 cards, along with Point Codes for SS7 communication and account passwordsfor the a7adm, scadm, lmf, and root accounts. Timezone information will also berequested from the operator. Please note that the Software Installation Document(available in the CIG library) is the authority on installs.



MM Operations- Tools & Tricks

• Problems may occur with any of a number of sys-tems on MM.

• Most problems are due to incorrectly configuredsoftware parameters.

• When SS7 links won’t align:- Use a7 terminal handler’s “displ-slk;” command.

- Use SS7 network analyzer.

- Double-check route to MSC (check DSUs).

• When the Supercell processes won’t initialize:- Check scsiim.logfile.

- Check output and error files.

- Turn up DBG levels using dbapi, or inscsiim_cdf.mm1.

OMC-R

MM

!#?&@^!



NOTES:

MM Operations- Tools & Tricks

Problems may occur on the MM which require user intervention- usually theseproblems will occur the first time the MM is to be placed into service, or whennew software has been loaded on the platform. The source of these problems isalmost always incorrectly configured software. The following is informationwhich will help in debugging problems:

• Become familiar with the tools available for use on the MM. Also become familiar withthe location of log files for such sub-systems as the SS7 system.

• The SS7 links from the MM to the MSC must be aligned properly in order for the MMto initialize all the devices in the Supercell network. See also the chapter on SS7- it hasmuch information on point codes and debugging link problems. A common problem inlink alignment is incorrectly set point codes. Check point codes by using the a7 terminalhandler- login as “a7adm”, and run “start_term”. Then enter: “displ-opc;” and “displ-ospc;”. Link status may be checked by entering “displ-slk;”. As a last resort, all pointcode and link information may be removed by performing the following commands:

# cd /usr/a7# find . -name \*.254 -exec rm -f {} \;

Beware of the above, since you will have to re-provision the SS7 links once all the*.254 files are removed. The links are re-provisioned through entering commandsat the a7 terminal handler. Another common problem is that the DSUs that servicethe MM-MSC link are faulty or incorrectly configured- verify the connections, andre-seat and re-power the DSUs as a possible debugging step.

• When the Supercell processes won’t initialize correctly, the first step is to look at the /sc/scsiim/scsiim.logfile file, and determine which process is causing a problem. Thenproceed to look at that process’s .err and .out files in /sc/processes. This should point tothe problem. If mm_slm refuses to initialize, it is almost always because of a defective orincorrectly loaded software release. If scdb refuses to initialize, it is almost always due tobad CDF files. Try to get CDF files which have been produced from a clean MIB.

• The debug levels of processes may be modified in order to display more information. TheSupercell process “dbapi” is capable of doing this through the “mDbg” command, and itcan display information of debug levels through the “S <processName> DbgStatus”command. Debug levels may also be changed through modifying the scsiim_cdf.mm1file, specifically the “Args” field for the processes.


Conclusions:• The MM is a Unix-based platform which has been

customized to work in the Supercell environment.

• MM is a fully hardware redundant platform- avail-ability across single hardware failures insured.

• The MM’s main functions are call processing, opera-tions & maintenance- other functions include statisticscollection, test & audits, implementation of recentchange to the Supercell network.

• The Unix OS is a powerful Operating System whichallows multiple streams of execution to run “simulta-neously” on the cpu.

• The MM utilizes the Veritas software to manage diskdrives and file systems for redundancy.

• Supercell software is composed of several processes,each of which implement particular functions.

• SCSIIM starts up all the Supercell processes, and re-starts them should they die (according to spec.).

• MM Software installation is done through the use ofKsh functions- MakeSC9600 is the main script.

• Several tools and “tricks” may be used to analyze fail-ures on the MM- among them are dbapi and the a7term handler.


PWID Training: CDMAThe OMC-R and UNO Platforms


Objectives:

• To present the OMC-R platform utilized in SupercellCDMA in detail- including architecture (software andhardware), theory, and operations.

• To present an overview of the UNO platform (moreinformation on UNO is available through a TEM vid-eotape).


Table of Contents:

OMC-R Architecture and Interfaces ........................................................................... 115OMC-R Physical Configuration................................................................................... 117The OMC-R Supercell Software Architecture............................................................. 119Platform Software Initialization .................................................................................. 121Informix & the MIB ..................................................................................................... 123OMC-R Operations....................................................................................................... 125Supercell Device Provisioning Hierarchy ................................................................... 127The UNO Platform ....................................................................................................... 129

PWID Center for Excellence Revision 0, 4/24/97 The OMC-R and UNO Platforms


OMC-R Architecture and Interfaces

• The OMC-R is the same fault-tolerant UNIX box asthat used for the MM, with some exceptions.

• The OMC-R connects to the MM and to the UNOplatform via TCP/IP over Ethernet.

• The OMC-R has an RS-232 connection to each XCin the system, in order to perform O&M specific tothe XCs.

• A modem with an external line may be connected forremote access.

Async.Cntrls

DiskCntrls

EnetCntrls

EnetCntrls

TCP/IP

OMC-R




UtilitiesUNIX

To XCs, Modems, etc.

Informix



NOTES:

OMC-R Architecture and Interfaces

The OMC-R is very similar to the MM, in that the same base platform is uti-lized. However, there are some important differences and attributes of the OMC-R which the following points address:

• The OMC-R’s CPUs are not as fast and have less on-board (local) memory than theMM: The CPUs runs= at 50 Mhz, while there is 64Mb of local RAM. There is still 16Mbof global RAM on the OMC-R.

• Asynchronous cards are used on the OMC-R in order to communicate with printers,modems, and RS-232 lines (such as the XC OMP GPROC connections).

• The OMC-R requires more disk space because of its role in life. (Current configurationas of this writing has 3 logical drives of 2 gig each for a total of 6 gigabytes).

• The SS7 software suite does not run on the OMC-R (there is no need for this).

• Informix is used extensively on the OMC-R in order to provide a Managed InformationBase (MIB), which contains much information on the sub-tending CBSCs.

• The OMC-R connects to the MM and the UNO platforms by using TCP/IP overEthernet. The Ethernet connections are through coaxial cable, commonly called10Base2. Ethernet transceivers (also called MAUs) must be used by every deviceconnected to the ethernet 10Base2 network.

• In order to perform operations & maintenance which is specific to the Transcoder sub-system, the OMC-R is connected via serial lines to the XCs in the system. Theseconnections may have to be done via modems if the XCs are not co-located with theOMC-R (however, most sites have co-located XCs and OMC-R).

• The OMC-R may also have a modem with a direct outside line connected to it, in orderto allow remote access to the OMC-R (and remote dial-out as well).

• In order to provide the above mentioned connections, the OMC-R has two Async cards,each capable of eight serial connections (RS-232).



OMC-R Physical Configuration

• Very similar to MM, except for I/O cards and Stor-age Device Bay.

• Total disk space is greater, due to requirements forstorage of CDLs and other data.

• Async cards replace the SS7 cards on the MM.

• The OMC-R’s filesystems are markedly differentfrom the MM.


Air intake

Air Exhaust(bottom, front and rear)

Fan Modules(3)

Fan Modules(3)

Bulk PowerSupply Mod-ule

151 152 161 162 163 164 165 166 167 168

SPC SPC Ether-net

SCSI SCSI Ether-net

Async Ether-net

Async Ether-netor

SPCBypass

}

c0d0

c1d1

c1d6

c1d0c0d1

c0d2

2 gig Disk

2 gig Disk

2 gig Disk

2 gig Disk

(tape)c0d4

201

202

203

211

212

213

221

222

223

c0d3

c1d5

DC PWR (75

Mem

ory

A

Mem

ory

B

CP

U-A

CP

U-B

CP

U-C

I/O P

roce

ssor

0

I/O P

roce

ssor

1

}

I/O Controller Card Shelf

Processor Card Shelf

Storage Device Bay

0 1 2 30 10 1

2 gig disk

2 gig disk



NOTES:

OMC-R Physical Configuration

The OMC-R and the MM have much in common- the physical cabinet is thesame, but the I/O cards and the Storage Device Bay are different. The followingpoints are given with respect to the OMC-R physical configuration:

• With the exception of the Async cards (which replace the SS7 cards), the I/O cage layoutis identical to that of the MM. The Async cards are used for serial connections to devicesin the Supercell network, and to modems for outside access.

• The OMC-R uses more disk drives since the requirements for storage are greater- CallDetail Logs (CDLs) are stored at the OMC-R, as well as information from statisticscollection from devices in the Supercell network.

• The following is the result from a “volprint -v” command on an OMC-R running release2.4:

TYPE NAME ASSOC KSTATE LENGTH COMMENTvol boot gen ENABLED 40000vol cdl fsgen ENABLED 2011144vol home fsgen ENABLED 342724vol mibrdbs gen ENABLED 204800vol miniroot fsgen ENABLED 20000vol rdbspace gen ENABLED 163840vol root fsgen ENABLED 131072vol sc fsgen ENABLED 2230511vol screl fsgen ENABLED 1212416vol swap gen ENABLED 65536vol swap1 gen ENABLED 393216vol swap2 gen ENABLED 262144vol usr fsgen ENABLED 512000vol var fsgen ENABLED 140000

The cdl filesystem is used to hold Call Detail Logs. The mibrdbs data space (noticethat it is not a filesystem) is used by the Informix engine to hold data for theSupercell Managed Information Base (MIB). The rdbspace is also used byInformix, but this space holds other Informix data bases (not the MIB). The scfilesystem holds performance management (PM) data (including statistics), as wellas some CDL logs and other logs for Supercell processes. The swap volumes arefor adding more virtual memory to the OMC-R.

• The “volprint -s” command may be used by the curious (or otherwise interested) partiesto find out where the above mentioned volumes reside (on which disks).

PW

ID C


Revision 0, 4/24/97

The O

MC

-R and U

NO

Platform

s

11

8 o

f 25

9M

oto

rola

Co

nfid

en

tial P

rop

rieta

ry

The O

MC

-R S

upercell Softw

are Architecture

omcmib

DatabaseAccess

LogEvents

TCP/IP-w- OMC

API

UNIX Operating System

CDLLog

File Transfer(Bulk via FTP)

sdc

Alarms &

omc_slm

Applications- Alarm Mgr- PM Database.

ASCIIPM Data

Command

UNIX Shell

EventReports

Spooler

EventReports

CLI Session

scevmgr

State ChngEvent Rpts

scsiim

pm_rqst

omcgw

ted

RS-232 /XC MMI

XCAlarms

APISegments

Logs

DB

Loads

Data

PM

sec_mgmt

SecurityAlarms

UNIXTrigger



NOTES:

The OMC-R Supercell Software Architecture

The OMC-R’s Supercell software architecture is different from that of the MM.The following points are made on the architecture:

• The API again serves as the interface between the OS and the application software. It isthe same API as in the MM, and provides the same services.

• Event reports, alarms, CDLs, and PM statistics are stored on the OMC-R’s disk drives,and may be browsed (or viewed) by using CLI commands on the OMC-R, or by usingthe more user-friendly interfaces provided by the UNO platform.

• The OMC-R receives information from each of the MMs that are connected to itthrough the omcgw process. This process has TCP/IP connections to each of the MMsthrough the Ethernet network. As of release 2.4, an OMC-R and the MMs that areconnected to it must be on the same release- this is due to the format of the messages thatare sent between the omcgw and mmgw processes- if a release changes the format (oradds or deletes messages), then both platforms must be on the same release. However, asof release 2.6, a new feature will be added which will allow the OMC-R and the MM totalk to each other across different releases.

• The scevmgr process performs much the same job as it does on the MM, which is toroute event messages and alarms to their proper destinations. Destinations usuallyinclude any CLI sessions which have registered for events or alarms. The scevmgr on theOMC-R also keeps a log of all the events in the system, which may be browsed by usingthe “BROWSE” CLI command.

• The omcmib process interfaces with the Informix software and with the rest of theSupercell software in order to implement commands which add, modify, or deletedevices in the MIB.

• The pm_rqst process requests on a periodic basis for statistics information to bedownloaded from the MMs through an FTP protocol transfer.

• The ted process is the Transcoder Event Daemon- it uses the RS-232 links to the XCs inorder to present XC alarms to the Event Management system (scevmgr, sdc, UNO, etc.).



Platform Software Initialization

• Unix Init process is initialized during the Unix start-up sequence.

• Script /etc/rc4.d/fast/S97informix initializes Infor-mix.

• SCSIIM is started by the Init process, and starts-upall the Supercell processes.

• SCSIIM will then idle until CLI INIT command isperformed, or until a process dies.

• Uppon process death, the appropriate action will beperformed.

Unix Init

/etc/inittab

Unix Stuff

SCSIIM

/screl/active/data/scsiim_cdf.omc

/sc/scsiim/scsiim.logfile

Supercell Stuff

CLI INIT

Informix



NOTES:

Platform Software Initialization

The OMC-R goes through a similar initialization process as that of the MM. TheUnix side of affairs is identical, except for the additional volumes and filesystemsthat must be checked, and the fact that the OMC-R has async card, but no SS7cards. The following points concern the software initialization of the OMC-R:

• The Unix Init process will read from the /etc/inittab, running in run-state four. The run-state four will cause the /sbin/rc4 script to be run, which in turn will run the scripts in /etc/rc4.d/fast directory.

• Among the scripts in the /etc/rc4.d/fast directory is a script called S97informix, whichwill initialize the Informix database system (more on this later). The informix databasesystem must be up and running in order for the Supercell software to initialize correctly(the omcmib process will not come up if the Informix system is not up).

• The SCSIIM process is started by the Unix Init process, and the SCSIIM processproceeds to establish shared memory and environment variables as specified in the /screl/active/data/scsiim_cdf.omc file.

• The SCSIIM process then starts bringing up the processes specified in thescsiim_cdf.omc file. The output file for scsiim is /sc/scsiim/scsiim.logfile (just as in theMM), and the output files for the processes brought up by scsiim are in /sc/processes aswell.

• Once SCSIIM has brought up all the Supercell processes specified in the scsiim_cdf.omcfile, SCSIIM will pause and wait for a CLI INIT command to be performed, or for aprocess to die. Should a process die, SCSIIM will perform the action specified in thescsiim_cdf.omc file (which could be restart, re-init the platform, or do nothing at all).



Informix & the MIB

• Installation scripts initialize disk space for root dbspace. tbspaces command used for mibrdbs dbspace.

• MIB must be loaded at installation time, or emptyMIB must be created and populated via CLI com-mands.

• dbimport utility used to load MIB into OMC-R,dbexport utility used to export a MIB.

• tbmonitor utility used to get status of databases,user information, and other statistics on informixsystem.

• dbaccess utility used to access information in a dbthrough SQL commands.

• tbmode utility used to force checkpoint, or to changemode of informix online engine.

/usr/informix/etc/tbconfig

tbinitUnix Init

S97informix dbexport

dbaccess

dbimport

tbmonitor

Informix Utilities

tbmodesqlturbo

sqlturbo



NOTES:

Informix & the MIB

• Informix is brought up through the Unix Init program, when the /etc/rc4.d/fast/S97informix script is executed by the /sbin/rc4 script. The S97informix script will setupsome environment variables, and then execute a tbinit process.

• The tbinit process is in charge of starting up the Informix software subsystem, includingthe initialization of the shared memory used for informix. An “ipcs” command at theOMC-R should show at least one shared memory segment owned by user “informix”.

• The sqlturbo processes are brought up by the processes needing access to the informixdatabase information. There are two processes on the OMC-R which access informationin the MIB: omcmib, and mmi_ping.

• As part of the OMC-R installation, the installation scripts will initialized the disk spaceto be used for the root db space (root db space is always required), and the “tbspaces”informix utility will be used to create a new db space which is the mibrdbs space createdon disk by the installation scripts. Information on this space is kept in the root db space.

• At installation time, the installation scripts will ask for a mib to be imported. Thecommand used to import a mib is called “dbimport”. It expects a directory to bespecified in which a set of files exists. These files are the result of a previous “dbexport”command done a properly configured MIB (usually Systems Engineering is responsiblefor providing a good MIB). The files are ASCII files which contain the data for each ofthe tables in the MIB, in addition to a file which contains Structured Query Language(SQL) commands which create the database structure and tables.

• The tbmonitor Informix utility is a menu-driven program which allows an operator tostatus the informix system, including information on users, databases, and certainactions (such as forcing a checkpoint) may be performed as well. The tbmode utility isused to change the mode of operation of the informix engine.

• The dbaccess Informix utility is a menu-driven program which allows an operator toenter SQL commands in order to display or modify data in the database. If a MIB is tobe created from CLI commands, this utility must be used to first create an empty MIB.See http://scwww.cig.mot.com/SC/test/cdma_foa/FOA/OMCR/emptymib.creation andhttp://scwww.cig.mot.com/SC/test/cdma_foa/FOA/OMCR/device.states (on the WWW-use Netscape or Mosaic) for information on how to create an empty MIB, and how to setthe proper device states.



OMC-R Operations

• OMC-R is a management point for CDMA network.

• However, functions are migrating to UNO platform.

• CLI environment on OMC-R may be used to per-form any of the following:

- Status queries on system devices.

- Add, delete, or modify system devices or systemparameters.

- Display statistics information (PM reports).

- Browse through alarms and event reports.

- Select certain alarms or event reports to be dis-played in real-time.

- Load, enable, disable system devices.

- Handle system alarms through the Alarm Man-ager application.

OMC-R

OMC-R



NOTES:

OMC-R Operations

• Many operations may be performed on the OMC-R in order to manage the CDMAnetwork. It should be noted that much effort is being placed by Motorola to make theUNO platform be the management interface to the CDMA network. However, it willalways be possible to perform network management of CDMA through the OMC-Rsince some customers may not opt for buying the UNO platform.

• The CLI environment is used on the OMC-R in order to manage the CDMA network.One function which may be performed is that of getting status of the different devices inthe network under the OMC-R is shown below. The “display” command may be used todisplay the status and the parameters on practically any device in the system (see SystemCommands Reference manuals for more information).

cli> display omcr allstatus

• Operations which effect changes on the CDMA network may be performed at the OMC-R’s CLI session. The “add” and “delete” commands are used to provision or removedevices in the CDMA network. More on provisioning on the next slide.

• The PM reports may be viewed by using the dbaccess menu-driven program (in absenceof the UNO platform). The SQL command “select * from <tableName>” may be usedinside the dbaccess program, where <tableName> is replaced by a table describe in the“Performance Analysis” manual published by TED. However, it is highly recommendedthat a UNO platform be used for viewing statistics, or an alternative software packagesuch as “PM Sum” be used (check out http://www.rochellepark.pamd.cig.mot.com/software.html).

• The command “BROWSE CDLLOG” is used to browse through CDLs, while“BROWSE EVTLOG” is used to browse through event reports. Both commands acceptmany options, including start and end information for the period of time desired, anddesire attributes of the records to be displayed.

• At a CLI session, the command “ENABLE EVENTS” is used to enable events andalarms to be displayed at the CLI session. In addition the “SUBSCRIBE EVENTS”command may be used to set filters so that only the desired alarms and events aredisplayed. A process called the “Alarm Manager” may be started at an X-terminalconnected to the OMC-R in order to manage alarms through a GUI.

• The commands CUTOVER, LOAD, ENABLE, and DISABLE are used to managedevices in the CDMA network.



Supercell Device Provisioning Hierarchy

• Devices shown include physical (MSI, LPA, etc.) andlogical (SCH, CBSC) devices.

• Devices usually provisioned through CLI commandson off-line MIB. Small changes can be done on-line.

SCH Add/Del SCH

GLI

RFDS

RGLI

MDM

BTS

BTSLINK

C7LINK

OMC

MSI

MSC SPANBTS SPAN

LPA

LCI

CBSC

MM

BTSDS0

TERCKT

MSC

SRCHAN

MCC

Add/Del BTS

Add/Del MCC

Add/Del OMCR Add/Del CBSC

Add/Del LPA

Add/Del GLI

Add/Del BTSLINK

Add/Del MSI

Add/Del BTSSPAN

Add/Del MSCSPAN

Add/Del C7LINK

Add/Del TERCKT

MMI

Add/Del MMI

LEGEND:

Required Assoc. when Provisioned

Optional Assoc. when Provisioned

XASECT

Add/DelXASECT

Non-Parental Assoc. enforcedwhen deleting

MCCCE

CARRIER

TCHwc

PCH Add/Del PCH

ACH Add/Del ACH

BDC

Add/DelCARRIER

SECTOR CSM

Add/DelMDM

Add/Del CSM

BBX(R) Add/Del BBX

-C

-C

-C

XCSECT

Add/DelXCSECT NCONAdd/Del FEP

XCLINK

FEP(R)

XCLINK

Add/Del XCLINK

(CPP)

(OMP)



NOTES:

Supercell Device Provisioning Hierarchy

The device hierarchy is presented here because the OMC-R is the central reposi-tory for the information, and it also is responsible for carrying out the necessaryactions when the hierarchy is changed. The following regards the hierarchy:

• The hierarchy follows (with two exceptions during deletion) a parent-child approach.

• None of the devices are allowed to be provisioned in the database unless the properparent-child relationship exists. For example, it is not possible to add a BTSSPANwithout specifying a valid MSI through which the span information will flow.

• In general, the devices will be provisioned in an off-line manner, usually by a SystemsEngineer who can utilize an OMC-R (or an OMC-R simulator) which is not incommercial service in order to generate a MIB. However, many times changes to devicesin the CDMA network will be done through CLI commands at the OMC-R. The OMC-R is responsible for updating the physical device, verifying that the update wasperformed, and then updating the MIB.

• Procedures for adding/deleting devices from the CDMA network are found in theCellular System Administration manual, published by TED (manual number as of thiswriting: 68P09226A21-O).

• The following are the expansions to some abbreviations which may be confusing:NCON Nailed ConnectionXASECT External Analog SectorXCSECT External CDMA SectorBTSDS0 BTS Digital Signaling 0 (this is a timeslot on a T1/E1 spanline)SRCHAN Sub-Rate Channels (each BTSDS0 has four of these, each at 16 Kbps)MDM Modem Cage (this is where the GLI, MCC, BDC, and BBX cards reside)MCCCE Multiple Channel Card Channel Element (each MCC-8 has 8 of these)TCHwc Traffic Channel Walsh CodeACH Access ChannelPCH Paging ChannelSCH Signaling Channel

• Many of the devices have associated parameters which may be set as part of theprovisioning operations. Check with Systems Engineering personnel for the latestrecommended parameters document.

• There is a BSS Sysgen tool available which facilitates the generation of MIBs- contactPWID System Engineering Software Tools Group for the latest version of this tool.



The UNO Platform

• UNO- Universal Network Operation platform.- Connects to OMC-R through a TCP/IP over eth-

ernet connection.

- Is a Sun Sparc20 with cdrom drive, several harddrives for large storage capacity.

- Is the preferred platform for CDMA manage-ment.

• UNO Provides:- Alarm Manager, Status Display, Geographical

Display GUIs.

- Data Analysis (analysis of PM data)

- On-line documentation (using WorldView)

• An open interface is used- other devices can be man-aged.

OMC-R

OMC-R

UNO



NOTES:

The UNO Platform

Much emphasis is being placed on moving management of the CDMA networkto the UNO platform (indeed, much has already been done). The following pointsconcern release 2.4 of the UNO platform:

• UNO stands for Universal Network Operation platform. The idea behind UNO is tointegrate all devices in a cellular network under one platform which will be the OperatorInterface to the devices. What used to be the Applications Processor has turned into“UNO”.

• The UNO connects to the OMC-R through a TCP/IP connection over ethernet. However,a protocol called CMIP is used at both the OMC-R and the UNO platform fortransferring data and commands related to Events, Control, and database access.

• CMIP stands for Common Management Interface Protocol, and is an open interfacedeveloped by Sun Microsystems for the management of devices in a network. Thecommands are generic enough so that any device that needs to be managed may build a“CMIP agent” which is capable of translating general CMIP commands into theircorresponding native language commands (for example, the OMC-R’s native languagewould be CLI commands, while the SwitchMate’s native language would be CATterminal commands).

• The UNO platform currently provides all customer documentation through an Interleafviewer called WorldView, an Alarm Manager GUI, a Status Display GUI, aGeographical System Display, and data analysis tools (for analyzing PM data).

• More information on the UNO platform is available through a TEM (TechnicalEnrichment Matrix). TEM libraries are accessible at Arlington Heights, Ft. Lauderdaleand (I think) Ft. Worth facilities.


Conclusions:• The OMC-R is the same platform as the MM, with

some modifications due to its different role in life.

• The OMC-R software architecture is also similar tothe MM- however, the process set brought up bySCSIIM is very different.

• Informix plays a large role on the OMC-R. Informix isbrought up by Unix, and used by the Supercell pro-cesses in order to access data in the MIB.

• All network management functions for Supercell maybe performed at the OMC-R. These include changesto the network, as well as status queries, device statetransitions, and statistics display.

• When adding or deleting devices in Supercell, theSupercell device hierarchy governs the relationshipswhich must exist between devices.

• The UNO platform is the preferred platform for man-aging the Supercell network, since its graphical capa-bilities and user-friendly interfaces make it a moreattractive alternative than the OMC-R’s CLI inter-face.


PWID Training: CDMAThe Transcoder Platform


Objectives:

• To present the Transcoder platform utilized in Super-cell CDMA in detail- including architecture, theory,and operations.


Table of Contents:

The Transcoder Architecture ....................................................................................... 135Clock Signals in the XC ............................................................................................... 137TDM Bus and KSW Card............................................................................................. 139The GPROCs................................................................................................................. 141The Token Ring............................................................................................................. 143The MCAP and SBUS Buses ....................................................................................... 145MSI Connections .......................................................................................................... 147XC Troubleshooting- Preparation & Definitions ........................................................ 149XC Troubleshooting- First Steps.................................................................................. 151XC Troubleshooting- Device States ............................................................................. 153XC Sudden Trouble- Token Ring................................................................................. 155XC Troubleshooting- Defective Token Ring................................................................ 157More XC Troubleshooting Notes ................................................................................. 159Loading Software into the XC...................................................................................... 163

PWID Center for Excellence Revision 0, 4/24/97 The Transcoder Platform


The Transcoder Architecture

• BTSs, MSC data transported through TDM bus.

• All buses are redundant.

• GPROC-II will dispose of router (future).

• MCAP provides asynchronous communicationbetween GPROCs and peripheral boards.

• Token Ring LAN allows MM to communicate withGPROCs, GPROCs to communicate with eachother.

• Serial Bus allows status and control of half-sizecards and AMR card by GPROCs.

To MM via Routerwith GPROC

To half-size boardsand AMR card

KSWMSI XCDR

TDM Bus

MCAP Bus

2.048 Mbps(To DSU/MSC) orBTS)

2 span-lines per MSI

Token Ring LAN

GPROC

Ethernet LAN

GPROC II

To MM with GPROC-II

Serial Bus



NOTES:

The Transcoder Architecture

Knowledge of the XC’s architecture will help in troubleshooting problems withthe XC. The following points are made regarding the XC’s architecture:

• The “backbone” for all the PCM, QCELP, SCAP, and A+ data flowing through the XCis the TDM highway. There is a TDM bus for inbound data, and another for outbounddata (inbound/outbound from the perspective of the KSW). The TDM buses areredundant (so there are four in all). Traffic on the TDM bus is managed by the KSWcard. Remember the following definitions:

• PCM is digitally coded voice data (at 64kbps) between the MSC and the XC. PCMis usually transported through a T1 or E1 span line.

• QCELP is digitally coded voice data (at 16Kbps) between the XC and the BTS.QCELP occupies one/fourth of a 64Kbps channel on a T1 or E1 span line.

• SCAP is digital control data between the MM and the BTS. SCAP is transportedfrom the MM with TCP/IP over Ethernet, and then the FEP changes TCP/IP pack-ets into LAP-D. The LAP-D packets are then packed into a 64Kbps channel on a T1or E1 span line which goes to the GLI at the BTS site.

• A+ is digital control data between the MM and the MSC. When the MSC and theMM are not co-located, this data is usually transported through a nailed connectionusing a 64Kbps channel of a T1 or E1 span line.

• All buses in the XC are 2n redundant. Redundancy in the XC cards is highlyrecommended, and if implemented, will be 2n for the KSW, 2n for GPROCs, and n+1 forthe FEP.

• This diagram is simplified- the LANX cards are not shown, XC is single cage.

• In the future, the GPROC-II card will allow the router to be eliminated, since theGPROC-II card will have a TCP/IP over Ethernet interface.

• The MCAP bus allows the GPROCs to talk to other non-GPROC boards for purposes ofinitialization, collection of alarms, download of firmware, and connection/disconnectionmessages to the KSW board.

• The Token-Ring LAN allows GPROCs to talk to GPROCs in other shelves, in additionto connecting the MM with the XC.

• The Serial Bus allows GPROCs to talk to half-size cards and the AMR board forpurposes of initializing and obtaining status.



Clock Signals in the XC

• All cards sync’ed to clock in GCLK card in cage 0.

• Four-cage XC shown. CLKx cards with fiber-opticsare used to extend clock to other cages.

• gclk5 feeds clock to clkx6, which sends clock tokswx1 in cages 0, 1, 2, and 3.

• gclk3 feeds clock to clkx3, which distributes clock tokswx0 in cages 0, 1, 2, and 3.

• kswx1 cards feed clock to the MCAP bus A in eachcage, while kswx0 card feeds clocks to the MCAPbus B in each cage.

MCAP Bus A

MCAP Bus B

KSW, GPROCMSI, or XCDR

kswx 0kswx 1


kswx 0kswx 1

MCAP Bus A

MCAP Bus B


kswx 0kswx 1

MCAP Bus A

MCAP Bus B


kswx 0kswx 1

clkx 6clkx 3

GCLK5GCLK3

cage 0

cage 1

cage 2

cage 3

Clock signal A



NOTES:

Clock Signals in the XC

• In order to maintain synchronization inside the XC and between the XC and the MSC,the same reference clock signal must be distributed to all cards in the XC. Typically, theMSC serves as the master clock reference, and the XC database has defined two MSCspans as reference signals for the XC GCLK cards. These cards then phase lock thetiming signal to that of the spans, and hence the XC is synchronized with the MSC.

• Cards called CLKx (Clock Extenders) allow the timing signal to be extended to multipleframes. A single frame XC could have a GCLK provide the timing directly to the MCAPbus in that frame, but as soon as a KSWx board is placed in the XC, a CLKx will benecessary in order to distribute the clock to the frame. Fiber optics are used to inter-connect the CLKx boards.

• The timing sub-system in the XC is fully redundant.

• The GCLK card in slot 5 distributes the timing reference to the CLKx card in (half-sizebay) slot 6. The CLKx card then distributes via fiber optic cable the signal to the KSWxboards in slot 1 of every cage. The KSWx boards in slot 1 of each cage will then feed thesignal to the MCAP Bus A.

• The other side is identical, with the exception of the slot numbers. The GCLKs card inslot 3 distributes the timing reference to the CLKx card in (half-size bay) slot 3. TheCLKx card then distributes via fiber optic cable the signal to the KSWx boards in slot 0of every cage. The KSWx boards in slot 0 of each cage will then feed the signal to theMCAP Bus B.



TDM Bus and KSW Card

• All switching of data is performed in KSW card.

• 1024 timeslots @ 64kbps for total of 65Mbps capac-ity in each TDM bus.

• Multiple KSWs may be inter-connected to increasecapacity.

• KSW performs sub-rate switching at 16Kbps, pack-ing 4 channels of 16Kbps each onto a single TDMchannel.

• XC has two separate TDM highways, 0 and 1.

KSW

from KSW

to KSW

timeslot=0

timeslot=1023

timeslot=400

timeslot=150

timeslot=1023

timeslot=0

timeslot=150

timeslot=400

copy data “x” from “to KSW” timeslot 150 to “from KSW” timeslot 400

copy data “y” from “to KSW” timeslot 400 to “from KSW” timeslot 150

1024 64kbps timeslots

y

y

x

x

x

x

y

y

TDM bus

MCAP bus



NOTES:

TDM Bus and KSW Card

• The KSW card is responsible for managing data on the TDM highway. A TDM highwayis composed of two buses, each capable of 1024 TDM channels at 64Kbps rate. There isan inbound TDM bus (inbound to the KSW card), and an outbound TDM bus.

• The KSW card receives orders from the OMP GPROC over the MCAP bus when aconnection is to be made or broken.

• Using KSWx cards, a TDM highway may be extended to other cages. Up to a maximumof four KSW cards may be configured on one TDM highway.

• Each of the KSW cards will be able to select from all of the timeslots on the inbound“lane” of the TDM highway, but will be able to place traffic only on its outbound lane(1024 ports). Thus, a XC with two KSW cards will be able to switch between 2048 inputports and 1024 output ports, with three cards it will be 3072-1024, and with four cards,4096-1024.

• The KSW card contains hardware which is capable of sub-rate switching data on theTDM highway, which means that any two bits from any inbound TDM channel may beplaced in any (multiple of two bits) location on a channel of the outbound TDM lane. A64Kbps timeslot will send 8 bits at a time- this is divided by four to achieve the 16Kbpsrate, which results in 2 bits at a time for every 16Kbps sub-rate channel.

• The XC platform has two separate TDM highways, 0 and 1. There are two entirelydifferent hardware paths for these highways in order to provide true redundancy.



The GPROCs

• FEP GPROC routes messages between MM andOMP, CPP.

• OMP GPROC uses MCAP, TDM, SBUS buses, man-ages KSW cards, device states in XC.

• CPF GPROC uses MCAP, TDM buses, performscall processing functions.

• Shelf GPROC provides OMP access to other cages.

TBUS 1 0

MCAP Bus

LAPD

FEP2

CPF 1

EXECDLSP CP

MMI

terminal

RS232

TBUS 0 0

LAPD

FEP1

OMP

EXECDLSP CA

MMIterminalRS232

MCAPDLSP

MCAPDLSP

token ringtoken ring

LAN

LAN

MCAP Bus

router

router

xc pc

MM

SM

(Motorola protocol)

(TCP/IP)



NOTES:

The GPROCs

• The GPROCs perform a variety of jobs, depending on the software which is loaded onthe processor. All GPROCs are capable of interfacing with 32 64Kbps serial paths viathe TDM switch highway, a 16 Mbps token ring LAN, the Motorola Cellular AdvancedProcessor (MCAP) bus, and four additional RS-232 serial interfaces (one of which isavailable on the front panel). The GPROCs are powered by 24 MHz Motorola 68030processors with 16 Mb of RAM, 32 Kb of non-volatile RAM, and 1 Mb of flash rom.

• The simplest type of GPROC is the Shelf GPROC, which only runs the processes whichmanage the buses of the cage in which it sits. Shelf GPROCs are necessary in order toallow access by the OMP and CPF GPROCs to shelves which have no OMP or CPFGPROC. Processes running on the Shelf GPROC include the MMI (Man MachineInterface) which provides a command interpreter, the MCAP DLSP (Motorola CellularAdvanced Processor Bus Data Link Service Process), which handles messages being sentto and from the peripheral boards (XCDR, MSI, GCLK, etc.), and the TTY DLSP whichmanages the serial bus for communications to the half-size cards and the AMR card inthat cage.

• The OMP GPROC has all the Shelf GPROC processes, plus a process called “CentralAuthority” (CA) and another called “Switch Manager” (SM). CA keeps track of XCdevice states and does whatever necessary to change their state (sending/receivingconfiguration messages, orders, and code load over the MCAP bus). SM managestimeslot connections in the KSW (ordering the KSW through the MCAP bus to make orbreak connections on the TDM highway).

• The CPF GPROC has all the Shelf GPROC processes, plus a process called “CallProcessing” (CP). The CP process terminates call processing SCAP messages sent by theMM through the FEP GPROC (which then places the SCAP message into a LAP-Dformat on the TDM bus for consumption by the CPF GPROC). The CP process makesand breaks call connections by sending messages to the SM process on the OMPGPROC via the token-ring bus.

• The FEP GPROC runs the LAN software, the LAPD software, and the router software.The LAN process on the FEP interfaces with the token ring in order to communicatewith other FEPs and with the MM. The LAPD software talks through the TDM bus inorder to exchange LAPD messages with the BTS devices (through the CSLDL links) andwith the OMP and CPF GPROCs (through the MXLDL links). The router processswitches messages going through the FEP to their proper destination.



The Token Ring

• LANx provides token ring access for all GPROCs ina cage. Token ring is 16 Mbps IEEE 802.5 standard.

• Token ring provides comm. link between non-FEPGPROCs, between FEPs, between MM and FEPs.

• FEPs use different language (TCP/IP) than otherGPROCs (Motorola Proprietary), but both useIEEE 802.5 transport.

xc pc

cage 2

cage 0

LANX

FEP 1OMP

LANX

FEP 3CPF 2

router

MM

ethernet

fiber optic token ring

throughbackplane

(over coax)

slot 25

slot 20

cage 3

LANX

FEP 4shelf gproc

slot 20LANX

FEP 2CPF 1

slot 21cage 1

slot 20

slot 25



NOTES:

The Token Ring

• A fiber optic token ring network is used in order to achieve high-speed communicationsbetween the MM and the FEPs, between FEPs, and between all non-FEP GPROCs inthe XC.

• The LANX card interfaces all GPROCs in a XC cage with the token ring.

• The token ring is a 16Mbps bi-directional communications path, governed by the IEEE802.5 standard.

• All GPROCs (including the FEP GPROC, before the FEP software is loaded on it by theMM) speak the same proprietary Motorola protocol over the token ring. The XC-PCalso speaks this language, and uses the language in order to download software to theXC GPROC cards.

• Once the FEP software is running on the FEP GPROCs, the FEPs start talking TCP/IPto the MM over the same token ring network. However, the TCP/IP packets are being“picked up” by the Router and sent over the TCP/IP over Ethernet network to the MM.All other IEEE 802.5 packets (which include those used by all other GPROCs in the XC)are ignored by the MM’s router.



The MCAP and SBUS buses

• GPROCs talk to MSIs, XCDRs, and KSW boardsthrough MCAP bus.

• Information sent over MCAP bus:- Initialization orders for the boards.

- Collection of alarms from boards.

- Download of firmware to boards.

- Connection/Disconnection messages to KSWcard.

• SBUS used to initialize and status half-size cards(LANX, CLKX, KSWX) and AMR card.

• MCAP and SBUS buses are local to their cages.Token ring used to communicate between cages.

LANX

LANX

OMPGPROC

CPFGPROC

cage 0cage 1

MSIXCDRGCLK

MSI

XCDR

KSW

KSW

fiberoptictokenring

MCAP

MCAP

MCAP: KSW initialization, download, and connect/disconnect messages:

MCAP: MSI, XCDR, GCLK initialization and download messages only:

KSWX

CLKX

AMR

KSWXSBUS

SBUS

SBUS: Initialization and status messages for half-size and AMR boards:



NOTES:

The MCAP and SBUS Buses

• The MCAP bus provides a path for the GPROC (the OMP GPROC) to communicatewith the peripheral boards (full-size boards other than the GPROCs).

• The MCAP bus is used to communicate with the KSW, MSI, and XCDR boards.

• The OMP GPROC will need to communicate with the peripheral boards for any of thefollowing functions:

• Initialization of the boards (during an XC initialization, or during an attempt torecover from a fault).

• Collection of alarms from the boards. Alarms are stored in the 32 Kb of non-vola-tile RAM on the OMP GPROC.

• Download of software to the boards. This is done during an XC initialization, orduring an attempt to recover from a fault.

• Connection/Disconnection messages to the KSW board(s) for the TDM highways.

• The serial bus (SBUS) is used to initialize and status half-size boards and the AMR(Alarm Monitoring and Reporting) card. An example of a serial bus message would be aGPROC poll message to an AMR asking, “What is the status of the fans and powerconverters?”, or a GPROC diagnostic message to a LANX asking it for some kind ofresponse to show that it is alive.

• The MCAP bus is local to a cage. If an OMP GPROC in cage 0 wants to send a messageto a KSW card in cage 1, the OMP GPROC will send a message over the token ring tothe MCAP bus administrator GPROC (usually the CPF GPROC, or maybe a shelfGPROC). This GPROC in cage 1 will relay the request over the local MCAP bus to theKSW card.

• The SBUS is also local to a cage. If an OMP GPROC wants to status a LANX card inanother cage, the same scenario above is repeated (except the SBUS is used in the remotecage, instead of the MCAP bus).



MSI Connections

• MSI Cards support either 2 T1 or 2 E1 span lines.

• Three MSI cards interface through XC backplane toa BIB card.

• BIB provides isolation of XC from T1 or E1 line.

• MSI board connects via backplane to TDM high-ways.

• MSI board under control of OMP GPROC throughMCAP bus.

• Cable pin-out for span cable to IDF on next page.

MS0MS1

MS2MS3

UPPERMS0

MSI

slot 13

MSI

slot 15

MSI

TOP of XC frame

throughbackplane

span cable to IDF

BIB card BIB 0

BIB 27,9,11

BIB 112,14,

BIB 36,8,10

BIB 013,15,

BIB 27,9,11

BIB 112,14,

BIB 36,8,10

Lower XC Shelf(0) Upper XC Shelf(1)

XC Frame Top View- BIB Interfaces

13,15,17 16 17 16

slot 17

MS0 MS1

MS2 MS3

MS0 MS1

MS2 MS3



NOTES:

MSI Connections

• Each MSI card is capable of supporting either 2 T1 or 2 E1 span lines. The MSI providesnormal clock recovery, span line build out, elastic buffering and jitter attenuation,insertion and alarms and loopbacks for remote diagnostic capability. The MSI uses aMotorola MC68302 Integrated Multiprotocol Processor (IMP) to perform dataprocessing and control functions.

• The T1 or E1 span lines are connected via the XC backplane to a connector whichconnects a ribbon cable to the top of the XC frame. At the top of the XC, a BIB(Balanced-Line Interconnect Board) provides isolation and impedance matching of theT1 or E1 to the XC signal path.

• A BIB board may be used for North American T1, Japanese T1, and European E1 (120ohm - 3V), while a T43IB board must be used to interface with European E1 at 75 ohm& 2.37 V.

• Each BIB supports up to 6 span lines (thus each BIB supports 3 MSI cards). The MSIslot numbers which each BIB supports are given in the diagram. The following tablegives the PIN wiring for the 37-pin D-connector at each of the BIBs:

• The MSI board is under control and status of the OMP GPROC through the MCAP bus.

Pins Signal Span Color

1,20 Tx t/r 1 white/blue,blue/white

2,21 Rx t/r 1 white/orange,orange/white

4,23 Tx t/r 2 white/green, green/white

5,24 Rx t/r 2 white/brown,brown/white

7,26 Tx t/r 3 white/gray,gray/white

8,27 Rx t/r 3 red/blue,blue/red

10,29 Tx t/r 4 red/orange, orange/red

11,30 Rx t/r 4 red/green,green/red

13,32 Tx t/r 5 red/gray, gray/red

14,33 Rx t/r 5 black/blue, blue/black

16,35 Tx t/r 6 black/blue,blue/black

17,36 Rx t/r 6 black/orange, orange/black



XC Troubleshooting- Prep & Definitions

• Preparing for trouble:- Empty non-volatile RAM.

- Verify that date is correct.

• Types of trouble:- Sudden.

- Redundancy Activated.

- Activity-Induced.

- New CBSC.

- Occasional Voice Loss.

Transcoder Frame



NOTES:

XC Troubleshooting- Preparation & Definitions

Troubleshooting of the XC is like troubleshooting any other complex system-knowledge of the system’s architecture is required, along with knowledge of thetools and methods available for analysis of problems, along with some imagina-tion on the part of the trouble-shooter.

• All XCs should be prepared for trouble before the trouble is ever encountered. All thenon-FEP GPROCs should have their non-volatile RAM cleared by issuing the “swfmclear fatal” and “swfm clear nonfatal”, mmi commands. Fatal SWFM messages areplaced in the non-volatile RAM until this space fills up- thus it is important that thememory be cleared. It is difficult to correlate the events happening in the XC to outside“real-world” events if the XC’s time is not correct- use the “chg_time” command toplace the proper time on the XC.

• There are several different types of trouble the XC may run into (indeed, almost anytype of trouble for any network element will fall into one of these categories):

TypeAffectsservice:

Cause

Sudden (During normaloperation, there is a sud-den loss of voice or links)

Yes This is usually due to hardware failure andlack of software redundancy in currentreleases. Immediate action may be required.

Redundancy activated(FEP, KSW, Clock redun-dancy)

No This is usually due to hardware failure.Replacement of a board may be required.

Activity-induced (DuringBTS expansion, BSC SWupgrade, or BTS reparent-ing)

Yes Most of the time, the trouble encountered isdue to the new software, new hardware intro-duced (expansion boards), or new databases.

New BSC Yes There are many possible factors. Hardware,databases are all new. Maybe the most diffi-cult to analyze, but there is time available totroubleshoot.

Occasional no voice Some If the number of no voice calls is very small,with no particular dependence on TERCKTand TCH used, then it may be a software bug.

PW

ID C


Revision 0, 4/24/97

The T

ranscoder Platform

15

0 o

f 25

9M

oto

rola

Co

nfid

en

tial P

rop

rieta

ry

XC

Troubleshooting- First S

teps

•D

etermination of G

PR

OC

state is the first step!

•G

PR

OC

s should all be in GP

RO

C R

AM

state.

•G

PR

OC

in slot 20 will be the O

MP

GP

RO

C.

•F

EP

s should all be in FE

P IN

S state.

•C

heck status of equipment through m

mi com

mand

“disp_equipment”.

BTC 0

BTC 1

XCDR 1

KSW A

GCLK B

GCLK A

KSW B

XCDR 0

RMT KSWX A4

RMT KSWX A3

RMT KSWX A2

RMT KSWX A1

RMT KSWX A0

KSWX A2

KSWX A1

KSWX A0

EXP

EXP

EXP

LANX A

LANX B

KSWX B0

KSWX B1

EXP

EXP

KSWX B2EXP

RMTKSWX B0

RMTKSWX B1

RMTKSWX B2

GPROC 1

GPROC 0

MSI 0

MSI 1

MSI 2

MSI 3

XCDR 2

XCDR 3

MSI 4

XCDR 4

XCDR 5

XCDR 6

XCDR 7

XCDR 8

XCDR 9

XCDR 10

XCDR 11

CLKX B0

CLKX B1

CLKX A0

CLKX A1

CLKX A2

CLKX B2

RMTKSWX B3

RMTKSWX B4

LOCAL KSWX A

LOCAL KSWX B

U0

U1

U2

U3

U4

U5

U6

U7

U8

U9

U10

U11

U12

U13

U14

U15

U16

U17

U18

U19

U20

U21

U22

U23

U24

U25

U26

U27

U28

L0

L1

L3

L5

L6

L7

L8

L9

L10

L11

L12

L13

L14

L15

L16

L17

L18

L19

L20

L21

L22

L23

L24

L25

L26

L27

L28

XCDR 12

XCDR 13

XCDR 14

XCDR 15

XCDR 16

XCDR 17

XCDR 18

MSI 5

MSI 6

MSI 7

MSI 8

MSI 9

MSI 10

MSI 11

GPROC 2

GPROC 3

GPROC 4

GPROC 5

GPROC 6

GPROC 7



NOTES:

XC Troubleshooting- First Steps

• The first step in debugging ANY Transcoder problem is to determine what state theGPROCs are in. Use the following table to do this:

• The prompt is presented when an RS-232 terminal at 9600 baud, 8 bits, no parity isconnected to the DB-25 pin connector on the front of the GPROC.

• There are two access levels possible using the RS-232 terminal. One is called the“Customer Prompt” and will be “Cust BSS MMI-0xxx ->”. The other is called the“emon” prompt and will be “RxM:emon_0xxx%”. When inside the customer prompt, a“chg_level” command will allow access to the emon prompt by pressing cntrl-n. InCDMA the password required is “5cardstud”.

• The OMP, CPFs, and shelf GPROC cards should be in GPROC RAM state, while theFEPs should be in the FEP INS state. If any of the GPROCs are not in their proper state,then start looking for token ring failure problems (more on this later).

• If the GPROCs are all in the RAM, then the following command may be used to displaythe status of the devices. All devices should be either unlocked busy, unlocked enabled,or equipped busy:

Cust BSS MMI-0115-> disp_equipment bsc status location

If LED is... and Prompt is... then State is...

Green steadyon or blinking

“Cust BSS MMI-0xxx” ->, control-n gives“ROM:emon_0xxx %”

GPROC ROM

Green steady “Cust BSS MMI-0xxx ->”, chg_level, con-trol-n gives “RAM:emon_0xxx %”

GPROC RAM

Green slowblinking

“RT>” only FEP, OOS ROM

Green fastblinking

“RT>”, esc/return gives “F>” FEP, OOS RAM

Green steady “RT>”, esc/return gives “F>” FEP, INSRed only None DeadRed and greenblinking fast

Garbage, “RAM:emon_0xxx %” or“ROM:emon_0xxx %”

EEPROM programmingin progress



XC Troubleshooting- Device States

• Above diagram illustrates valid states most XCdevices may be in.

• Valid states are:- Not Equipped

- Locked Enabled/Disabled

- Unlocked Busy/Enabled/Disabled

• See results of “disp_equipment bsc status location”mmi command.

LOCK

EQUIP

UNEQUIP

LOC

K

UN

LOC

K

UNEQUIP

UNLOCKUNLOCKFAILURE

LOC

K

UN

LOC

K FA

ILUR

E

DISABLE

ENABLE

DISABLE

ACTIVE

STANDBY

ENABLEFAILUR E

Operational Transition: Administrative Transition:

LockedDisabled

LockedEnabled

NotEquipped

UnlockedBusy

UnlockedEnabled

UnlockedDisabled

ENABLE

DISABLE



NOTES:

XC Troubleshooting- Device States

• Three commands are used at the XC OMP MMI in order to manage devices: EQUIP,LOCK, and UNLOCK.

• The following are descriptions of the different XC device states:

• Not Equipped- device does not have an entry in the XC database.• Locked- device is equipped, but is not allowed to go into service. Two sub-states

exist: disabled or enabled- the XC’s CA process automatically transitions betweenthese states, depending on whether or not the device can be placed into service.

• Unlocked Busy- device is equipped, and is being used (active).• Unlocked Enabled- device is equipped, and is ready to go into service if necessary

(but is not currently being used).• Unlocked Disabled- device is equipped, but an attempt to place it into service has

failed. CA will continue to attempt to place device in service.

• The following command may be used at the OMP GPROC in order to display the stateof all devices in the XC:

- disp_equip 0 location status NAME ID 1 2 3 | DEVICE’S PARENT(S)/ ADDITIONAL INFO | C/S | Admin OP============================================================================= GPROC 0 0 0 | .................................. | 0/20 | Unlocked Busy GPROC 1 0 0 | .................................. | 0/21 | Unlocked Busy GPROC 2 0 0 | .................................. | 0/25 | Unlocked Busy GPROC 3 0 0 | .................................. | 0/23 | Unlocked Busy OMP 0 0 0 | GPROC 0 0 0 ..................... | 0/20 | Unlocked Busy MSI 0 0 0 | .................................. | 0/6 | Unlocked Busy MMS 0 0 0 | MSI 0 0 0 ..................... | 0/6 | Unlocked Busy MMS 0 1 0 | MSI 0 0 0 ..................... | 0/6 | Unlocked Busy GCLK 0 0 0 | .................................. | 0/5 | Unlocked Busy KSW 0 0 0 | .................................. | 0/27 | Unlocked Busy CAGE 0 0 0 | TDM HWY 0 ......................... | ----- | Unlocked Busy CAB 0 0 0 | .................................. | ----- | Unlocked Enabled SITE 0 0 0 | .................................. | ----- | Unlocked Busy XCDR 0 0 0 | .................................. | 0/26 | Unlocked Disabled CSLDL 0 0 0 | MMS 0 0 0[ 1]-FEP 0 0 0[ 3] | ----- | Unlocked Busy CSLDL 1 0 0 | MMS 0 0 0[ 2]-FEP 0 0 0[ 4] | ----- | Unlocked Busy CSLDL 2 0 0 | MMS 0 0 0[ 3]-FEP 0 0 0[ 5] | ----- | Unlocked Busy MXLDL 0 0 0 | GPROC 0 0 0[ 2]-FEP 0 0 0[ 1] | ----- | Unlocked Busy MXLDL 1 0 0 | GPROC 1 0 0[ 2]-FEP 0 0 0[ 2] | ----- | Unlocked Busy FEP 0 0 0 | GPROC 2 0 0 ..................... | 0/25 | Unlocked Busy CPF 1 0 0 | GPROC 1 0 0 ..................... | 0/21 | Equipped Busy CPF 2 0 0 | GPROC 3 0 0 ..................... | 0/23 | Equipped Busy



XC Sudden Trouble- Token Ring

• First sign of trouble- MXLDL or XCLINK alarm.

• Token Ring Failure: FEPs, OMP are OOS RAM,other GPROCs are in ROM.

• MM normally polls FEPs. If FEPs don’t see the poll,they drop to OOS.

OMP

cage 1

cage 0

FEP 1FEP 3

FEP 2FEP 4

router

MM

ethernet

fiber optic token ring

throughbackplane

(over coax)

slot 25

slot 25

cage 2

FEP 5FEP 7

slot 25

cage 3

FEP 6FEP 8

slot 25

POLL

BAD TOKEN RING, POLL DOES NOTREACH DESTINATION!!!

FEPs wait for poll, but when it does not come,they drop to OOS




LANX

LANX

LANX

LANX

token ring died, sousually only one LEDon token ring card is lit.



NOTES:

XC Sudden Trouble- Token Ring

• The first sign of trouble with the XC is usually an MXLDL or XCLINK alarm observedat the OMC-R. This alarm is an indication that something is wrong, and the first step (asalways) should be to determine what state the GPROCs (including FEPs) are in.

• The following is a table detailing the different “sudden trouble” types, and their usualsymptoms:

• The MM “pings” the FEPs periodically over the TCP/IP network. If the FEPs do not seethis message from the MM over a period of time, the FEPs will go OOS RAM.

• Please note that the “ping” from the MM may not be getting to the FEPs due to severalreasons: defective router, defective LANX card, or a defective LAN fiber optic cable. Thelast reason is less likely in a non-lab environment.

Trouble type Symptoms

Token ring failure(The important thing toremember about the tokenring failure is that it willaffect all GPROCs: FEPs,OMP, CPF, and shelfGPROCs.)

FEPs will drop to OOS RAM or go red LED. Since thetoken ring is down, there is no messaging between theMM and FEPs, so the FEPs die due to lack of polls fromthe MM. (WARNING: This could also be an indicationof an MM or router problem)CPF or shelf GPROCs will all reboot, and will be foundin ROM. Since the token ring is down, there is no mes-saging between the OMP and other GPROCs, so thenon-OMP GPROCs (CPF and shelf) will die from lackof OMP polls.OMP is in RAM as it should be, but if you check theemon prompt, you will see errors, indicating that thetoken ring is in beacon mode, and that links have beendropped.

OMP GPROC failure OMP, CPF and shelf GPROCs have all rebooted andappear to be rebooting/initializing.FEPs still INS

CPF or shelf GPROC fail-ure

1 CPF or 1 shelf GPROC seems to have rebooted andautomatically recovered.FEPs still INS



XC Troubleshooting- Defective Token Ring

• Use problem isolation techniques- reset all GPROCs and KSWs between iterations

- observe for “link established” message at OMP

• Once problem isolated:- replace defective device

- reset all GPROCs and KSWs one last time

- observe for “XC init complete” message at OMP

3/20

RX

TX

1/20

RX

TX

0/202/20Router

TX

RX

0-R

1-R

0-2E

1-3E

RX

TX

RX

TX

All boards pictured are LANXs.Location x/y = cage x, slot y

Normal Fiber Optic Cabling- 4 cages

3/20

RX

TX

1/20

RX

TX

0/202/20Router

TX

RX

0-2E

1-3E

RX

TX

RX

TX

#1: Is it the Router?

3/20

RX

TX

1/20

RX

TX

0/202/20Router

TX

RX1-3E

RX

TX

RX

TX

#2: Is Cage 0 Ok?

3/20

RX

TX

1/20

RX

TX

0/202/20Router

TX

RX

0-R

1-R

RX

TX

RX

TX

#3: Are router, cage 0 Ok?



NOTES:

XC Troubleshooting- Defective Token Ring

• When a defective token ring is suspected, the GPROCs and KSWs should be reset (bytoggling the switches on their front panels). A defective token ring will result in the lackof a “link established” message at the OMP GPROC, and all GPROCs will reboot with afatal SWFM message with a “no good lan available” message.

• Use problem isolation techniques by utilizing a spare fiber optic cable to “short” outcertain sections of the fiber optic network, as seen in the diagram. Between eachiteration, reset all the GPROCs and KSW cards, and check to see if the “linkestablished” message appears.

• If the “link established” appears for case #1, then the router must be checked, andpossibly replaced. However, replacing the router requires some programming as well- soreplacement should be the last alternative (more on this later). If the message does notappear, then continue with case #2 (since it could be a bad router OR a bad componentin the XC).

• If the “link established” message appears for case #2, then you can assume that thetoken ring path for cage 0 is okay. If not, then consider replacing the LANX card forcage 0.

• If the “link established” message appears for case #3, then you can assume that therouter is okay. If not, then you can assume that the router is defective, and the routermust be checked.

• Continue in the above manner until the culprit has been found.

• Note: removing and inserting fiber optic cables is a delicate operation, which when doneproperly will pose no mechanical resistance. If you find that the plug does not slide intothe socket easily, then make sure that the bumps on the side of the bayonet plug aresliding into the grooves on the socket connector.

• Note: Replacing a LANX will require a torx screwdriver. There is a rotary switch on theLANX card indicating the cage ID of the parent cage- make sure this switch is setproperly if any cards are replaced.

PW

ID C


Revision 0, 4/24/97

The T

ranscoder Platform

15

8 o

f 25

9M

oto

rola

Co

nfid

en

tial P

rop

rieta

ry

More X

C Troubleshooting N

otes

•O

MP

GP

RO

C failure- F

EP

s stay INS

.

•C

PF, S

helf GP

RO

C failure- G

PR

OC

reboots.

•F

EP

Redundancy- analyze data, replace F

EP.

•G

CLK

Redundancy- replace bad board.

BTC 0

BTC 1

XCDR 1

KSW A

GCLK B

GCLK A

KSW B

XCDR 0

RMT KSWX A4

RMT KSWX A3

RMT KSWX A2

RMT KSWX A1

RMT KSWX A0

KSWX A2

KSWX A1

KSWX A0

EXP

EXP

EXP

LANX A

LANX B

KSWX B0

KSWX B1

EXP

EXP

KSWX B2EXP

RMTKSWX B0

RMTKSWX B1

RMTKSWX B2

GPROC 1

GPROC 0

MSI 0

MSI 1

MSI 2

MSI 3

XCDR 2

XCDR 3

MSI 4

XCDR 4

XCDR 5

XCDR 6

XCDR 7

XCDR 8

XCDR 9

XCDR 10

XCDR 11

CLKX B0

CLKX B1

CLKX A0

CLKX A1

CLKX A2

CLKX B2

RMTKSWX B3

RMTKSWX B4

LOCAL KSWX A

LOCAL KSWX B

U0

U1

U2

U3

U4

U5

U6

U7

U8

U9

U10

U11

U12

U13

U14

U15

U16

U17

U18

U19

U20

U21

U22

U23

U24

U25

U26

U27

U28

L0

L1

L3

L5

L6

L7

L8

L9

L10

L11

L12

L13

L14

L15

L16

L17

L18

L19

L20

L21

L22

L23

L24

L25

L26

L27

L28

XCDR 12

XCDR 13

XCDR 14

XCDR 15

XCDR 16

XCDR 17

XCDR 18

MSI 5

MSI 6

MSI 7

MSI 8

MSI 9

MSI 10

MSI 11

GPROC 2

GPROC 3

GPROC 4

GPROC 5

GPROC 6

GPROC 7



NOTES:

More XC Troubleshooting Notes

• If an OMP GPROC is in GPROC ROM state, then reset the entire XC to recover. If theOMP still does not initialize (no “XC init complete” message observed), then read theOMP error messages at a terminal which has logging capability by entering thefollowing commands repeatedly, until no messages are left:

RAM:emon_0115% swfm read nonfatalRAM:emon_0115% swfm read fatal

In order to replace an OMP GPROC, switch the reset switch to “disabled”, replace theboard, and reset all GPROCs and KSW cards. Wait for the “XC InitializationComplete” message, and then do an “ENABLE CBSC” command at the MM.

• If a CPF or Shelf GPROC is rebooting of its own accord, monitor the GPROC via aterminal from its serial port. At the OMP, place another serial terminal to its serial port,with logging enabled in order to log errors that may appear. At the faulty GPROC, readthe swfm error indications (by entering the above two commands repeatedly). Theoutput should be analyzed by XC support personnel- call MCSC in order to get an XCsupport person on-line. If a CPF or Shelf GPROC needs replacement, first lock thedevice by entering the following at the OMP GPROC:

Cust BSS MMI-0115-> lock_device bsc GPROC <id> 0 0

Then disable the GPROC, replace it with a new one, and unlock the GPROC with thefollowing command at the OMP GPROC:

Cust BSS MMI-0115-> unlock_device bsc GPROC <id> 0 0

Monitor the new GPROC, and make sure it initializes correctly.

• If a FEP fails, the redundant FEP will be enabled automatically, and it will take over thelinks that were on the failed FEP. In order to collect information from the failed FEP foranalysis, connect a serial port to a computer which can log output to a file, and performthe following command at the FEPSCOPE prompt:

F> execlist system

The document for all fepscope commands is in the Arlington Heights network in /usr/test/supcell/ife/frame/fepscope.

• The following commands may be used to check the status of the GCLK boards:Cust BSS MMI-0115-> disp_dev_stat bsc GCLK 0 0 0Cust BSS MMI-0115-> disp_dev_stat bsc GCLK 1 0 0

The “master” GCLK will be unlocked-busy, and have its green LED lit. Look at theevent log in the OMP GPROC (use swfm read command) to check for bad GCLK.



Loading Software into the XC

• Use XC-PC, placing PC on the Token-Ring LAN.

• Release and database files on PC (or database to becreated by MMI commands).

• Reset XC with LANLOAD program running on XC-PC- XC will be downloaded.

• “Sysgen” will be needed to build database fromscratch.

- Script used on PC connected to OMP GPROCserial port.

• Take PC off Token-Ring LAN, reset XC, enableCBSC.

Transcoder Frame



NOTES:

Loading Software into the XC

• The XC software load is done through the token ring LAN, using the XC-PC. Generally,the XC-PC is kept off the LAN with the XC in order to avoid placing a single-point-of-failure into the network. So, the fiber optic LAN will need to be modified such that thePC is placed into the LAN.

• The XC-PC will have a software program call “lanload” loaded on its hard disk. Inaddition, the XC-PC needs to have all the files which compose the XC software releaseloaded into a directory on the PC’s hard drive. This will usually be a sub-directory ofc:\xc.

• In addition, if an XC database is to be loaded (as opposed to building one right on theXC through OMP MMI commands), the database object (a file with a “.002” extension)must be copied into the XC software release directory. If this is a software upgrade, youwant to upload the current XC database by performing the following:

• Place PC on the LAN (PC’s Rx to LANx Tx, PC’s Tx to LANx Rx).• Start up lanload program in the XC release directory: lanload -f• Reset the XC (reset OMP GPROC< followed by all other GPROCs and KSW cards)• Observe that “Link Established” messages appear at the XC-PC (if not, XC will

reboot).• Wait for XC Initialization Complete message.• Type “u” at the lanload prompt on the PC.• Respond to prompts to upload object ‘2’ from GPROC 0115.

• Download the code objects to the XC by re-setting the XC while the XC-PC is in thelanload program. Code objects will be downloaded to the OMP GPROC, and it willcrossload other GPROCs and cards in the XC. If a database object was included in theXC-PC’s XC release directory, at some point the XC will observe the “Cust BSS MMI-0115 ->” prompt. Go to the emon prompt by typing chg_le, followed by a cntrl-N. Thiswill complete the XC initialization and the XC will come into service.

• If you do not have a database object, the XC will say that a sysgen is needed, and youmust type “sysgen on”. After changing security levels, run a script with commands toconfigure the devices on a PC connected to the OMP GPROC. Type “sysgen off”, thenreset the XC. Observe the initialization sequence at the OMP GPROC by putting it inemon monitor mode (control-n, or change_level followed by control-n).


Conclusions:• The Transcoder is fully hardware redundant.

• GPROCs talk to other full-size cards in a shelfthrough the MCAP bus.

• GPROCs talk to full-size cards in other shelvesthrough the token ring LAN.

• The TDM bus handles all data traffic to/from theMSC, BTSs, and between FEPs and other GPROCs.

• The SBUS allows a GPROC to communicate withhalf-size cards, and with AMR card.

• MSIs handle two T1 or E1 interfaces.

• Troubleshooting an XC requires knowledge of XCsub-systems and internal XC connectivity.

• Sudden Trouble on an XC is usually indicated byXCLINK or MXLDL alarms.

• The “lock_device” and “unlock_device” commandsare used to disable/enable devices in the XC.

• The XC-PC is used with a program called “lanload”in order to download an XC’s software.


PWID Training: CDMAThe SS7 Sub-System: MM and MSC


Objectives:

• To present the SS7 sub-system, with an emphasis onoperation and debugging typical problems.


Table of Contents:

SS7 Definitions ............................................................................................................. 167Point Codes, A7 Terminal Handler .............................................................................. 169More on SS7 Links ....................................................................................................... 171The MSC & SS7- creating the SS7 sub-system ........................................................... 173SS7 Commands at the EMX......................................................................................... 175Sample Status Commands for SS7 on EMX2500 ....................................................... 177Debugging SS7 Link Problems .................................................................................... 179SS7 Definitions ............................................................................................................. 167Point Codes, A7 Terminal Handler .............................................................................. 169More on SS7 Links ....................................................................................................... 171The MSC & SS7- creating the SS7 sub-system ........................................................... 173SS7 Commands at the EMX......................................................................................... 175Sample Status Commands for SS7 on EMX2500 ....................................................... 177Debugging SS7 Link Problems .................................................................................... 179

PWID Center for Excellence Revision 0, 4/24/97 The SS7 Sub-System: MM and MSC


SS7 Definitions

• SS7- transports telephony signaling information.

• Point Codes- used to identify network elements.

• STP- Signalling Transfer Point

• DPC- Destination Point Code

• OPC- Origination Point Code

• SCP- Signaling Control Point

EMX

MSC

EndEndOffice

Office

STP

STP STP

STP

EndOffice

SP

CCSSO

SSP

STP STP

SCP SCPAIN AIN

SCP SCP

Signaling Point Codes

NETWORK

Cluster

Member

Network ID1 to 254

Cluster ID0 to 255

Member ID0 to 255

Signal Point0 to 7

Area/Netwrk 0 to 255

Zone0 to 7

C7 Point Cd.

SS7 Point Cd.



NOTES:

SS7 Definitions

• Common Channel Signalling System Number 7 (SS7) is an internationally standardizedsignaling system for telephony applications.

• SS7 defines a set of messages and flows which handle call control (setup, tear-down,billing, etc.). It also defines a communications protocol which implements managementfunctionality for the SS7 links.

• STP is a “Signaling Transfer Point”, and can be thought of as a switch capable ofrouting traffic to other points in the network.

• SCP is a “Signaling Control Point”. The signaling control point usually originates SS7messages, and/or is the “sink” for many SS7 messages in the system. The SCP is usuallywhere a database will reside (for example, the 800-number database for the PSTN).

• Links which connect any two points in an SS7 network are called “linksets”. The linksetcan be composed of one to sixteen links of 56 Kbps or 64 Kbps.

• Each linkset is identified by its Signaling Link Code (SLC) in order to route messages.

• A Destination Point Code (DPC) is a three octet number which uniquely identifies thenode that an SS7 message is destined to.

• An Originating Point Code (OPC) is a three octet number which uniquely identifies thenode that an SS7 message came from.

• Nodes in the SS7 network also have NODENAMEs, which are associated with pointcodes such that the nodename may be used to route messages to particular point codes.



Point Codes, A7 Terminal Handler

• “A7 Terminal Handler” allows operator interface toSS7 sub-system on MM.

• SS7 parameters configured on MM throughport_commands file.

• OPC & DPC are defined relative to the platform.

• CDF file on MM must match DPC & OPC informa-tion on MSC & MM’s SS7 sub-system.

V.35 SS7 Link, 56/64 Kbps

EMX2500 MMOPC: 250-150-001DPC: 251-151-001

DPC: 250-150-001OPC: 251-151-001S7Dpc: 104186S7Opc: 104443



NOTES:

Point Codes, A7 Terminal Handler

• The SS7 package on the MM includes an MMI (Man-Machine Interface) which allowsan operator to perform actions on the SS7 software stack. The interface is called the “A7Terminal Handler”. The SS7 links and link parameters may be modified through thisinterface, as well as other operations such as obtaining status of links, and setting pointcodes.

• The SS7 links must be provisioned as part of an initial MM install. Usually, a file called /usr/a7/conf/port_commands contains A7 Terminal Handler commands for provisioningthe links. The first line will have the MM’s point code, while all other point codes are forthe MSC.

• The point codes in the /screl/active/loadable/cbsc-1.cdf MUST MATCH those used bythe MSC and in the SS7 software package. In the CDF files, the point codes are decimalrepresentations of the hexadecimal point codes entered into the MSC and the A7 termhandler, when the first and third octets are swapped. For example, a point code of 251-151-001 will be converted into hex: FB.97.1, then read from left to right: 1.97.FB, andconvert this to decimal: 104443. This would be the number used in the CDF file.

• All the commands available from the A7 terminal handler (which is started as follows:)a7termhandler -node A7 -logicalname a7th -destname PM

ALW-MON:NAME=”string”; ALW-PRINT[:ID=”string”]; ALW-SNAPSHOT[:FILE=”string”];BKUP-APPL:FROM=”string”,TO=”string”; BKUP-NODE; DISPL-BKUP;DISPL-MON[:PRT=YES/NO]; DISPL-PSTATUS; DISPL-PURGE;DISPL-QSTATUS; INH-MON; INH-PRINT;RST-APPL:FROM=”string”,TO=”string”;RST-NODE:SOURCE={PRIMARY/SECONDARY/TERTIARY},[CHGLOG=YES/NO];RTRV-ALARM[:DATE=ALL]; RTRV-ERROR[:DATE=ALL]; RTRV-ILLOG[:DATE=ALL];RTRV-INFO[:DATE=ALL]; SCH-BKUP:DAYS=number; SCH-PURGE:DAYS=number; STOP-NODE;ACTV-LSET:LSET=identifier; ACTV-SLK:SLK=identifier; ALW-MEAS:ID=MTPXMT/MTPRCV;ALW-RSET:RSET=identifier; CHG-CPC:SSN=number,ADD/DEL=number;CHG-REMSSN:PC=number,ADD/DEL=number; CHG-REPCPC:SSN=number,PC=number;CHG-RSET:RSET=identifier,ADD/DEL=identifier,PRIORITY=number;CHG-RSET:RSET=identifer,LOADSHR=YES/NO;CHG-TIMER:TIMER=identifier,TVAL=numberM/S/MS; CRTE-CPC:SSN=number,PC=number;CRTE-LSET:LSET=identifier,PC=number; CRTE-OSPC:PC=number;CRTE-REMSSN:PC=number,SSN=number; CRTE-REPCPC:SSN=number,PC=number;CRTE-RSET:RSET=identifier,PC=number,RTES=identifier,LOADSHR=YES/NO;CRTE-SLK:SLK=identifier,LSET=identifier,SLC=number,SPEED=number,PORT=number;DEACT-LSET:LSET=identifier; DEACT-SLK:SLK=identifier;DISPL-CPC; DISPL-LSET; DISPL-OSPC; DISPL-REMSSN; DISPL-RSET; DISPL-SLK;DISPL-SSN; DISPL-TIMER;DLT-CPC:SSN=number; DLT-LSET:LSET=identifier; DLT-OSPC; DLT-REMSSN:PC=number;DLT-REPCPC:SSN=number; DLT-RSET:RSET=identifier; DLT-SLK:SLK=identifier;INH-MEAS:ID=MTPXMT/MTPRCV; INH-RSET:RSET=identifier;RTRV-MEAS:ID=SCCP/MTP,TIME=00:00;



More on SS7 Links

• Example port_commands file for MM.

• Looking at SS7 logs on the MM- Turning logging on/off.

- Log file: /usr/a7/Logs/Monitor.log

• Gathering information when problems arise- Turn logging on for aprox. 5 minutes.

- Run get.tandem.data script.

- Turn logging off.

- Send information to Motorola.

OMC-R

Hmm... I’m gatheringinfo as fast as I can!



NOTES:

More on SS7 Links

• An example port_commands file with all the commands to provision links on the MMfollows. Please note that port 0 on the SS7 card is actually the bottom-most port!

CRTE-OSPC:PC=251-151-003;CRTE-LSET:LSET=LSET1,PC=250-150-50,TYPE=F;CRTE-RSET:RSET=RSET1,PC=250-150-50,RTES=LSET1;CRTE-SLK:SLK=SLK00,LSET=LSET1,SLC=0,SPEED=64K,PORT=0;CRTE-SLK:SLK=SLK01,LSET=LSET1,SLC=2,SPEED=64K,PORT=1;CRTE-SLK:SLK=SLK02,LSET=LSET1,SLC=4,SPEED=64K,PORT=2;CRTE-SLK:SLK=SLK03,LSET=LSET1,SLC=6,SPEED=64K,PORT=3;CRTE-SLK:SLK=SLK04,LSET=LSET1,SLC=1,SPEED=64K,PORT=4;CRTE-SLK:SLK=SLK05,LSET=LSET1,SLC=3,SPEED=64K,PORT=5;CRTE-SLK:SLK=SLK06,LSET=LSET1,SLC=5,SPEED=64K,PORT=6;CRTE-SLK:SLK=SLK07,LSET=LSET1,SLC=7,SPEED=64K,PORT=7;ALW-RSET:RSET=RSET1;CRTE-CPC:PC=250-150-50,SSN=254;CRTE-REMSSN:PC=250-150-50,SSN=254;ACTV-SLK:SLK=SLK00;ACTV-SLK:SLK=SLK01;ACTV-SLK:SLK=SLK02;ACTV-SLK:SLK=SLK03;ACTV-SLK:SLK=SLK04;ACTV-SLK:SLK=SLK05;ACTV-SLK:SLK=SLK06;ACTV-SLK:SLK=SLK07;CHG-SLK:SLK=SLK00,SLT=NO;CHG-SLK:SLK=SLK01,SLT=NO;CHG-SLK:SLK=SLK02,SLT=NO;CHG-SLK:SLK=SLK03,SLT=NO;CHG-SLK:SLK=SLK04,SLT=NO;CHG-SLK:SLK=SLK05,SLT=NO;CHG-SLK:SLK=SLK06,SLT=NO;CHG-SLK:SLK=SLK07,SLT=NO;

• A useful logfile for the SS7 sub-system is /usr/a7/Logs/Monitor.log. However, loggingmust be enabled if very useful information is expected (otherwise, only the basics arelogged). Logging is turned on for the SS7 sub-system with the following command:

/usr/a7/bin/monitor -ln A7.SCCP -m 15

Logging is turned off with the following command. Please note that logging should notbe turned on for commercial systems during busy hours.

/usr/a7/bin/monitor -ln A7.SCCP -m 0

• A script called “get.tandem.data” is run in order to analyze and collect information forthe SS7 sub-system, for later analysis by Tandem in order to debug problems with theSS7 software and/or hardware. This script will be included in /usr/a7/bin in laterreleases of the software, but is available now in /home/hilton0/perez/supercell/notes/tandem/get.tandem.data in the Arlington Heights Sun network.



The MSC & SS7- creating the SS7 Sub-System

• On the left, EMX2500 Common Channel SignalingProcessors.

• On the right, steps necessary for creation of SS7Database.

Call ProcMgr. Ext.

Link Ext.Mgr.

(LNKEXT)

CommonChan. Mgr.

(CCM)

CommonChan. Dist.

(CCD)

CommonChan. Link.

(CCLK)(0-31)

(0-1)

EMX2500 CCS

Make CBSC Node

Disable Build Mode

Equip A+ Link

Set Link Parameters

Add link to Linkset

Add linkset to routeset

Enable Route

Refresh Procs

Activate links, linksets

Set Timers

Enable Build Mode

Equip Processors



NOTES:

The MSC & SS7- creating the SS7 sub-system

• The Link Extension Manager processor (LNKEXT) provides tandem message transportbetween SS7 subsystem and the rest of the switch. The LNKEXT allows the SS7subsystem to pass call processing signalling message to the line trunk managerprocessors for call setup and trunk circuit maintenance functions, and for the LTMs tosend messages to the SS7 system.

• The Common Channel Manager processor (CCM) provides management of the SS7network. It is necessary to remove/restore this component when changes are made to theSS7 network since its memory holds information on the network. The CCM routesmessages between the LNKEXT and CCD processors and contains an interface to theTraffic Metering and Measurement (TM&M) subsystem.

• The Common Channel Distributor processor (CCD) provides message formatting androuting to and from the links. The CCD is responsible for Link Selection for outgoingmessages. It also determines if an incoming message is destined for this EMX, and routesthe message accordingly.

• The Common Channel Link processor (CCLK) implements the link layer protocol. Itprovides the physical link through either RS449, V.35, or RS232 (depending on theimplementation). The CCLK will normally be connected via V.35 to the MM, through aboard in the rear of the CCS frame which will have either a DTE or a DCE Winchester-type connector for the cable to the DSU. Although the CCLK may have 2 physical ports,this is only possible for bit rates below 19.2 Kbps- since our links are 56 Kbps, we end upusing only one link per CCLK.

• The equipage of the processors is usually done only once. There is information availablein the Arlington Heights FOA group’s home page on the SS7 subsystem. MosesKunamalla has written a document which describes the commands necessary to buildthe database.

• Making the CBSC node involves creating sub-system names and indexes for the EMXand the CBSC. One or more A+ links are then equipped, and their parameters(including speed, time-outs, format, and others) are set. The A+ link is then added to alinkset. Then the linkset is added to a routeset, and the route is enabled.

• The processors for the CC subsystem are then reset, the links and linksets are thenactivated. Then the proper timers are set for the links. The CCLK processors will begintrying to align the links (the orange light will alternate on/off with a period of aprox. 3seconds). When the links are aligned, the orange light on the CCLK will stay on.



SS7 Commands at the EMX

• Excerpt from commands used to build the SS7 sub-system on the EMX.

• Voice Trunks and CICs must be added as well.

EMX2500

I’m building a database,building a database,building a database,ha, ha, ha, ha!



NOTES:

SS7 Commands at the EMX

• The following is an excerpt from a list of commands which are input at the MSC in orderto configure the SS7 links. Commands such as these normally must be performed whenthe MSC is first installed, and when any changes are performed to the SS7 network.The following information is assumed in this setup:Local Node: EMX1 250-150-001 CBSC Node: CBSC1 251-151-001Local node subsystem: EMXSS CBSC node subsytem: CBSCSSInitial number of A+ links is one (using link id 0)Initial number of voice spans is one with 23 circuits (span 16, ckts 385-407)Voice trunk group is 700Links configured are V.35 DTE at 64Kbps external clocking from DSU with NRZ!Commands !Network Assignment !Enable Build ModeBUILD CCS NETWORK E YCHANGE CCS NODE CBSC1 Y N 001 151 251 Y !Establish CBSC NodeASSIGN SUBSYS INDEX EMX1 10ASSIGN SUBSYS INDEX CBSC1 10 YCHANGE SCCP SSNAME EMXSS Y N EMX1 254 YCHANGE SCCP SSNAME CBSCSS Y N CBSC1 254 YCHANGE SCCP SSSTAT CBSCSS Y A N YCHANGE SCCP SSSTAT EMXSS Y A N YCHANGE SCCP SUBSYS EMXSS Y EMX NCHANGE CCS USERPA SCCP YONDMX N PA YBUILD CCS NETWORK D !Disable Build ModeSTATUS CCS PROCS ; !Verify Processors are in ServiceCHANGE CCS LINK 0 EQ Y !Equip A+ linkCHANGE CCS LINKPA 0 1 1 4 4 NRZ !Establish Link ParametersCHANGE CCS LINKPA 0 2 N 256 N YCHANGE CCS LINKPA 0 3 50 29 64 80 60 100 110 90 130 140 120 160 YCHANGE CCS LINKPA 0 4 D S YCHANGE CCS LINKPA 0 5 400 400 4 YCHANGE CCS LINKSE LSCBSC Y AS CBSC1 ; ; F 1 0 1 0 Q Y Y !Add Link to LinksetCHANGE CCS RSMAST 4 LSCBSC 1 Q Y !Add Linkset to RoutesetCHANGE CCS DPCRS EMX1 E Y !Enable RouteCHANGE CCS DPCRS CBSC1 4 E D Y Y A Y YREMOVE LNKEXT PROC A B !Refresh ProcessorsRESTORE LNKEXT PROC A I ; REMOV LNKEXT PROC B B ; RESTORE LNKEXT PROC B IREMOV CCM PROC A B ; RESTOR CCM PROC A I ; REMOV CCM PROC B BRESTOR CCM PROC B I ; REMOV CCD PROC A B ; RESTOR CCD PROC A IREMOV CCD PROC B B ; RESTOR CCD PROC B I ; REMOV CCLK PROC 0 BRESTOR CCLK PROC 0 IDEACTIVATE CCS LINKSE LSCBSC !Activate Link’s and LinksetsACTIVATE CCS LINKSE LSCBSC ; DEACTIVATE CCS LINK 0 ; ACTIVATE CC LINK 0 Y!Set TimersCHANGE CCS TIMERS MTP 1 1 0 Y MTP 2 1 0 Y MTP 3 1 0 Y MTP 4 1 0 Y MTP 5 1 0 YCHANGE CCS TIMERS MTP 6 1 0 Y MTP 8 1 0 Y MTP 10 30 0 Y MTP 11 30 0 YCHANGE CCS TIMERS MTP 12 1 0 Y MTP 13 1 0 Y MTP 14 3 0 Y MTP 15 3 0 YCHANGE CCS TIMERS MTP 16 2 0 Y MTP 17 1 0 Y MTP 18 30 0 Y MTP 20 100 0 YCHANGE CCS TIMERS MTP 21 100 0 Y MTP 98 60 0 Y MTP 99 60 0 YCHANGE CCS TIMERS SCCP 0 180 0 Y SCCP 1 300 0 Y SCCP 2 1320 0 Y SCCP 3 20 0 YCHANGE CCS TIMERS SCCP 5 30 0 Y SCCP 9 30 0 Y SCCP 10 30 0 Y!Add Voice TrunksCHANGE MTN T1THRS 16 EQ 1 3 15 20 30 3 10 20 30 35 ;;;;;;;;;;;;; CCHANGE MTN T1THRS 16 ;;;;;;;;;;; 100 30 100 4 10 12 N Y Y Y N Y Y CCHANGE MTN TRKCKT 385-407 EQ MCHANGE MTN TRKGRP 700 EQ 0 0 0 0 0 0 YCHANGE CP TRKGRP 700 ; ; 0 M 0 N N N BSS ALL YCHANGE CP CKTTBL 385-407 CC7TRK 0 0 Y MCHANGE CP TRUNK 385 700 TWOWAY N 1 YCHANGE CP CIC 385 ADD 385 CBSC1 YDISPLAY BSS BSSRTE !Check to see if the BSS route is correctDISPLAY BSS CELRTE !For Hard-Handoffs checkt the cell route table



Sample Status Commands for SS7 on EMX

• Display point code and sub-system name informa-tion.

• Display link parameter information.

• Display linkset information.

• Status the links.

OMC-R

MM

These SS7 links won’talign!! What do I do?



NOTES:

Sample Status Commands for SS7 on EMX2500

• To display point code and node name information:DISP SCCP SSSTATDISPLAY SCCP SSSTATUS DATE: 1/11/96 TIME: 16:02:19SUBSYSTEM NAME : EMXSSSS SS SEND SSNAME NODE NAME POINT CODE NO. TRAF ASSIGNMENT REPL REPL LOCATION------ ----------- ----------- --- ---- ---------- ---- -------------EMXSS HKEMX 250-150-001 254 YES ASSIGNED NOSUBSYSTEM NAME : BSSSSSS SS SEND SSNAME NODE NAME POINT CODE NO. TRAF ASSIGNMENT REPL REPL LOCATION------ ----------- ----------- --- ---- ---------- ---- -------------BSSSS HKMM1 251-151-001 254 YES ASSIGNED NO

• To display link parameter information:>DISP CCS LINKPALINK# : 0PARAMETER GROUP : 1DISPLAY CCS LINKPARMS FOR LINK: 0 DATE: 1/11/96 TIME: 16:07:22CURRENT PARAMETERS : PHYSICAL CHANNELINTERFACE BAUD RATE CLOCKING SIG ENCODE-------------- --------- -------- ----------RS449/V.35 DCE 64KB internal NRZ

• To display linkset information (it’s important that SLC be the same as for the MM):DISP CCS LINKSELINK SET : LS001CCS LINKSET : LS001 DATE: 1/11/96 TIME: 16:10:00ASSIGN ADJACENT SRTM COMBINED LS ACT LINKSSTATUS NODE COMP LINK SET TYPE ALLOWED====== =========== ======= ======== ==== =========AS HKMM1 Issue 1 ******** A 4

LINKS IN LINKLINK SET PRIORITY SLC======== ======== ===0 1 32 1 24 1 16 1 0

• Status links:STAT CCS LINK N 0-2CCS LINK STATUS DATE: 3/19/96 TIME: 12:34:23 SERVICE BLOCKED INHIBITLINK # STATUS STATUS STATUS LINK SET CHANNEL STATUS------ ------------------------- -------- -------------- 0 AOS UB UI LS001 in service 1 NE UB UI ******** in service 2 AOS UB UI LS001 in service

• Other “interesting” commands:>DISP CCS DPCRS -> To find information on the destination point code>STAT SCCP SUBSYS -> Shows status of the EMX CCS subsystem>LIST CCS NODE ; ; -> Lists all the node names and their pointcodes>LIST SCCP SSNAME ; ; -> Lists the SS Name, Node Name, Point Code and SS Number.



Debugging SS7 Link Problems

• Trace signal path- make sure there is a physical con-nection.

• Check SLC number assignments, point codes.- Use a7 terminal handler’s “displ-slk;, displ-ospc,

displ-dpc” commands.

- Use MSC’s DISP CCS LINKSE, DISP SCCPSSSTAT commands.

• Check DSU programming.

• Check link speed assignment.

• Re-seat DSUs.

• Replace cables.

• Replace DSUs.

V.35 56/64 Kbps

EMX2500 MMOPC: 250-150-001DPC: 251-151-001

DPC: 250-150-001OPC: 251-151-001S7Dpc: 104186S7Opc: 104443

DTE/DCE Interface Board

V.35, to SS7 Board 0, port 0

DSUs- provide Timing,link isolation.

T1 Spans- connected through DSX



NOTES:

Debugging SS7 Link Problems

• One way to avoid problems altogether (and to make them easier to find when they doappear) is to label all components correctly. All DSX (Digital Cross-Connect) jacksshould be labeled properly, as well as all cables.

• The V.35 connections are present on the MM’s SS7 card. It is important to note that SS7card 0 (in slot 165) will have port 0 on the bottom. Port 0 is usually the first to beconfigured- this port will be defined with SLC of 0 as well (see the port_commands file).The SLC on the MSC must be the same for this link! (use the DISPL CCS LINKSEcommand).

• The DSUs provide link isolation, timing for the link, and also they provide a means tomonitor the links through the Digital Cross Connect (DSX). An SS7 LinkSet Monitor(such as the Wandel & Goltermann PA-41, or the HP 37900) may be connected to the“monitor” ports on the DSX.

• The DSUs should be checked if the links don’t align. A loop-back may be performed atthe DSX in order to isolate the faulty DSU. For example, if the link aligns at the EMXwhen the EMX’s DSU’s T1 connection is looped-back, then the MM’s DSU is suspect.

• The proper settings for the DSUs are documented in the Arlington Heights FOA grouppage, under FOA/DSU. Note that when two DSU’s are connected back-to-back (as in thediagram on the previous page), one of them must be setup as a DCE in order to providetiming on the link.

• A common source of problems is improper link speed specification on one of thecomponents. Note that the link speed is specified in various places: on the EMX (DISPCCS LINKPA), in the MM’s SS7 software (a7 term handler: displ-slk;), in the MM’scbsc-X.cdf file (check for C7LinkSpeed), and in the DSUs as well. All of these speeddefinitions must be the same for the link to align.

• Yet another problem which may cause link alignment failure is the point code settings onthe MM and MSC. Again, the operator must make certain that the point codes havebeen set the same in the MSC (use DISP SCCP SSSTAT), in the MM’s SS7 software (usea7 term handler: displ-ospc; and displ-dpc commands), and in the MM’s cbsc-X.cdf file(check for S7Dpc, S7Opc).

• As part of the troubleshooting procedures, the DSUs should be re-seated. If problemspersists after all the data checks out okay (point codes, link speeds, link parameters,etc.), then the DSUs should be replaced. The cables could also be replaced as adebugging step. Also, an SS7 linkset analyzer (if available) can prove to be an excellentdebugging tool (most analyzers can emulate link alignment).


Conclusions:• The SS7 communications protocol is used for commu-

nication between MM and MSC.

• Point codes must be defined correctly on MM and onMSC.

• The a7 terminal handler is used on MM to display,change, and status the SS7 sub-system on the MM.

• Several commands must be run on both the MM andthe MSC in order to provision the SS7 links correctly.

• Any of a number of components in the SS7 path maybe incorrectly configured or damaged.

• Debugging and problem isolation is used to find theculprit for faulty SS7 links.


PWID Training: CDMAThe SC9600 BTS


Objectives:

• To present the SC9600 bts utilized in Supercell CDMAin detail- including architecture, hardware, and oper-ations.

• To present briefly the other BTS alternatives forCDMA.


Table of Contents:

The SC9600 BTS Architecture..................................................................................... 185BTS Operations ............................................................................................................ 187More BTS Operations Notes ........................................................................................ 189Other BTSs for CDMA................................................................................................. 191

PWID Center for Excellence Revision 0, 4/24/97 The SC9600 BTS


The SC9600 BTS Architecture...

...

Rx Rxd Rxd1a 1b 3b

12

6

...1a 1b 3b

126

...

2

1a 1b 3a 3b...

6

2 ...1...

12

33

6 6

3 3 3 3 3 3

Preselector /Splitter

3 3

3

1a 1b3a 3b...

...1

R

PR

43

1 ...20

PR

PR

Rx/Rxd ... 12

PR

PR

Tx

32

1

32

1

Tx

Dir. Coupler

MIO Rx MIO Tx

Tx Combiner

Dir. Coupler

Cell Site Filter

Multicoupler

Preselector

Preselector I/O

BBX

BDC

MCC

GLI

MSIO

AMR

CSM

GPS

LFRClk/Sync

Timing andSynchronization

MCC Bus

CCP Shelf

AMR BUS (RS485)

CustomerAlarms

Alarms

From MIO’s ofother CCP shelves

To MIO’s of otherCCP shelves

T1

T1/E1

= primary= redundant

Note:

3-sector system shown,Assumes Tx & Rx non-shared antennas

To/From

FRAME

LPA FRAME

RF MODEMFRAME

(RS485)

Trial system will use omni xmit & rcv(i.e.: 2 rcv, 1 xmit antenna)

To other GLI’sin same frame

ISB

Ethernet

RS485

RS485

RFDS(not in CDMA Trial yet)

PR

(SCAP/TCP/IP)

To other GLI’sin other RF Modem

Frames

To LMF

T1/E1

To/Fromother BTS’CBSC

CDMA Trial will not use RFDS

AM

R B

US

(R

S48

5)

Eth

erne

t

RGLI

33 33

12

Clk/Sync

AMR BUS

41a,bI,Q ...3a,b

I,Q

1a,b..3a,bI,Q

RS232

36

6

Clk/Sync

To otherRF ModemFrames

In Trial System, MIO Tx outputs willbe routed directly to (NAMPS) LPA’s

SITE INTERFACE

LCILPA



NOTES:

The SC9600 BTS Architecture

In this diagram we present the internal connectivity of the BTS, including RFand Baseband signal flow.

• Note the demarcations for each of the frames: SIF frame on the top, LPA frame in themiddle right, RF Modem Frame on the bottom.

• The SIF frame presents all the RF interfaces to the RFMF on the Rx path. Note that theRFDS uses the directional coupler in order to test Tx and Rx paths at the BTS. TheRFDS is not currently available for CDMA.

• From the Multicouplers, the Rx RF signal is fed to the RFMF. The RF signal is deliveredto the BBX cards (one per CDMA sector). The BBX cards perform all the up-conversionand down-conversion.

• The baseband signals from the receive antennas are then fed to the BDC card, which willperform antenna diversity functions to select the best signal, and it will remove the Pnoffset for the particular cell or sector. The resulting signal is forwarded to the MCCcards.

• The MCC cards will be directed by the MM to “listen” for particular codedconversations on the reverse link (Rx path), and the de-coded conversation will be sent tothe XC via the GLI and MSIO cards through a T1 or E1 interface.

• On the transmit path, the information to be sent will be received from the XC throughthe T1 or E1 interface, through the MSIO and GLI cards, and “picked up” from theMCC bus by the MCC card which has been assigned to the particular conversationbeing transmitted.

• The MCC card will code the conversation so that it can be mixed with all otherconversations and still be de-coded by the mobile (Spread-Spectrum spreading with aparticular Walsh code), and the resulting signal will be sent to the BDC.

• The BDC will combine all signals being transmitted from all MCC cards, and place thePn offset for the cell or sector on the combined digital signal. This signal is then fed tothe BBX.

• The BBX will up-convert the digital signal to RF, and send the resulting RF signal to theLPA frame where it will be amplified and sent to the SIF for transmission on the Txantennas.



BTS Operations

• Optimization- See guide by TED.

• LMF is a Unix-based platform.- Presents a CLI interface.

- Allows opeator to download code, data to BTSdevices.

- Allows operator to optimize BTS through auto-mated scripts (see notes on Ksh on page 92).

- Use Ethernet “tips & tricks” on page 244 whenLMF can’t ping the GLI.

• GLI has an MMI interface at which commands maybe entered for troubleshooting.





NOTES:

BTS Operations

Very good material is available from TED on installing the BTS, and on optimiz-ing the BTS- therefore we will not go into these topics. However, the followingpoints are presented on the LMF and on the BTS devices:

• The LMF is a unix-based (Sun Solaris) laptop. Many of the “tips & tricks” for Unix maybe used on the LMF. In particular, many times the LMF cannot login to the BTS. Whenthis occurs, try to ping the GLI (look in the /etc/hosts file to find the GLI’s address).

• If the GLI has no address in the /etc/hosts table on the LMF, then lookup the GLI’saddress by connecting an MMI cable to its front-panel RS-232 port to the LMF’s serialport. Press <esc><return> to obtain the RT> prompt. At this prompt, enter “config”,and this should print the IP address of the GLI. Enter this IP address into the /etc/hostsfile on the LMF. Then the LMF should be able to ping the GLI, and the LMF applicationshould initialize okay.

• When problems are suspected at the GLI (E1 link coming in/out of service, for example),connect a terminal to the MMI port of the GLI, and enter the following commands(make sure to capture the output from the commands in a file on the terminal):

RT> printf on <ret>RT> status <ret> <esc> <esc>RT> crashp 80 80 80RT> rd_path 1 (notice the output when you see .... Link ID=X .... )RT> rd_link X (get X from last command)RT> configRT> scap_addrRT> dpll <esc><ret>RT> lapd_stat 0 <ret><esc><ret>RT> dmsg 50<ret)RT? printf off

• Another usefull troubleshooting command for the GLI is:RT> lstat <ret><esc>

This command shows all the device the GLI sees on the LAN. The GLI will be able tocommunicate with the device if the “Conn” field says “CONN”, and the “State” field is“OK”.

• The “ping” command may be used at the GLI in order to see if the GLI can “see” theMM: ping 128 0 0 0 0. The dmsg command will print the last <n> messages. After a ping,you should see a message type “3600” (ping), followed by a message type “3602” (pingresponse).



More BTS Operations Notes

• Resetting a GLI remotely (used when comm link toMM is down).

• Checking span line information (use span_viewcommand at GLI’s MMI).

• Changing GLI’s IP address.





NOTES:

More BTS Operations Notes

• Resetting the GLI:

• Dial into the BTS through a “cu” command at the OMC-R, or through a ProComm(or similar) package on a PC with a modem.

• A CONNECT mesage will be seen, followed by several numbers (ignore the num-bers). Type the <esc> key, followed by <return> key to get the RT scope prompt(“RT>”).

• Use the “status” command to get the status of the GLI. If GLI has cage control,enter sndtype 16387 <ret> <esc> (16387d = 4003h = SCAP message for OOS_RAM).If not, enter send 0 0 128 48 255 16387 <ret><esc> . The '128 48 255’ is the decimalrepresentation of the SCAP address for the GLI, the GLI in the bottem cage = 48and in the top cage = 47.

• Now check the status of the LAN connections with the “lstat ” command. Noticethat a column labeled ‘CTL’ is decoded as follows:B - INS_STANDBY, A -

INS_ACTIVE, 3 - OOSROM, 4 - OOSRAM . If the GLI state is OOSROM, do the following:RT> cmd; set 40004eccA number will be displayed, after this number type: 4003. <ret>RT> ping 0 0 128 48 255 (this will reboot the GLI)RT> cmd; set 40004eccA number will be displayed, after this number type the original number that wasdisplayed the first time followed by .<ret>

• To check span configuration at the GLI, try this command (example output shown):RT>span_viewSpan Type = T1_3Span Rate = 56Span A Type = T1 long haulSpan B Type = T1 long haulLapd slot for Span A = 0Lapd slot for Span B = 0

• To change the GLI’s IP address from 128.0.0.2 to 136.182.108.51:> RT>cmd> RC>set.b 04000006> 04000006 80 ? 88> 04000006 08> 04000007 00 ? b6> 04000007 36> 04000008 00 ? 6c> 04000008 AC> 04000009 02 ? 33> 04000009 B3> 0400000A 00 ? .> RC>go



Other BTSs for CDMA

• Naming Conventions:- Four digits used, first omitted when no narrow-

band channels included.

- When four digits used, first two digits indicate # ofnarrow-band transceivers.

- Third digit will be ‘2’ for CDMA-only, ‘5’ formixed.

• SC4850- 48 NAMPS channels, 1.9 GHz, 108 CDMA chnls.

• SC600- No NAMPS channels, 36 CDMA chnls.

• SC2450- 24 AMPS/NAMPS channels, 800/900 MHz, 108

CDMA chnls.

SC600

SC4850 SC2450



NOTES:

Other BTSs for CDMA

• The naming convention for BTSs in CDMA is defined as follows. A “series” is a BTStype which can have several options. A series is identified with four digits. Those serieswhich do not include narrow band transceivers have only three digits (i.e. SC 600). Aseries which includes narrow band transcievers has four digits (i.e. SC9600). The thirddigit with be a ‘2’ if a CDMA-only BTS, ‘5’ if its a mixed CDMA-analog BTS. Thefourth digit will be used if necessary in the future for major product variations.

• The SC4850 BTS for CDMA is a 1.9 GHz-only system, capable of providing CDMA andNAMPS air interfaces. Input voltage is +27 VDC.

• Dimensions are 2100mm height, 800mm width, 600mm depth. Maximum weight is430 kg (950lbs).

• Capacity is 48 analog (NAMPS) transceivers, and 2 CDMA carriers of 3 sectorseach. Using 18 CDMA channels per sector (practical maximum determinedthrough field trials), this gives 108 CDMA channels.

• The SC2450 BTS (currently called the SC2400) provides AMPS/NAMPS/CDMA airinterfaces at 800 or 900 MHz ranges. Input voltage is +27 VDC.

• Dimensions are 2100mm height, 800mm width, 600mm depth. Maximum weight is636 kg (1400lbs).

• Capacity is 24 analog (AMPS or NAMPS) channels, and 2 CDMA carriers of 3 sec-tors each, giving 108 CDMA channels.

• The SC600 BTS is a CDMA-only BTS designed for easy instllation outdoors. It is a self-contained, convection-cooled unit. An SC600 can support one Omni site with 36 CDMAchannels (maximum). The SC600 can be expanded with other SC600 units (stacked ontop of each other) to provide a maximum of 108 CDMA channels.

• With the exception of the SC600, all BTSs will have (sometime in the future) an optionfor a Radio Frequency Diagnostic System (RFDS), which is being developed in Ft.Worth.


Conclusions:• The SC9600 is a three-frame BTS, composed of the

RF Modem Frame, LPA Frame, and Site InterfaceFrame.

• The BBX is the demarcation point for the RF path.

• The GLI handles T1/E1 termination, acts as the site’scontroller.

• The BTS is designed such that fault tolerance is pro-vided for those elements which could malfunction intime.

• The LMF allows an operator to login to the GLI, anduse a CLI to input enable/disable commands andother commands to optimize the site.

• The GLI uses Motorola’s EXEC environment, whichpresents a prompt at which several commands may beinput for debugging/maintenance functions.

• The GLI may be reset remotely via a modem to thesite.


PWID Training: CDMAOutput from Work Sessions


Objectives:

• Present example flows from initializations, in order tofamiliarize the reader with the sequence of events.


Table of Contents:

Sample Output: Unix Initialization ............................................................................. 196Sample Output: Unix Initialization (Continued), SCSIIM Initialization .................. 198Sample Output: SCSIIM Initialization (Continued)................................................... 199Sample Output: Enable CBSC..................................................................................... 200Sample Output: Enable CBSC (Continued)................................................................ 201Sample Output: Enable CBSC (Continued)................................................................ 202Sample Output: Enable CBSC (Continued), BTS Enable.......................................... 203Sample Output: BTS Enable........................................................................................ 204Sample Output: BTS Enable........................................................................................ 205

PWID Center for Excellence Revision 0, 4/24/97 Output from Work Sessions


Cabinet configuration: Central OfficeIdentified Interphase Ethernet Controller in iop 0, slot 0Identified Interphase Ethernet Controller in iop 1, slot 0Identified Interphase SCSI controller in iop 0, slot 1Identified Interphase SCSI controller in iop 1, slot 1Identified Interphase SCSI controller in iop 0, slot 2Identified Interphase SCSI controller in iop 1, slot 2Identified Interphase Ethernet Controller in iop 0, slot 3Identified Interphase Ethernet Controller in iop 1, slot 3Identified Interphase Ethernet Controller in iop 0, slot 5Identified Interphase Ethernet Controller in iop 1, slot 5Identified Interphase Ethernet Controller in iop 0, slot 7Identified Interphase Ethernet Controller in iop 1, slot 7slot0: initialization succeededslot1: initialization succeededslot1: disk0 disk3 tape4 onlineslot2: initialization succeededslot2: tape0 disk1 disk5 onlineslot3: initialization succeededslot5: initialization succeededslot7: initialization succeeded

CPU Proms - t7437A14 - Fri Nov 05 13:23:02 CDT 1993Hit space within 10 seconds to interrupt rcboot sequence:Initiating rcbootcould not load disk(1,0,6)could not load disk(2,1,6)loading file disk(1,0,5)unix2349616+390992+1116272text entry:0x80031000, start: 0x80031000, end:0x8026ea2fdata start:0x8026ea30, end:0x802ce17fbss start:0x802ce180, end:0x803de9ef

CPU: MIPS/R4400 Processor Chip Revision: 5.0FPU: MIPS/R4400 Floating Point Coprocessor Revision: 0.0consattr=19200,8,n,1,n2boot_end=135266304red_zone_end=138412031Iop0: IOP is present (inserted)Iop1: IOP is present (inserted) cpu: cpua,cpub,cpuc 64MB local memory tmrc: tmrca,tmrcc[Primary] 16MB (base 0xa8000000) iop: iop0,iop1mlsetup: initializing memoryDelay multiplier = 27, cnt = 59mlsetup: reintegration diagnostic save area, size 4096 bytes, at a8fff000Event log driver initialized

NonStop-UX SVR4.0 -- Seq: 4 -- root -- Tue Mar 19 23:28:32 GMT 1996UNIX(R) System V Release 4.0 AT&T Version B32Total real memory = 82837504 (20224 pages)Root on dev 5/0Swap on dev 5/0, 32768K bytestotal real memory = 82837504total available memory = 76259328

Identified Interphase Ethernet Controller in iop 0, slot 0Identified Interphase Ethernet Controller in iop 1, slot 0Identified Interphase SCSI controller in iop 0, slot 1Identified Interphase SCSI controller in iop 1, slot 1Identified Interphase SCSI controller in iop 0, slot 2Identified Interphase SCSI controller in iop 1, slot 2Identified Interphase Ethernet Controller in iop 0, slot 3Identified Interphase Ethernet Controller in iop 1, slot 3Identified Macrolink async board, serial #390 in iop 0, slot 4Identified Macrolink async board, serial #390 in iop 1, slot 4Identified Interphase Ethernet Controller in iop 0, slot 5Identified Interphase Ethernet Controller in iop 1, slot 5Identified Interphase Ethernet Controller in iop 0, slot 7Identified Interphase Ethernet Controller in iop 1, slot 7Jaguar Version (077-34-01N) Date 01271992 with 126 Kbytes ram.Port0: Single EndedPort1: None

Sample Output: Unix Initialization



NOTES:

scsi0: initialization succeeded

scsi0 unit0: unit onlinescsi0 unit3: unit onlinescsi0 unit4: unit onlineAsync0: Initialization CompletedJaguar Version (077-34-01N) Date 01271992 with 126 Kbytes ram.Port0: Single EndedPort1: Nonescsi1: initialization succeededscsi1 unit0: unit onlinescsi1 unit1: unit onlinescsi1 unit5: unit online

OSI/TCP Application Facility Release 1.0IO subsystem initialization complete.Volume Manager startedStarting root volume... Donelog replay in progressreplay complete - marking super block as CLEANSwitching to root volumefsgen/volume: Volume cdl: log read from plex cdl-0 failed: No such device or addressfsgen/volume: Volume cdl: log read from plex cdl-1 failed: No such device or addressfsgen/volume: Volume cdl: No valid logs were foundfsgen/volume: Volume cdl: Cannot reconcile logs, volume not startedNode: rpomcrChecking for new unix kernelkeepalive daemon startedSystem V STREAMS TCP Release 1.0(c) 1983,1984,1985,1986,1987,1988,1989 AT&T.(c) 1986,1987,1988,1989 Sun Microsystems.(c) 1987,1988,1989 Lachman Associates, Inc. (LAI).All Rights Reserved.adding daemon ‘/usr/sbin/rpcbind’ (pid 92) respawn ‘/etc/keepalive.d/rpcbind’adding daemon ‘/usr/sbin/inetd’ (pid 93) respawn ‘/etc/keepalive.d/inetd’adding daemon ‘/usr/bin/ism/configd’ (pid 99) respawn ‘/etc/keepalive.d/configd’adding daemon ‘/usr/bin/ism/evlogd’ (pid 98) respawn ‘/etc/keepalive.d/evlogd’adding daemon ‘/usr/sbin/sproute’ (pid 120) respawn ‘/etc/keepalive.d/sproute’adding daemon ‘/usr/mds/midas’ (pid 149) respawn ‘/etc/mdsinit’adding daemon ‘/usr/sbin/syslogd’ (pid 154) respawn ‘/etc/keepalive.d/syslog’The system is going into fast multi-user mode. Please wait.vxfs mount: warning: <> mounted as </sc>vxfs mount: warning: <> mounted as </screl>vxfs mount: warning: <> mounted as </home>/sbin/spawndaemon: daemon ‘/usr/sbin/sproute’ is already executing/sbin/spawndaemon: daemon ‘/usr/sbin/inetd’ is already executingadding daemon ‘/usr/sbin/tp0d’ (pid 238) respawn ‘/etc/keepalive.d/tp0d’K: tp0open calledMar 19 23:38:28 1996: MIDAS: Boot has occurred.Async0: Ready/sbin/mvcload: Option ‘-i’ board information: Firmware part number: 450146 Firmware revision: 4.0 Board address: 0x0 Number of lines: 16 Printer port available: YES Amount of FIFO memory: 66K (0x10800) Nvram message: ‘PN221200RB1#390 ‘

Sample Output: Unix Initialization (continued)



The system is ready.Beginning background state 4 processingChecking for system core dump. Saving the system core dump (if present) will be done in the background.298keepalive daemon restarted

rpomcrWelcome to the Tandem S2 UNIX SystemConsole Login: could not find daemon ‘/sbin/vold’ (pid=0) to delete.Wed Mar 20 21:41:59 1996 rpmm1 PID: 316 scsiim: scsiim process started at time: Wed Mar 20 21:41:591996Wed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: initLevel=2 scsiim: initForced=0Wed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: This platform is an MM, with platform number: 1Wed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: Warning: CDF file has no RESOURCES::MutexLocks field.Defaulting to creating 150 locks and 10 bucketsWed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: Warning: Preserve item in CDF file is empty or notpresent. Defaulting to false.Wed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: Warning: Preserve item in CDF file is empty or notpresent. Defaulting to false.Wed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: Type for shared memory segment SC_DEVOM_CDF has beenspecified as MMAPWed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: Warning: Preserve item in CDF file is empty or notpresent. Defaulting to false.Wed Mar 20 21:42:00 1996 rpmm1 PID: 316 scsiim: Warning: Preserve item in CDF file is empty or notpresent. Defaulting to false.Wed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: SyslogdRefreshWed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: SyslogdRefresh instantiated.Pid: 2618Wed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: popperWed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: popper instantiated. Pid: 2619Wed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: MonitorAPIWed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: MonitorAPI instantiated. Pid:2620Wed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: scemlogdWed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: scemlogd instantiated. Pid: 2621Wed Mar 20 21:42:02 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: mmgwWed Mar 20 21:42:03 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: mmgw instantiated. Pid: 2622Wed Mar 20 21:42:03 1996 rpmm1 PID: 316 scsiim: Pausing for an additional 10 seconds after mmgwWed Mar 20 21:42:03 1996 rpmm1 PID: 316 scsiim: Pausing for an additional 10 seconds after mmgwWed Mar 20 21:42:13 1996 rpmm1 PID: 316 scsiim: Pausing for an additional 1 seconds after mmgwWed Mar 20 21:42:14 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: scevmgrWed Mar 20 21:42:14 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: scevmgr instantiated. Pid: 2727Wed Mar 20 21:42:14 1996 rpmm1 PID: 316 scsiim: Pausing for an additional 3 seconds after scevmgrWed Mar 20 21:42:17 1996 rpmm1 PID: 316 scsiim: Pausing for an additional 1 seconds after scevmgrWed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: sec_mgmtWed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: sec_mgmt instantiated. Pid: 2766Wed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: vdt_1Wed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: vdt_1 instantiated. Pid: 2767Wed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: vdt_2Wed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: vdt_2 instantiated. Pid: 2768Wed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: mm_slmWed Mar 20 21:42:18 1996 rpmm1 PID: 316 scsiim: Checking lockstep for process: mm_slmWed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: lockstep process: mm_slm brought up successfully. Pid:2769Wed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: dgenWed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: dgen instantiated. Pid: 2773Wed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: ddWed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: dd instantiated. Pid: 2774Wed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: scdbWed Mar 20 21:42:23 1996 rpmm1 PID: 316 scsiim: Checking lockstep for process: scdbWed Mar 20 21:42:39 1996 rpmm1 PID: 316 scsiim: lockstep process: scdb brought up successfully. Pid:2775Wed Mar 20 21:42:39 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: sc_devomWed Mar 20 21:42:39 1996 rpmm1 PID: 316 scsiim: Checking lockstep for process: sc_devomWed Mar 20 21:43:14 1996 rpmm1 PID: 316 scsiim: lockstep process: sc_devom brought up successfully.Pid: 2780Wed Mar 20 21:43:14 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: mmtadadmWed Mar 20 21:43:14 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: mmtadadm instantiated. Pid: 2782Wed Mar 20 21:43:14 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: cpmoWed Mar 20 21:43:14 1996 rpmm1 PID: 316 scsiim: Checking lockstep for process: cpmoWed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: lockstep process: cpmo brought up successfully. Pid:2783Wed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: callproc1Wed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: callproc1 instantiated. Pid:

Sample Output: Unix Initialization (Continued), SCSIIM Initialization



NOTES:

2785

Wed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: clcpWed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: clcp instantiated. Pid: 2786Wed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: Attempting to spawn process: fepMonWed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: Non-lockstep process: fepMon instantiated. Pid: 2787Wed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: scsiim has finished attempting to bring up processesas specified in CDF file.Wed Mar 20 21:43:20 1996 rpmm1 PID: 316 scsiim: scsiim of pid: 316 is pausing, waiting for INITindication.

Sample Output: SCSIIM Initialization (Continued)



rpmm1-010012 > enable cbsc-1 unc010012-00003 COMMAND ACCEPTED CLICMD-5179 96-03-16 16:58:03 rpmm1 MM-1 M010012.00003 000062/000062INFO:1 “Command Received and Accepted” COMMAND=”ENABLE cbsc-1 unc “

CBSC-1 96-03-16 16:58:03 rpmm1 MM-1 M010012.00003 000063/000063 INFO:3 “Command in Progress” STATUS=STARTED

CBSC-1 96-03-16 16:58:03 rpmm1 MM-1 M010012.00003 000064/000064 STCHG:1 “State Change Event” OLDSTATE=OOS_MANUAL PROCEDURE=INITIALIZE NEWSTATE=INS ACTION=START

CBSC-1 96-03-16 16:58:03 rpmm1 MM-1 M010012.00003 000065/000065 STCHG:1 “State Change Event” OLDSTATE=INS PROCEDURE=INITIALIZE NEWSTATE=OOS_AUTOMATIC ACTION=FINISH

CBSC-1 96-03-16 16:58:03 rpmm1 MM-1 M010012.00003 000066/000066 INFO:5 “Command Failed” REASON_CODE=”Primary FEPs Not INS”rpmm1-010012 > enable fepr-1-1-1 unc010012-00004 COMMAND ACCEPTED CLICMD-5180 96-03-16 16:58:13 rpmm1 MM-1 M010012.00004 000067/000067 INFO:1 “Command Received and Accepted” COMMAND=”ENABLE fepr-1-1-1 unc “

FEPR-1-1-1 96-03-16 16:58:13 rpmm1 MM-1 M010012.00004 000068/000068 INFO:3 “Command in Progress” STATUS=STARTED

FEPR-1-1-1 96-03-16 16:58:13 rpmm1 MM-1 M010012.00004 000069/000069 STCHG:1 “State Change Event” OLDSTATE=OOS_MANUAL PROCEDURE=NONE NEWSTATE=OOS_PARENT ACTION=NOCHANGE

FEPR-1-1-1 96-03-16 16:58:13 rpmm1 MM-1 M010012.00004 000070/000070 INFO:5 “Command Failed” REASON_CODE=”Parent Device In Improper State”rpmm1-010012 > status cbsc-1010012-00005 COMMAND ACCEPTED

CBSC-1 96-03-16 16:58:21 rpmm1 MM-1 D010012.00005 000071/000071 INFO:3 “Command in Progress” STATUS=STARTED

CBSC-1 96-03-16 16:58:21 rpmm1 MM-1 D010012.00005 000072/000072 INFO:2 “Device Status Response” TELSTATE=OOS_AUTOMATIC PROCEDURE=NONE PHYSTATE=NONErpmm1-010012 > load fep1 -1-1-2010012-00006 COMMAND ACCEPTED CLICMD-5182 96-03-16 16:58:31 rpmm1 MM-1 M010012.00006 000073/000073 INFO:1 “Command Received and Accepted” COMMAND=”LOAD fep-1-1-2 “

FEP-1-1-2 96-03-16 16:58:31 rpmm1 MM-1 M010012.00006 000074/000074 INFO:3 “Command in Progress” STATUS=QUEUED

FEP-1-1-2 96-03-16 16:58:31 rpmm1 MM-1 M010012.00006 000075/000075 PROC:1 “Procedural State Change Event” PROCEDURE=QUEUED_TO_DOWNLOAD TELSTATE=OOS_MANUAL ACTION=START

FEP-1-1-2 96-03-16 16:58:31 rpmm1 MM-1 M010012.00006 000076/000076 INFO:3 “Command in Progress” STATUS=STARTED

FEP-1-1-2 96-03-16 16:58:31 rpmm1 MM-1 M010012.00006 000077/000077 PROC:1 “Procedural State Change Event” PROCEDURE=QUEUED_TO_DOWNLOAD TELSTATE=OOS_MANUAL ACTION=FINISH

Sample Output: Enable CBSC



NOTES:

FEP-1-1-2 96-03-16 16:58:31 rpmm1 MM-1 M010012.00006 000078/000078 PROC:1 “Procedural State Change Event” PROCEDURE=DOWNLOAD ACTION=START TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:01:31 rpmm1 MM-1 M010012.00006 000079/000079 PROC:1 “Procedural State Change Event” PROCEDURE=DOWNLOAD ACTION=FINISH TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:01:31 rpmm1 MM-1 M010012.00006 000080/000080 INFO:4 “Command Successfully Completed” REASON_CODE=”No Reason”rpmm1-010012 > enable cbsc-1 unc010012-00007 COMMAND ACCEPTED

CLICMD-5200 96-03-16 17:03:00 rpmm1 MM-1 M010012.00007 000081/000081 INFO:1 “Command Received and Accepted” COMMAND=”ENABLE cbsc-1 unc “

CBSC-1 96-03-16 17:03:00 rpmm1 MM-1 M010012.00007 000082/000082 INFO:3 “Command in Progress” STATUS=STARTED

CBSC-1 96-03-16 17:03:00 rpmm1 MM-1 M010012.00007 000083/000083 STCHG:1 “State Change Event” OLDSTATE=OOS_AUTOMATIC PROCEDURE=INITIALIZE NEWSTATE=INS ACTION=START

FEP-1-1-2 96-03-16 17:03:00 rpmm1 MM-1 M010012.00007 000084/000084 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:03:00 rpmm1 MM-1 M010012.00007 000086/000086 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=CANCELLED TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:03:00 rpmm1 MM-1 M010012.00007 000087/000087 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=START TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:03:11 rpmm1 MM-1 M010012.00007 000090/000090 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=FINISH TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:03:11 rpmm1 MM-1 M010012.00007 000091/000091 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_AUTOMATIC

FEP-1-1-2 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000092/000092 STCHG:1 “State Change Event” OLDSTATE=OOS_AUTOMATIC PROCEDURE=NONE NEWSTATE=INS_ACTIVE ACTION=NOCHANGE

XCLINK-1-1-2-1 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000093/000093 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=INITIALIZE NEWSTATE=INS ACTION=FINISH

CPP-1-1-1 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000094/000094 PROC:1 “Procedural State Change Event”

Sample Output: Enable CBSC (Continued)



PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

CPP-1-1-1 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000096/000096 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=INITIALIZE NEWSTATE=INS ACTION=FINISH


FEP-1-1-2 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000098/000098 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS_ACTIVE

FEPR-1-1-1 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000099/000099 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

FEPR-1-1-1 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000102/000102 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=CANCELLED TELSTATE=OOS_PARENT

FEPR-1-1-1 96-03-16 17:03:12 rpmm1 MM-1 M010012.00007 000103/000103 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=START TELSTATE=OOS_PARENT

FEPR-1-1-1 96-03-16 17:03:23 rpmm1 MM-1 M010012.00007 000105/000105 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=FINISH TELSTATE=OOS_PARENT

FEPR-1-1-1 96-03-16 17:03:23 rpmm1 MM-1 M010012.00007 000106/000106 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

FEPR-1-1-1 96-03-16 17:03:24 rpmm1 MM-1 M010012.00007 000107/000107 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=NONE NEWSTATE=INS_STANDBY ACTION=NOCHANGE


FEPR-1-1-1 96-03-16 17:03:24 rpmm1 MM-1 M010012.00007 000109/000109 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS_STANDBY

MSC-1 96-03-16 17:03:30 rpmm1 MM-1 M010012.00007 000111/000111 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

C7LINK-1-1-1 96-03-16 17:03:30 rpmm1 MM-1 M010012.00007 000113/000113 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=INITIALIZE NEWSTATE=INS ACTION=FINISH

MSC-1 96-03-16 17:03:30 rpmm1 MM-1 M010012.00007 000114/000114 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=NONE NEWSTATE=INS ACTION=NOCHANGErpmm1-010005 > enable bts-52 unc010005-00014 COMMAND ACCEPTED CLICMD-2340 96-03-20 20:40:26 rpmm1 MM-1 M010005.00014 000231/000231 INFO:1 “Command Received and Accepted” COMMAND=”ENABLE bts-52 unc “

BTS-52 96-03-20 20:40:26 rpmm1 MM-1 M010005.00014 000232/000232 INFO:3 “Command in Progress” STATUS=STARTED

BTS-52 96-03-20 20:40:26 rpmm1 MM-1 M010005.00014 000233/000233 PROC:1 “Procedural State Change Event”

Sample Output: Enable CBSC (Continued)



NOTES:

PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_MANUAL

BTSLINK-52-1 96-03-20 20:40:26 rpmm1 MM-1 M010005.00014 000235/000235 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

BTSLINK-52-1 96-03-20 20:40:26 rpmm1 MM-1 M010005.00014 000236/000236 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=NONE NEWSTATE=INS ACTION=NOCHANGE

BTSLINK-52-1 96-03-20 20:40:26 rpmm1 MM-1 M010005.00014 000237/000237 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS

MGLI-52-1 96-03-20 20:40:41 rpmm1 MM-1 M010005.00014 000238/000238 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_MANUAL

MGLI-52-1 96-03-20 20:40:41 rpmm1 MM-1 M010005.00014 000240/000240 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=CANCELLED TELSTATE=OOS_MANUAL

MGLI-52-1 96-03-20 20:40:41 rpmm1 MM-1 M010005.00014 000241/000241 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=START TELSTATE=OOS_MANUAL

MGLI-52-1 96-03-20 20:40:51 rpmm1 MM-1 M010005.00014 000243/000243 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=FINISH TELSTATE=OOS_MANUAL

MGLI-52-1 96-03-20 20:40:51 rpmm1 MM-1 M010005.00014 000244/000244 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_MANUAL

MGLI-52-1 96-03-20 20:40:56 rpmm1 MM-1 M010005.00014 000245/000245 STCHG:1 “State Change Event” OLDSTATE=OOS_MANUAL PROCEDURE=NONE NEWSTATE=INS_ACTIVE ACTION=NOCHANGE

CSM-52-1 96-03-20 20:41:04 rpmm1 MM-1 M010005.00014 000246/000246 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

LCI-52-1 96-03-20 20:44:56 rpmm1 MM-1 M010005.00012 000250/000250 PROC:1 “Procedural State Change Event” PROCEDURE=DOWNLOAD ACTION=FINISH TELSTATE=OOS_AUTOMATIC

CSM-52-1 96-03-20 20:45:53 rpmm1 MM-1 M010005.00014 000251/000251 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=INITIALIZE NEWSTATE=INS_ACTIVE ACTION=START

CSM-52-1 96-03-20 20:45:53 rpmm1 MM-1 M010005.00014 000252/000252 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS_ACTIVE

LCI-52-1 96-03-20 20:45:54 rpmm1 MM-1 M010005.00014 000253/000253 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_AUTOMATIC

Sample Output: Enable CBSC (Continued), BTS Enable



LCI-52-1 96-03-20 20:46:04 rpmm1 MM-1 M010005.00014 000255/000255 STCHG:1 “State Change Event” OLDSTATE=OOS_AUTOMATIC PROCEDURE=NONE NEWSTATE=INS_ACTIVE ACTION=FINISH

LCI-52-1 96-03-20 20:46:04 rpmm1 MM-1 M010005.00014 000256/000256 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS_ACTIVE

BDC-52-1 96-03-20 20:46:04 rpmm1 MM-1 M010005.00014 000257/000257 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

BDC-52-1 96-03-20 20:46:04 rpmm1 MM-1 M010005.00014 000259/000259 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=CANCELLED TELSTATE=OOS_PARENT

BDC-52-1 96-03-20 20:46:04 rpmm1 MM-1 M010005.00014 000260/000260 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=START TELSTATE=OOS_PARENT

BDC-52-1 96-03-20 20:46:16 rpmm1 MM-1 M010005.00014 000261/000261 PROC:1 “Procedural State Change Event” PROCEDURE=DATALOAD ACTION=FINISH TELSTATE=OOS_PARENT

BDC-52-1 96-03-20 20:46:16 rpmm1 MM-1 M010005.00014 000262/000262 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

BDC-52-1 96-03-20 20:46:17 rpmm1 MM-1 M010005.00014 000263/000263 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=NONE NEWSTATE=INS_ACTIVE ACTION=NOCHANGE

BDC-52-1 96-03-20 20:46:17 rpmm1 MM-1 M010005.00014 000264/000264 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS_ACTIVE

BBX-52-1 96-03-20 20:46:17 rpmm1 MM-1 M010005.00014 000265/000265 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=START TELSTATE=OOS_PARENT

BBX-52-1 96-03-20 20:46:22 rpmm1 MM-1 M010005.00014 000267/000267 STCHG:1 “State Change Event” OLDSTATE=OOS_PARENT PROCEDURE=INITIALIZE NEWSTATE=OOS_AUTOMATIC ACTION=FINISH

MGLI-52-1 96-03-20 20:46:22 rpmm1 MM-1 M010005.00014 000268/000268 PROC:1 “Procedural State Change Event” PROCEDURE=INITIALIZE ACTION=FINISH TELSTATE=INS_ACTIVE

BTS-52 96-03-20 20:46:22 rpmm1 MM-1 M010005.00014 000269/000269 STCHG:1 “State Change Event” OLDSTATE=OOS_MANUAL PROCEDURE=INITIALIZE NEWSTATE=INS ACTION=FINISH

BTS-52 96-03-20 20:46:22 rpmm1 MM-1 M010005.00014 000270/000270 INFO:4 “Command Successfully Completed” REASON_CODE=”No Reason”rpmm1-010005 > rpmm1-010005 >

Sample Output: BTS Enable



NOTES:

Page intentionally left blank (use for notes).Sample Output: BTS Enable




PWID Training: CDMACDMA Call Processing


Objectives:

• To present CDMA call processing in detail, from theperspective of actual message flow through the variousdevices in the network.


Table of Contents:

Call Processing Functionality Split ............................................................................. 211Observing CDMA Call Processing at the MM ............................................................ 213Call Processing Flows- General................................................................................... 215Call Flow: Mobile Termination.....................................................................................223Call Flow: Mobile Registration.....................................................................................226Hand-Offs in CDMA .................................................................................................... 229Call Flow: Soft Hand-Off Add Success ........................................................................230Looking at SCAP and A+ Messages ............................................................................ 235

PWID Center for Excellence Revision 0, 4/24/97 CDMA Call Processing


Call Processing Functionality Split

• Functions in “Real Time” category left at BTS.

• “Background” functions at CBSC.

• High volume transactions were left at cell site due tobandwidth demands.

• Review of interfaces used:- SCAP/LAP-D over DS0 for control of BTS.

- STRAU over DS0 for traffic channels of BTS.

- A+/SS7 over V.35 or DS0 for MSC, CBSC control.

- PCM over DS0 for traffic channels to/from MSC.

MSC

MS

CBSC

BTS

PSTN InterfaceHandoff ArbitrationFeature ControlTone InsertionNumber TranslationBillingSubscriber Data BaseTrunk Management

Resource Allocation (TCH)Dynamic Equipment SharingHandoff ControlCall ModelAccess ThresholdPriority Access

Signal Quality MeasurementsLayer 2 ProtocolsPower Change ManagementHandoff MeasurementsC/I & MRI ReportingMAHO integration

SCAP/LAP-D

IS-95 Air Interface Spec.

A+/SS7

E1- 29 DS0s x 4 tch/DS0 = 116 tch/E1

E1- 30 PCM



NOTES:

Call Processing Functionality Split

• Supercell takes much functionality away from the MSC, as compared to thefunctionality assigned to the MSC in typical analog systems. The CGM (Cell GroupManager) on an EMX acting as an MSC no longer performs any work. (However, theEMX will not initialize properly if the CGM is not in place, so the EMX still has theCGM present).

• The MSC holds the subscriber database, performs hand-off arbitration on inter-cbscand inter-system hand-offs (hard hand-offs), provides the interface to the PSTN (bothtraffic and signaling trunks), performs number translations, and is the centralrepository for billing information.

• The CBSC is responsible for allocating resources in the system which are needed forperforming call processing. These resources include MCCCE (Channel elements in anMCC), and connections through the XC’s KSW card for timeslots coming fromMCCCEs to the proper XCDR card which will have a TERCKT assigned to it. The MMhandles hand-offs for BTSs that it controls, and it acts as the controlling agent for hardhand-offs which originate in its BTSs. In addition, it implements call processing featuressuch as Access Threshold and Priority Access. The MM also handles dynamicequipment sharing- allocating free devices from a pool of available resources.

• The BTS handles the more time-critical functions which could not be handled by theCBSC due to delays in transmission and processing at the CBSC. These functionsinclude power control, hand-off measurements, and signal quality measurements,among others.

• Reviewing the interfaces:

• IS-95 Air Interface is used over an RF carrier from the BTS.• SCAP (Supercell Application Protocol) is used over a LAP-D transport for control

information at 64 Kbps to the BTS.• E1s are used (in the International Market) for transport of traffic channels to the

BTS. Each E1 has 30 channels at 64 Kbps each, out of which 1 is used for the LAP-D control link. The remaining 29 channels can each carry 4 traffic channelsencoded with CDMA STRAU frames, giving 116 channels per E1 to a BTS.

• A+ is used over SS7 for transport of control data between MSC and MM. Usuallythis link will have physical resources dedicated solely for it (V.35 cables, when MMand MSC are co-located). In that case, each E1 between the XC and the MSC cancarry 30 channels of PCM-encoded data.



Observing CDMA CP at the MM

• Utility called “scapdbg” used to look at message flowon MM.

• Utility presents ladder diagram of message flows,based on information from DBG output files.

• Getting scapdbg: http://scwww.cig.mot.com/SC/test/cdma_foa/public_html

• Turning on DBG statements for SCAP for callproc1and clcp: dbapi mDbg command.

• Looking more closely at SCAP messages: The SCAPdefinition document.

OMC-R

MM

scapdbg: ---------------------------- 96 Mar 22 -------------------------------------------- CDMA Ch Required -->03:04:08.243 BTS: 52 ESN: 9f038ca4 Dialed: 992-0071 --- Complete L3 Info --> 0x003903:04:08.248 <-- Connect Confirmed --- 0x003903:04:08.345 --- Setup --> 0x003903:04:08.347 <-- Call Proceeding --- 0x003903:04:08.416 <-- Assignment Request --- 0x003903:04:08.479



NOTES:

Observing CDMA Call Processing at the MM

• CDMA call processing may be observed at the MM through the use of a tool called“scapdbg”. This tool is not a comprehensive call analysis tool, but will suffice in givingthe operator a general idea about call processing & device O&M activity. The tool wascreated by members of the Arlington Heights FOA group, along with some members ofthe development comunity.

• This tool is generally used when trying to debug problems with call processing when thesystem is first being tested. Once a system is commercial, the call processing problems inthe system tend to be due to bad coverage in certain areas, or due to high traffic volume.These problems are analyzed using another set of tools (more on this later).

• The tool presents a ladder diagram of messages going between the MM, XC, and MSCdevices. The tool may be obtained through the Arlington Heights’ FOA group homepage on the WWW at: http://scwww.cig.mot.com/SC/test/cdma_foa/public_html.

• In order for the scapdbg program to present the ladder diagram, the DBG attributes forthe processes to be studied must be set to “ScapTrace 3” by performing the followingcommand from inside the dbapi debugger:

dbapi> mDbg callproc1 ScapTrace 3

Typically, the callproc1 and clcp processes will be modified as in the above commandwhen an operator wants to trace call processing; while sc_devom, sc_vdt1 and sc_vdt2will be modified when the operator wants to trace code downloads and device operations& maintenance.

• scapdbg usage:scapdbg v11.3Usage: scapdbg -cesxlhvdtTEB <ESN in hex> <BTS in decimal> -c Show detailed cause -e Show errors -s Show ESN, dialed digits and Markov -h Show handoff info -x Show XC connections -l Separate calls -b Ring bell between calls (off by default) -v Verbose (shows all the above) -d Use when decoding sc_devom or vdt -t Suppress Trouble Notifications -T Show only Trouble Notifications -E Only show calls with a particular ESN -B Only show sc_devom messages for a particular BTS

• The SCAP definition document (available through the CIG library) is the referencedocument for SCAP messages of all types.



Call Processing Flows- General

• Call Flows illustrate messages through system.

• Successful Mobile Origination.

• Successful Mobile Termination.

• Successful Mobile Registration.

• Errors can occur at any step in a flow.

• Timers are set up so that expected responses mustarrive in a given time frame.

MMXC

BTS

MSC

BTS BTS

CBSC

MS



NOTES:

Call Processing Flows- General

• The next few pages illustrate the flow of messages which go through the CDMA networkelements in order to perform different types of CDMA functions. Although this materialis somewhat tedious, it is important to follow the flow in order to get an idea as to whathappens during phone calls.

• An example mobile origination, as output from the scapdbg tool is as follows:---------------------------- 96 Mar 22 -------------------------------------------- CDMA Ch Required --> 12:15:54.825 BTS: 52 ESN: 9f03733f Dialed: 001-305-999-9999 --- Complete L3 Info --> 0x019e 12:15:54.829 <-- Connect Confirmed --- 0x019e 12:15:54.928 --- Setup --> 0x019e 12:15:54.929 <-- Call Proceeding --- 0x019e 12:15:55.000 <-- Assignment Request --- 0x019e 12:15:55.060<-- CDMA XC Ch Assigned --- 0x19f 12:15:55.062 CIC=78 MSC Span=2 TS=14 MSI=2 Span=A TS=2 Grp=2 CPP=1 BTS=52 MCC=1 CEP=1 Channel=1 PN=110 WC=18 ESN=9f03733f--- CDMA XC Assmnt Success--> 0x19f 12:15:56.565 --- Assignment Complete --> 0x019e 12:15:56.566--- CDMA HO Recognized --> 0x19c 12:15:57.665 HOCause: Soft HO Add 1 - TAdd Event UplinkFER: 0 ActvPNs [Ec/Io] [Phase]: 110 [-5.0dB] [0] ttdrop timer not expired CandPNs [Ec/Io] [Phase]: 310 [-12.0dB] [19840]<-- CDMA XC HO Direction --- 0x19c 12:15:57.668 HOCause: Softer HO - Add 1 SrchWinN: 8 (60 chips) TAdd: 24 (-12.0dB) SrchWinA: 5 (20 chips) TDrop: 26 (-13.0dB) SrchWinR: 9 (80 chips) TTdrop: 4 (6 sec) NghbrMaxAge: 3 TComp: 4 (2.0dB) FwdPwrThresh: 2 (erasures) PN: 310 WC: 8 BTS=52 MCC=2 CEP=1 CHNL=1--- CDMA HO Successful --> 0x19c 12:15:58.525<-- CDMA Update Parameters--- 0x19c 12:15:58.527 Neighbor list: 10 20 210 30 40 410 60 160 No. of neighbors: 8

< LOTS of Hand-Offs (deleted for space constraints) >

--- CDMA XC Radio Ch Rlsd --> 0x19c 12:17:04.304 CDMA call processing cause: normal release --- Release --> 0x019b 12:17:04.306 <-- Release Complete --- 0x019b 12:17:04.385 <-- Clear Cmd --- 0x019b 12:17:04.389 --- Clear Comp --> 0x019b 12:17:04.391<-- CDMA Release Radio Ch --- 0x19c 12:17:04.393 BTS=52 MCC=2 CEP=1 Ch=1 CDMA call processing cause: normal release <-- Release --- 0x019c 12:17:04.395<-- CDMA XC Rls Radio Ch --- 0x19d 12:17:04.397 CDMA call processing cause: normal release <-- SCCP Disconnect Req --- 0x019b 12:17:04.464

CDMA Call Processing

Motorola Confidential Proprietary

Revision 0, 5/23/96PWID Center for Excellence

216 of 259

Call Flow: Mobile OriginationMSC MM FEP GPROC XCDR KSW GLI PG-ACC TCH MS

[1] Origination Message [MSID(MIN,ESN),AUTH_MODE=0, SLOT_CYCLE_INDEX,SCM, service option 1, 2, 8002, or 8003 hex, dialed digits]

[2] Base Station Ack Order

[4] CHI: Channel Required [Location area, Antenna Pair, Origination message][5] LAPD: Channel Required

[6] LAN: Channel Required

[10] SS7: SCCP Connection Request + Complete Layer 3 Information Message (CM Service Request) [Cell Identifier,CM Service Type, Ciphering Key Sequence Number=111, Classmark Information Type 2,Mobile Identity (MID), Mobile Identity (ESN)] [If SCM in IS-95 Origination message indicates mobile is slotted mode, the Slot Cycle Index element is included in the CM Service Request.]

[12] SS7: SCCP Connection Confirm

[14] SS7: Setup

[Called Party BCD Number]

[16] SS7: Call Proceeding

[19] SS7: Assignment Request

[Channel Type, Layer 3 Header Info, Circuit Identity Code][20] TC marked busy; MCC TCH and WC allocated

[21] LAN: XC Channel Assigned [CDMA TCH information,encryption mode=0,frame offset=0,antenna pair, MIN,public long code mask,Circuit Identity Code, source XCVR routing info(TCH MCC),source XCVR routing info(PG MCC), ESN,CODEC mode,traffic channel code,call mode,service option]




217 of 259

Call Flow: Mobile Origination (Continued)MSC MM FEP GPROC XCDR KSW GLI PG-ACC TCH MS

[22] LAPD: XC Channel Assigned

****************************************************************** XCDR is operating in default service option mode - no dim and burst operation - PCM silent tone generated toward MSC - IS-95 Null TCH Data generated toward MCC The XCDR and MCC are not yet connected thru the KSW

*****************************************************************

[24] MCAP: Traffic Channel BTS Link Request [XCDR Channel,CL2=Clear Layer 2,B0=1]

[25] MCAP: Port connection instructions

[26] PCM silent tone[27] STRAU Speech frames containing IS-95 Null TCH data XCDR generated Null TCH data

[28] STRAU Idle frames[29] PCM silent tone

Generated by XCDR[30] MCAP: Traffic Channel State Change [XCDR channel,New_State=MCC_idle, Invalid_Speech, or Valid_Speech] **************************************************** XCDR and TCH MCC have acheived STRAU frame synchronization ****************************************************

[32] LAPD: Channel Assigned

[CDMA TCH information,encryption mode=0,frame offset=0,antenna pair, MIN,public long code mask,source XCVR routing info,ESN,call reference number]

[33] LAPD: Channel Assigned

[34] CHI: Channel Assigned[36] IS-95 frames




218 of 259


[37] CHI: Target Channel Designation [CDMA TCH information,encryption mode=0, frame offset=0,MIN]

[38] CHI: Target Channel Designation

[40] Channel Assignment Message [ADDR_TYPE=0,ADDRESS(MIN), ASSIGN_MODE=0, FREQ_INCL=1,CODE_CHAN, CDMA_FREQ,FRAME_OFFSET=0, ENCRYPT_MODE=0]

Upon detection of two consec. good frames on Fwd. TCH, MS starts

to tx TCH Preamble.

[43] IS-95 frames

TCH Preamble

MCC is in default service option mode: - IS-95 traffic from MS formatted as QCELP and sent to xcdr in STRAU Speech frames - STRAU traffic from XCDR formatted as QCELP and is expected in STRAU Speech frames

***********************

***********************[44] STRAU Speech frames

Full-rate, inner & outer CRC indicators not equal to 00[46] PCM silent tone generated in XCDR

[47] CHI: Suspend TCH Assignment [MIN] ******************************* TCH MCC has acquired the MS *******************************




219 of 259


[48] CHI: Suspend TCH Assignment[50] MCAP: Traffic Channel State Change [XCDR channel,New_State=Invalid Speech] *********************************************** XCDR has detected TCH acquisition of the MS ***********************************************[52] MCAP: Layer 3 Message Request

[XCDR Channel,Magic Number,ackr=1,Base Station Acknowledgment Order]

[53] Base Station Acknowledgement Order [ack_req]

[54] Mobile Station Ack Order[55] IS-95 frames Null TCH Data[56] STRAU Speech frames

Eighth-rate, inner & outer CRC = 00[57] PCM silent tone generated by XCDR

[58] MCAP: Layer 3 Message Confirm [XCDR Channel, Magic Number]

[59] MCAP: Traffic Channel State Change [XCDR channel,New_State=Valid Speech ]

[61] MCAP: Layer 3 Message Request [XCDR Channel,Magic Number,ackr=1,Service Option Response Order] [ack_req,service option 1, 8002 or 8003 hex]

[62] Service Option Response Order

[63] Mobile Station Ack Order




220 of 259


[64] IS-95 frames

QCELP, if SO=1 Markov data, if SO=8002 hex Data, if SO=8003 hex

[65] MCAP: Layer 3 Message Confirm [XCDR Channel, Magic Number]

[66] MCAP: Service Option Change Request [Service Option (1-QCELP, 8002 hex-markov, 4-m-m data trial), Codec Mode]

The XCDR will start operating in the requested service option mode. SO 1: - Valid STRAU Speech frames from MCC will be transcoded into PCM and sent to the MSC - Traffic from the MSC will be transcoded from PCM into QCELP in STRAU speech frames - Dim and Burst operation will be used for downlink signalling SO 8002 hex: - Variable rate Markov data will be originated in the transcoder and sent to the MCC in STRAU Markov frames - Data received from the MCC in STRAU Markov frames will be treated as received variable rate Markov data - PCM silent tone or equivelent will be sent toward the MSC - Received PCM from the MSC will be discarded SO 8003 hex: - Synchronization with remote transcoder will be attempted (and will not be successful until MSC

connects local and remote circuits) - Valid data traffic from the MCC will be sent to remote transcoder - Framed traffic from remote transcoder will be sent to MCC in STRAU Data frames - If no framed data traffic from remote transcoder, send IS-95 Null TCH Data to MCC in STRAU Data frames

[67] STRAU frames If SO = 1: STRAU Speech frames containing PCM->QCELP transcoded silent tone If SO = 8002: STRAU Markov frames containing XCDR generated Markov data If SO = 8003: STRAU Data frames containing XCDR generated IS-95 Null TCH Data




221 of 259


[68] STRAU frames If still receiving STRAU Speech frames downlink, continue STRAU Speech frames containing QCELP uplinkIf STRAU Data frames received downlink, send STRAU Data frames containing IS-95 traffic uplink If STRAU Markov frames received downlink, send STRAU Markov frames containing IS-95 traffic uplink

[69] LAPD: XC Assignment Successful

[call reference number]

[70] LAN: XC Assignment Successful

[71] SS7: Assignment Complete

[Channel Number]

[73] SS7: Alerting

[74] PCM ringback tone




222 of 259

Call Flow: Mobile Origination (Last Slide)MSC MM FEP GPROC XCDR KSW GLI PG-ACC TCH MS

[75] STRAU Frames If SO=1: STRAU Speech frames containing PCM->QCELP transcoded ringback tone If SO=8002: STRAU Markov frames containing XCDR generated markov data If SO=8003: STRAU Data frames containing XCDR generated IS-95 Null TCH Data[77] SS7: Connect

******************** Called has Party Answered If SO=1, a PCM connection is now established between the called and calling parties If SO=8002, PCM received from the MSC is discarded by the XCDR, the XCDR sends PCM silent tone or equivelent to the MSC, the XCDR transmits markov data in STRAU Markov frames to the MCC, the XCDR receives STRAU Markov frames from the MCC If SO=8003, the local and remote transcoders achieve frame synchronization, and a data connection is now established between the called and calling parties********************

[78] Through connected

If SO=1: PCM (voice) If SO=8002 hex: dont care If SO=8003 hex: rate adapted data frames

[79] STRAU Frames

If SO=1: STRAU Speech frames containing PCM->QCELP transcoded voice If SO=8002 hex: STRAU Markov frames containing XCDR generated markov data If SO=8003 hex: STRAU Data frames containing transcoder bypassed data

[80] MCAP: Synched to remote xcdr circuit

SO 8003 hex only: Indicates to GPROC that frames from the remote transcoder are being received by the local transcoder from the MSC direction.

[81] SS7: Connect Acknowledge




223 of 259

Call Flow: Mobile TerminationMSC MM FEP GPROC XCDR KSW GLI TCH MS

[1] Release Order [normal release or power down indication]

[2] IS-95 frames Null TCH Data[3] STRAU Speech frames

Eighth-rate, inner & outer CRC=0[4] PCM

[5] MCAP: Layer 3 Message Indication [XCDR Channel,Release Order message]

[6] Use fast repeat to send Release to MS

[8] MCAP: Layer 3 Message Request

[XCDR Channel,Magic Number,Fast_repeat_flag=use fast rep., ACKR=no ack req.,Release Order message] [no reason given]

[9] Release Order [no reason given] [10] MS stops tx

[11] TCH loses MS[12] MCAP: Layer 3 Message Confirm [XCDR Channel,Magic Number] [Note that in fast repeat case, L3 Msg Confirm and State Change Indication may reverse order.]

[13] STRAU Traffic frames (bad quality) ********************************************************** XCDR detects loss of the MS on the Reverse Traffic Channel **********************************************************

[14] MCAP: Traffic Channel State Change Indication

[XCDR Channel,New_State= bad quality traffic frames][16] MCAP: Traffic Channel BTS Link Request

[XCDR Channel,CL2=clear layer 2,B0-B2=inactive link][17] XCDR is now in service optionless state




224 of 259

Call Flow: Mobile Termination (Continued)MSC MM FEP GPROC XCDR KSW GLI TCH MS

[18] STRAU Speech frames containing IS-95 Null Traffic Channel Data

[19] PCM Silent tone or equivalent

[20] PCM

[21] MCAP: Port connection instructions

[22] All TCH MCCs in call are disconnected from XCDR

[23] STRAU Speech frames containing IS-95 Null Traffic Channel Data

[24] KSW idle pattern

[25] KSW idle pattern on BTS links into XCDR[26] IS-95 frames

Null TCH Data[27] LAPD: XC Radio Channel Released

[cause=normal release]

[28] LAN: XC Radio Channel Released[30] SS7: Release

[cause layer 3=normal clearing]

[31] SS7: Release Complete

[35] SS7: Clear Command

[Layer 3 Header Information, cause=call control]

[37] SS7: Clear Complete

[39] Idle TC




225 of 259

Call Flow: Mobile Termination (Last Slide)MSC MM FEP GPROC XCDR KSW GLI TCH MS

[41] Release all TCHs associated with call

[42] LAN: Release Radio Channel

[detailed cause]

[43] SS7: SCCP RLSD

[44] SS7: SCCP RLC

[45] LAPD: Release Radio Channel

[46] CHI: Release Radio Channel

[47] TCH stops dl tx

[48] CHI: Radio Channel Released

[49] LAPD: Radio Channel Released

[50] LAN: Radio Channel Released

[51] Stop MmCpT3

[52] Deallocate TCH MCC and associated Walsh Code




226 of 259

Call Flow: Mobile RegistrationMSC MM FEP GLI PG-ACC MS

[1] Registration

[ACK_REQ, MSID, REG_TYPE, SLOT_CYCLE_INDEX, SCM]

[2] BTS Ack Order

[ACK_SEQ,VALID_ACK] This message is sent using quick repeats.

[3] CHI: Location Registration Request

[ MIN, Location Area, ESN Registration Type, SCM, Slot Cycle Index ]

[4] LAPD: Location Registration Request

[5] LAN: Location Registration Request

[7] SS7: SCCP Connection Request + Complete Layer 3 Information (Location Updating Request)

[Location Area Identification, Mobile Id (MID), Mobile Id (ESN)] [If the SCM indicates that the mobile is slotted mode, the Slot Cycle Index element is included]




227 of 259

Call Flow: Mobile Registration (Last Slide)MSC MM FEP GLI PG-ACC MS

[8] SS7 SCCP Connection Refused + Location Updating Accept

[Location Area Identification]

[10] LAN: Location Registration Accepted

[MIN] [If the MS is operating in slotted mode, the Slot Cycle Index element is included]

[11] LAPD: Location Registration Accepted

[12] CHI: Location Registration Accepted

[13] Registration Accepted Order

Sent using quick repeats. If L2 info for the MS id still exists, include an ack to the Registration message in the Registration Accepted Order.



Hand-Offs in CDMA

• Mobile Assisted Hand-Offs (MAHO) used inCDMA.

• TAdd, TComp, TDrop, and TTDrop- four of severalparameters governing HOs.

• The XC, MM, and MS are involved in HOs.

• XC detects need to HO indications sent by MS,selects the best 20ms frames from two or more BTSs.

• MM determines resources to use for the HO, sendsinstructions to XC and MSC to perform HO.

MMXC

BTS

MSC

BTS BTS

CBSC

MS



NOTES:

Hand-Offs in CDMA

• HO detection processing takes place in the MS, the XC, and the MM. The MS hasparameters which define which Pilots will be in the Active Set, and in the Candidate Set(the two most important sets). Some of these parameters are:

• TAdd- Threshold above which a pilot’s signal strength must rise in order for theMS to transmit a pilot strength measurement message. The pilot strength measure-ment message is the mobile’s response to the infrastructure indicating that the MShas seen a strong Pilot.

• TComp- Threshold above which a candidate set pilot strength must rise above anactive set pilot to cause the MS to transmit a pilot strength measurement message.

• TDrop- Pilot signal strength threshold below which a pilot will be dropped from itslist (either Active, or Candidate list). Pilot must stay under the threshold forTTDrop time or more in order to be dropped from its list.

• The XC will detect the need to handoff (by detecting the pilot strength measurementfrom the MS), perform handoff preprocessing and identify hand-off events. The XC alsoselects the best 20ms frame of STRAU data to transcode, from the 2 or 3 availableframes.

• The MM will determine the handoff type, perform target selection and perform channelallocation (Walsh codes, available MCCCE, XCDR circuits, etc.).

• The following is a section of the output from the scapdbg script for a hand-off:--- CDMA HO Recognized --> 0x19c 12:15:57.665 HOCause: Soft HO Add 1 - TAdd Event UplinkFER: 0 ActvPNs [Ec/Io] [Phase]: 110 [-5.0dB] [0] ttdrop timer not expired CandPNs [Ec/Io] [Phase]: 310 [-12.0dB] [19840]<-- CDMA XC HO Direction --- 0x19c 12:15:57.668 HOCause: Softer HO - Add 1 SrchWinN: 8 (60 chips) TAdd: 24 (-12.0dB) SrchWinA: 5 (20 chips) TDrop: 26 (-13.0dB) SrchWinR: 9 (80 chips) TTdrop: 4 (6 sec) NghbrMaxAge: 3 TComp: 4 (2.0dB) FwdPwrThresh: 2 (erasures) PN: 310 WC: 8 BTS=52 MCC=2 CEP=1 CHNL=1--- CDMA HO Successful --> 0x19c 12:15:58.525<-- CDMA Update Parameters--- 0x19c 12:15:58.527 Neighbor list: 10 20 210 30 40 410 60 160 No. of neighbors: 8

• The following pages illustrate message flow through the CDMA system under differenthand-off situations.




230 of 259

Call Flow: Soft Hand-Off Add SuccessMM XC KSW XCDR ADD_TCH ALL_TCH MS

[1] RF: Pilot Strength Measurement Message [Pilot strengths, Pilot phases, reference pilot strength & phase]

[2] SCAP: CDMA Handoff Recognized [Cause field = soft handoff add, XC starts timer XcHoT7.]

[3] Channel Hunt Process

[MM recognizes "add", finds available TCH in target cell]

[4] SCAP: CDMA XC Handoff Direction

[Soft handoff "add" decision, new set of active pilots, etc.]

[5] MCAP: KSW Connection Request

[Route downlink QCELP to new target, time align uplink.]

[6] MCAP: CDMA TCH BTS Link Request [Connect XCDR to new TCH.]




231 of 259

Call Flow: Soft Hand-Off Add Success (continued)MM XC KSW XCDR ADD_TCH ALL_TCH MS

[7] SCAP: CDMA Handoff Channel Assigned

[The target MCC programs its searchers, setpoint, forward power, and "keys" PA.]

[8] SCAP: CDMA Forward Channel Transmission Indication

[Target MCC is now simulcasting on downlink.]

[9] RF: Extended Handoff Direction Message

[HDM_SEQ, search windows, T_ADD, T_DROP, T_TDROP, T_COMP.] [Pilots and Walsh codes of active TCHs.] [Start timer XcHoT1.]

[10] RF: Handoff Completion Message

[Pilots involved in active set and HDM_SEQ.] [XC clears timers XcHoT7 and XcHoT1.]

[11] RF: Base Station Acknowledgment Order

[12] SCAP: CDMA Handoff State Change [Number of forward links and receivers assigned to the call]

[13] SCAP: CDMA Handoff Successful

[MM clears timer]




232 of 259

Call Flow: Soft Hand-Off Add Success (Last Slide)MM XC KSW XCDR ADD_TCH ALL_TCH MS

[14] SCAP: CDMA Update Parameters Message

[15] RF: In-Traffic System Parameters Message

[16] RF: Mobile Station Acknowledgment Order

[17] RF: Neighbor List Update Message


[19] RF: Power Control Parameters Message


The following messages are sent as required




233 of 259

Page intentionally left blank.



Looking at SCAP & A+ Messages

• SCAP document provides full definition of all mes-sages for control of the BTSs and the XC.

• A+ document provides full definition of all A+ mes-sages for communication to the MSC.

• DBG output files will contain all the messages asseen above when “Trace 3” is turned on.

MXC

MSC

BTS BTS

CBSC

MS

------------HEADER(20 bytes)------Header Version: 0x0

Data Version: 0x0Hop Count: 0x2Channel ID: 0x1Data Length: 0x48Message Type: 0x12Dest Address: sys(0x80) subsys(0xffff)

grp(0xff) dev(0xff) elem(0xff)Response Type: 0x0Source Address: sys(0x81)

subsys(0x0001) grp(0x80) dev(0x42)elem(0x22)

-------------BODY(bytes)----------0x00 0x22 0x00 0x00 0x00 0x0a 0x42 0xc60xd5 0x80 0x00 0x0b 0x39 0x00 0x00 0x000x00 0x20 0x00 0x00 0x00 0x01 0x01 0x000x00 0x17 0x00 0x34 0x00 0x18 0x00 0x1c0x00 0x0f 0x1a 0x00 0x00 0x24 0x06 0x010x9e 0x00 0x00 0x10 0x06 0x01 0x01 0x000x00 0x21 0x18 0x13 0x01 0x08 0x01 0x050x04 0x03 0x80 0x87 0xaa 0x6c 0x06 0x800x84 0xa7 0xaa 0xaa 0x0a 0x70 0x01 0x00

file: ../AplusComm.Cfunc: ServiceInbound()AplusMsg is:---------SCCP HEADER(6 bytes)--------Message Type: 0x1201Ref Number: 0x00000000------A+ BSSMAP HEADER(2 bytes)------Message Discrimination: 0x00Length Indicator: 0x07---------BODY(7 bytes)---------------

---------Message Type: 0x3401 00 37 04 01 07



NOTES:

Looking at SCAP and A+ Messages

At times it could be necessary to go down to the bits-and-bytes level in order todebug a particular problem, or just to get more information on what’s going onwith the CDMA network elements.

• The messages must be captured first, by using the dbapi debugger and turning on theDBG statements for “Trace” at level 3 for processes to be studied. In general, turning on“Trace” for callproc1, clcp, cpmo, and sc_devom will give most messages of interest.

• The SCAP ICD document defines all the messages and message formats for all messagesgoing over the BTSLINKs, and the XCLINKs. This document can be found in theArlington Heights Supercell Library, which has an on-line copy in the AH Sun networkat /usr/test/scdocrel/COM/GEN/ICD/003.

• The A+ ICD document defines all the messages and formats for messages going over theA+ links from the MM to the MSC. This document can also be found in the ArlingtonHeights Supercell Library (on-line path is not available at time of writing, but readercan try the Arlington Heights Web page: http://www.sc.cig.mot.com/SC/library/index.html for more information.


Conclusions:• Call Processing in CDMA is carried out at all points in

the network. However, the MM is the “work horse”.

• Call Processing message flow may be observed at theMM by using the “scapdbg” program.

• DBG statements must be turned on for “ScapTrace”at level 3 for the above program to work.

• Call Processing flows help understand what is goingon in the system when a call is placed.

• Hand-Offs are considered part of call processing.

• Hand-Offs are handled mostly by the MM, althoughthe MS, XC are always involved. The MSC is involvedfor Hard Hand-Offs.

• Several parameters govern the behavior of Hand-Offs.These parameters are editable.

• The SCAP and A+ ICD documents are the authoritieson messages going over the BTSLINKs, XCLINKs,and over the A+ link.


PWID Training: CDMATricks, Tips & Tools for Unix and CDMA


Objectives:

• To present various tricks & tips on general topicsrelated to Supercell and Unix.

• To present a brief overview of the tools available forCDMA data analysis (statistics and drive test data).


Table of Contents:

On-Site Access to Outside World.................................................................................. 241Setting up a Dial-In Modem......................................................................................... 243Ethernet Configuration ................................................................................................ 245Other Unix “Tricks”: Find, dps, high priority session................................................ 247More Tricks & Tips: ftp, telnet, anonymous ftp, etc.................................................... 249Tricks & Tips: tar, compression utilities, framemaker batch files .............................. 251Tools for Analysis of CDMA Data ............................................................................... 253

PWID Center for Excellence Revision 0, 4/24/97 Tricks, Tips & Tools for Unix and CDMA


On-Site Access to Outside World

• For accessing the external world, hardware needed:- Portable PC with an ethernet card (preferably

PCMCIA- very convenient).

- PC Modem (at least 14.4 baud).

- Direct-connect telephone lines to the site.

- Activated SecureID card (at least one).

• Software needed (two options):- Obtain Linux for PCs (vendor, the Internet, or

through Motorola WWW), or-

- Obtain Connection Kit (through MotorolaWWW- soon) for Windows.

• Second option will be easier, but first option allowsUnix OS on your portable PC!



NOTES:

On-Site Access to Outside World

It is very useful that the CBSC site be equipped to communicate with the outsideworld, specifically with the Motorola internal network. An operator can neverguess exactly when a particular file will be necessary, or some information whichis on the Motorola WWW is crucial to the proper functioning of a new release.Access to email, phone lists, etc. will also be provided with a link to Motorola.The following are points are given on establishing a link.

• Hardware requirements: A portable PC with an Ethernet PCMCIA card and a modem.A combo card (28.8 Modem, Ethernet 10BaseT and 10Base2) PCMCIA Ethernet cardsuch as 3Com’s Etherlink III Combo card is recommended.

• In addition, direct-connect lines to the outside world for exclusive use by modems shouldbe installed at the CBSC site. This usually must be arranged with the customer, since thecustomer will <usually> have another MSC at the site (assuming that most customersinstall CDMA in the same physical facilities as their existing analog or digital telephonyequipment).

• At least one activated SecureID card is essential in order to access the Motorolanetwork. Contact ITS at the CIG help desk ([email protected]) in order to getinformation on activating a SecureID card, or contact your local support person forinformation on how to obtain a new card.

• There are two approaches to obtaining the software necessary for the portable PC toaccess the Motorola network- one is through the Linux OS (which is a version of Unixavailable on the internet for free), or through a “Connection Kit” which will allow PCswith Windows to login to the Motorola network and use such utilities as ftp, telnet, andNetscape.

• Linux is more complicated to install, but at the same time it provides a full version ofUnix which the operator might find interesting to “play around” (hack) with. Linuxincludes PPP (Point-to-Point protocol), which is necessary to create a link with Motorolathrough a dial-up modem. Linux also has ftp, telnet, and Netscape applicationsavailable. Linux is available through the WWW (search for “Linux Distributions”), or itmay be purchased on CDROM through Walnut Creek CDROM(1-800-786-9907). Abare-bones version of Linux is available on the Motorola WWW at http:

• The “Connection Kit” will be available on the Motorola WWW soon (look fornotification on the “cdmaprod” mailing list).



Setting up a Dial-In Modem

• Imperative at times to allow outside access of OMC-R (and thus, MM and XC as well) to qualifiedMotorola personnel.

• Access must be restricted!!!!

• Additional hardware needed: at least 9600 baudmodem (1) with RS-232 connection. One RJ-48 toRS-232 cable.

• Procedure Overview:- Connect modem to power supply, telco, async

card.

- Add a port (software command for port monitor).

- Add a service to the port (software command forport monitor).

- Dial in, and get a prompt!

MTSORemote Support Site



NOTES:

Setting up a Dial-In Modem

At times it is critical for outside support personnel to be able to login to theOMC-R in order to debug problems with Supercell equipment such as theOMC-R, MM, or XC. The following are notes and procedures for setting upaccess for the outside world.

• First of all, it is critical that this be done only when absolutely necessary, and that accessto the system be monitored such that only authorized personnel have access.

• In addition, the account through which the remote personnel will login must beprotected by a good password, and the modem should actually be turned off when not inuse in order to avoid un-authorized access.

• The hardware needed:

• One modem with RS-232 port (at least 9600 baud). The modems used in the SC9600BTSs may be used on the OMC-R, provided there is a +27 VDC power supply. Mostof the time a BTS will be co-located with the OMCR, so the BTS modem could beused temporarily. The modem must be connected to a telephone line with directaccess to the outside world (PSTN).

• One RJ-47 to RS-232 cable. The cable used for connecting the OMC-R console ter-minal to the IOP card may be used (it’s better to use a spare cable, if there is a spareavailable). This cable is connected from the first available terminal on the firstasync card in the back of the OMC-R.

• Add a port monitor process (ttymon), using the sacadm command. An example is thefollowing:

# sacadm -a -p ttymon1 -t ttymon -c /usr/lib/saf/ttymon -v `ttyadm -V` -n 1000

• Please note that the word following the “-p” argument is the tagname of the portmonitor being added. This tagname will be used in the following command, which mustbe executed next (all one line):

# pmadm -a -p ttymon1 -i root -s 0 -f u -m `ttyadm -p tty0login: -h -d /dev/term/a0 -l 9600 -s /usr/bin/login` -v `ttyadm -V`

• The above command will tie the port monitor process (established with the sacadmcommand) of tagname ttymon1 with a login service. Note that the “-d” argument mustbe replaced with the actual device number (a0 is the topmost connection on async card 1,a7 is the bottom-most). The 9600 entry is the speed of the modem, while the “tty0login:”is the prompt presented upon login.



Ethernet Configuration

• Ethernet sub-system first established by adding thecards with a “cfinstall” (on Tandem).

• Several files govern ethernet configuration: /etc/inet/strcf, /etc/inet/rc.inet, and /etc/hosts.

• The /etc/rc4.d/fast/S69inet script starts and stops theethernet stack.

• The “ifconfig” command allows management of net-work interfaces to unix.

• The “netstat” command displays status and configu-ration (routes, etc.) for network interfaces.

198.175.180.10

198.175.181.10

OMC-R

198.175.180.11

198.175.181.11

MM-1

198.175.180.1n

198.175.181.1n

MM-n

...

OMCLAN-2-omcOMCLAN-1-omc

198.175.182.01 198.175.182.01

198.175.183.01

198.175.183.01

198.175.182.01FEP-1

FEP-2

FEP-x

Bridge

Bridge

IFELAN-1

IFELAN-2

IFELAN-1

IFELAN-2

FEP-1

FEP-2

FEP-x

Bridge

Bridge

enet0 (slot 161)

enet1 (slot 164)

enet2 (slot 166)

enet3 (slot 168)

enet0 (slot 161)

enet1 (slot 164)

enet2 (slot 166)

enet3 (slot 168)

enet0 (slot 161)

enet1 (slot 164)

enet2 (slot 166)

enet3 (slot 168)198.175.183.01

MMI LAN-1

OMCLAN-1-mm1

OMCLAN-2-mm1

OMCLAN-1-mm<n>

OMCLAN-2-mm<n>MMI LAN-2



NOTES:

Ethernet Configuration

The commands and comments given here apply to the MM and OMC-R plat-forms specifically, but most Unix-based platforms will be very similar.

• The ethernet cards are installed as part of the MakeSC9600 script. The followingcommand will install the first ethernet card on the MM or OMC:

cfinstall -i iop1 -c enet0 -s ioc0

• The files that configure the ethernet sub-system are:

• /etc/inet/strcf. This file configures the TCP/IP system, including the ethernet cards.An example command in this file for ethernet card 0 is as follows:

cenet ip /dev/enet enet 0 # S2 eagle enet

• /etc/inet/rc.inet. This file performs the ifconfig files for each ethernet card, givingeach card an identify on the network, and the file also sets up a route for traffic.Example config lines in this file are as follows:

/usr/sbin/ifconfig enet3 vulcan5 up -trailers netmask 0xffffff80 broadcast136.182.112.255/usr/sbin/ifconfig enet0 OMCLAN-1-mm1 up -trailers/usr/sbin/route add default 136.182.112.254 1

• /etc/hosts. This file contains name to IP address translations. The “vulcan5” and“OMCLAN-1-mm1” names given in the above command must have entries in thisfile with proper IP addresses.

• The script /etc/rc4.d/fast/S69inet can be used to stop and start the ethernet services (justinvoke the script with either a “stop” or “start” argument).

• The “ifconfig” command may be used as follows to print information on ethernet:vulcan5:/etc/inet > ifconfig enet0enet0: flags=4023<UP,BROADCAST,NOTRAILERS,MULTICAST> inet 198.175.180.11 netmask ffffff00 broadcast 198.175.180.255

• The “netstat” command may be used as follows to print information on ethernet:vulcan5:/etc/inet > netstat -irName Mtu Network Address Ipkts Ierrs Opkts Oerrs Collisspsl1 296 192.150.151 SP_HOST 1201495 0 1213658 0 0spsl0 296 192.150.151 SP_HOST 1211464 0 1223859 0 0lo0 2048 loopback localhost 2857 0 2857 0 0enet2 1496 198.175.182 IFELAN-1 1366502 0 1847960 0 12enet0 1496 198.175.180 OMCLAN-1-mm1 365109 0 491086 0 205enet3 1496 136.182.112 vulcan5 160374 0 150622 0 274enet1 1496 198.175.181 OMCLAN-2-mm1 0 0 0 0 0



Other Unix “Tricks”

• The ‘find’ command.- Useful for performing operations on files which

match certain parameters.

• The ‘dps’ command.- Useful for checking process consumption of mem-

ory, CPU resources.

• High-priority console.- Necessary to enable access to MM when MM is

under “stress”.

OMC-R

MM

These tricks are reallymaking me smart!



NOTES:

Other Unix “Tricks”: Find, dps, high priority session

• The Unix find command is very convenient for performing operations on files thatcomply with a desired attribute. For example, to remove all files with the extension ‘254’,the following find command may be performed:

# find . -name \*.254 -exec rm -f {} \; -print

The above command will find files starting in the current working directory, and willexecuted the command “rm -f” on the files which end in a ‘254’ extension. Thecommand to be executed may be any command which may be executed by the Unixshell. The brackets ‘{}’ will be substituted with the file name, while the escaped-semi-colon ‘\;’ signals the end of the command. The -print directive will print those files whichwere removed.

• In order to find out which processes are consuming the most time and memory resourcesin the system, a command called “dps” may be used. Command usage:

Usage: dps [-T] [-r] [t [n]]

Where ‘-T’ specifies to print time-of-day for each report,‘-r’ specifies relative time reporting (default is absolute),and ‘t’ specifies time (no default) between ‘n’ samples (default 1).

Example: dps 5 2

Reports two 5 second samples with respect to absolute time.

• On very RARE occasions, an MM may lock-up (no response from the MM tty sessions).Usually this is due to a run-away Supercell process which is running in the Real-Timepriority class. In order to provide a means to debug the situation, it is recommended thatthe operator establish a high-priority tty session at the console of the MM through thefollowing command (after logging into the console as root):

# priocntl -e -c RT -p 59 -t 1000 ksh

In addition, the following command will list all processes currently running in the RTclass:

# priocntl -d -i class RT

Check to make sure that the shell running at the console window is among the listresulting from the above command with the following (which prints the shell’s processID):

# echo $$



Still More Tricks & Tips

• IP addresses, and NameServers.

• Using FTP, TELNET.

• Using PWID’s anonymous FTP server.

OMC-R

MM

Wow! I’m really becominga power user!



NOTES:

More Tricks & Tips: ftp, telnet, anonymous ftp, etc.

• Computer platforms which have a TCP/IP communications stack are identified on theInternet with numbers called IP addresses, such as 160.4.22.1 (this is the IP address ofthe file server in Ft. Lauderdale). Names may also be used, but you must have a “NameServer” which is capable of translating the name into an IP address (or the name mustbe in the /etc/hosts file of your local platform). Most name servers have a program called“nslookup” which allows you to find out the IP address of the Name Server, and of themachine you are looking for. For example, to find the address of the anonymous FTPserver in Rolling Meadows:

segws005 $ nslookup ftp.rollingmdwsName Server: segfs001.ftlauderdale.pamd.cig.mot.comAddress: 160.4.22.1

Non-authoritative answer:Name: ftp.rollingmdws.pamd.cig.mot.comAddress: 160.4.64.1

Anonymous FTP means that the user may enter “anonymous” as the user id for theremote system, and the password is usually the user’s email address. Many anonymousftp sites exist all over the internet for different types of “stuff”.

• FTP stands for “File Transfer Protocol”, and is used to transfer files from one system toanother. If you have the IP address of the platform, you can start an FTP session, andtransfer files to your local platform. Inside FTP, you should set the “bin” mode (so thatfiles get transferred in binary mode), and you should enter the “hash” command (so thatyou can see the progress of a transfer). Other commands are available, such as “lcd” (forchanging the local directory), “cd” (for changing the remote directory), “get” (whichgets a file from the remote system, and “put” (which puts a file into the remote systemfrom the local system).

• Telnet is used to setup a session on a remote unix session. Again, an IP address is neededby the telnet command in order to login to the remote system. When a connection isestablished with the remote system, a login prompt is presented, and the user must entera username and password to gain access to the remote system. One trick that is used is toenter the command “DISPLAY=<LOCAL_IP>; export DISPLAY” on the remotesystem, and “xhost+” on the local system, and then start a program that uses X-windows. The program will be displayed on the local system, but will actually berunning on the remote system.



And Still More Tips!

• The tar command.

• Compression Utilities.

• Framemaker batch files.

OMC-R

MM

Okay! I’m learning as fastas I can!



NOTES:

Tricks & Tips: tar, compression utilities, framemaker batch files

• The ‘tar’ command is used to place all the files in one or more directories into one “flat”file which can then be compressed and transferred using ftp. To make a tarfile called /tmp/scap.tar which contains all the files and sub-directories under /usr/test/scap:

segws005 $ cd /usr/testsegws005 $ tar -cvf /tmp/scap.tar scap

Notice that we made /usr/test the current working directory before doing the tar. Wecould have specified /usr/test/scap as the directory to tar, but then when you “un-tar” thetar file, the command would attempt to create /usr/test/scap, and many times the userwill not have the access privileges to do that. To un-tar a tarfile:

segws005 $ tar -xvf /tmp/scap.tar

• Usually a tar file will be created in order to transfer the information over a link to a localsystem. It is convenient to compress this information so that the transfer does not take aslong. Several compression utilities exist for Unix. The most common is “compress”, andits counter-part, “uncompress”. These programs create a file with a “.Z” extension.

segws005 $ compress /tmp/scap.tarsegws005 $ ls -la /tmp/scap.tar*-rw-r--r-- 1 ashapple engr 1329 May 23 10:36 /tmp/scap.tar.Z

Another set of programs are the “gzip” and “gunzip” utilities- these are made by thefamous “GNU” programmers, and is shareware (freely available on the Web). These aremore efficient than compress/uncompress, and are generally available for manyplatforms (however, check to make sure both platforms have this utility before using it).

• Much documentation is now being done in Motorola using Framemaker (a powerfuldocument processing software package). When Framemaker versions change, the usermust convert the files to the new version. A script which will do this for all the files in agiven directory (except those ending with a “.ps” or “.backup” extension) is as follows:

#! /bin/kshfind . ! -name \*.ps -type f -print > /tmp/files.framefind . -name \*.backup -exec rm {} \;> /tmp/frame.scriptfor file in `cat /tmp/files.frame`do echo “Open $file” >> /tmp/frame.script #echo “Update $file” >> /tmp/frame.script echo “Save $file” >> /tmp/frame.script echo “Quit $file” >> /tmp/frame.scriptdoneecho “Quit” >> /tmp/frame.scriptcat /tmp/frame.scriptfmbatch -v /tmp/frame.scriptrm /tmp/files.framerm /tmp/frame.scriptfind . -name \*.backup -exec rm {} \;



Tools for Analysis of CDMA Data

• Data collected through:- PM reports stored at OMC-R.

- Drive tests (using Safeco, LCC, or MotorolaCAMPS).

- SMAP platform- used at CBSC to collect forwardlink information.

• Data analyzed through:- UNO platform GUI, reports.

- XFreq, pm_sum programs.

- Automatic Overnight Analysis program.

• Sources of information given for the above.

OMC-R

OMC-R

How do I analyze thissystem’s performance?



NOTES:

Tools for Analysis of CDMA Data

There are several tools available for analysis of data from a CDMA system. Thefollowing points are presented concerning this:

• Data is collected automatically by the system, and stored at the OMC-R platform. Thisdata is analyzed by programs running on the UNO platform which present the data in auser-friendly manner. The tables and the reports which may be generated from them arespecified in the “Performance Analysis” manual from TED- part # 68P09226A23-2.

• Data collected from the system will be due to use of the system- the data will includerecords on calls which failed with given CFCs. This data is usually of much interest tothe operator, in that call failures will need to be analyzed.

• A software program called Xfreq by Allan Shedlo (see http://www.rochellepark.pamd.cig.mot.com/~ashedlo/) will present a graphical interface to thedata, while another software program (also by Allan) called “pm_sum” will generateuseful reports on the Performance Management data for purposes of analyzing networkproblems.

• Data may also be collected through drive tests. Drive tests will usually be conductedusing Safeco equipment, although recently another company called LCC is producingCDMA drive test equipment, and Motorola has produced a product called “CAMPS”for drive testing. The CAMPS web page is: http://scwww.cig.mot.com/SC/mgmt/tools/CDMA/Test_Tools/CAMPS/index.html.

• The data collected from drive testing is analyzed using a software program called AOA(Automatic Overnight Analysis). The name is somewhat of a misnomer, since it usuallydoesn’t take all night to analyze the drive test data. The AOA home page is: http://www.fwrdc.rtsg.mot.com/TPD/Projects/AOA/product.html.

• Every two weeks there is a conference call held by a group of engineers and managerswho are involved with the analysis of CDMA network performance. Jay Moore isheading up this group- the group currently has representatives from all the majormarkets (Antonio Shappley is representing PWID). The notes from the meetings andother notes or questions are sent to a mailing list: [email protected]. Sendemail to [email protected] with a line which states “subscribe cdmadata”(in the body) in order to become a member of this mailing list.

• Another tool used to collect data from the system during operation and/or drive tests isSMAP. TED publishes an operator’s guide (68P09226A14). The project manager forSMAP is John Bernas- 847-632-3557.


Conclusions:• Having on-site access to the outside world from the

MTSO can be a life-saver!

• Allowing access to the MTSO (MSC, CBSC, etc.) toremote engineers can also be a life-saver!

• Commands “netstat -ir”, and “ifconfig” are used tofind status and configuration of communications sub-systems.

• The find command is convenient for performing oper-ations on many files.

• The dps command works on MM and OMC-R to seewhich processes are using the most resources.

• A high-priority session must be established on theMM and OMC-R console.

• Ftp, telnet, anonymous ftp are used to access remotesystems.

• The tar program can be used along with compressionutilities to ship many files to other systems.

• Framemaker batch files can be used to bring manydocuments up to the latest frame release.


References:Many sources of information exist for CDMA. Following are most of the sources used by theauthor to put this course together, and to learn about the CDMA technology and the CDMASupercell system in general.

• Supercell Library (http://www.sc.cig.mot.com/SC/library/index.html). Most documents inthe library are available on-line in the Arlington Heights network under /usr/test/scdocrel.

• The PWID Rolling Meadows /users/CDMA directory. Under this directory are many usefulldocuments for CDMA, including the CDMA RF Guide.

• TEM (Technical Enrichment Matrix) Presentations (http://www.acpg.cig.mot.com/w3/APD/TSG/TEM/tem.html). TEM video tape libraries are in Ft. Worth and in Arlington Heights.In Arlington Heights, call Jennifer at 632-4133. Of particular interest to CDMA are:Supercell Software Architecture, by Paul Steinberg et.al., Supercell Applications Processor,by Chris Shenefiel, and the SC600 Overview, by Craig Reilly et.al. Some presentations willsoon be available at Ft. Lauderdale- contact Felix Garcia about this (954-489-2051).

• The PWID Core Tools home page (http://www-home.pamd.cig.mot.com/group_pages/groups_html/core_tools). Check out the CDMA BSS Sysgen Creator Guide (CBSC DatabaseTool), under “Digital Database”.

• The Arlington Heights FOA group’s home page (http://www.sc.cig.mot.com/SC/test/cdma_foa). Notes from FOA group members may be found under this link- including theCDMA Recommended Parameters document.

• Mobile Cellular Telecommunications, William C. Y. Lee. Available from McGraw-Hill.Purchasing line: 800-722-4726. Product Code: 0070380899. This is overall one of the bestbooks on Cellular Technologies available. It covers analog and digital technologies, includingCDMA.

• U.S. CDMA Commercial System Specification (The B1 Document), Motorola CIG. Thisdocument presents Motorola’s CDMA to Motorola customers. It’s a good overview of theMotorola CDMA product. Available from the Supercell library.

• SC9600 Architecture Document (The ARD Document), Motorola CIG. This documentdiscuses the SC9600 equipment architecture in some detail. Also available at the CIGlibrary.

• CDMA Call Processing System Functional Specification (The CP SFS Document). Alsoavailable from the Supercell Library.

• CDMA Hand-Off and Power Control System Functional Specification (HOPC Document).Also availabel from the Supercell Library.

• TED Courses GNL180 (CDMA: IS-95 Implementation & Operation), and GNL190 (CDMACall Processing). These courses are given by Space 2000, usually by Dr. Simo, or by Dr.

PWID Center for Excellence Revision 0, 4/24/97 References


Nored. Both courses are highly recommended.

• The CDMA Command Reference Guides. Available from TED. Call George Reed (847-632-2747) or Dora Lopez (847-435-7661).

• The CDMA SC9600 and SC4950 BTS Installation and Optomization Guides, Availablefrom TED.

• The cdmadata, cdmaprod, and cdmadbcm mailing lists. Send email [email protected], with the line “subscribe <listname>” in the body. Youwill be added to the mailing list, pending that the list is “open”. Approval of thesubscription may be necessary for “closed” lists.


Acknowledgements:• Paul Steinberg and others- Supercell Software Overview TEM.• Marilyn Escue- CDMA Call Processing TEM.• Steve Wasilew- Introduction to SuperCell Digital Cellular Technology.• Phil Tsubaki- XC Overview.• Mike Jones from TED- Much documentation on CDMA- manuals, instructor

notes, etc.• Many, many others who helped out in one way or another- Thanks!

PWID Center for Excellence Revision 0, 4/24/97 Acknowledgements


Page intentionally left blank.


Last Page!Please use this page for any notes you want to make.

CDMA Basic

Documents

accepted commandenable

progress statusstarted

motorola confidential

motorola confidential

technical

single ended

optional echo

a7 term handler