General Packet RadioSystem (GPRS) Testing:Why and How?

Application Note 1384

2

Introduction

Since GPRS is intended to be usedfor data transfer, quality and devicereliability are important factors formaximizing the system network’soverall performance. Given that networks have limited capacity, marginal RF performance effects the data rate of the entire system.Higher quality GPRS devices allownetwork resources to be used moreeffectively, improve network efficiency,and facilitate faster data rates.

This document is intended to highlightconflicts of interest surroundingGPRS test in the mobile phoneindustry and give an insight intosome of the technical difficultiesencountered in the transition from GSM to GPRS mobile phoneproduction. Rather than explainingGPRS or GSM technology, this paperoutlines issues that make GPRS testing during device manufacturingan important procedure, explainswhy GPRS testing is necessary, and shows how testing can beaccomplished in order to producegood, reliable GPRS devices.

3

Table of contentspage

1 Why test GPRS 4

1.1 NOP/NEM conflicts of interest 4

1.2 Progression from GSM to GPRS and the related difficulties 4

1.2.1 Timing of up/down link bursts 5

1.2.2 Receiving adjacent bursts at different power levels 6

1.2.3 Transmitting adjacent bursts at different power levels 7

1.2.4 Output RF Spectrum (ORFS) due to switching 7

1.3 GPRS without test 8

2 Methods of testing GPRS 9

2.1 Test modes 9

2.1.1 BLER mode 9

2.1.1.1 Using BLER mode to assess receiver sensitivity 10

2.1.1.2 What do the specifications state 11

2.1.2 ETSI test modes 12

2.1.2.1 ETSI test mode A 12

2.1.2.2 ETSI test mode B 13

2.1.2.3 BER testing 13

2.1.3 Manufacturers test mode 14

2.2 Why use a protocol based test mode 15

3 Conclusion 15

4

1 Why test GPRS

Data transmission for the mobilestation (MS) is heralded as the nextbig market for the communicationsindustry. The deployment of GPRStechnology is viewed as a key step-ping stone towards migration to 3Gand is fueling a headlong rush to getnew products out.

The manufacturers' race to be one ofthe first to offer a cost effective GPRShandset, has given rise to problems.GPRS is a complex system and devicesare much more complicated thantheir predecessors were, which hascreated difficulties in providing GPRSterminals that operate reliably onnetworks. The problem has been madeworse by the lack of requirement forthird-party type approval for GPRS.

As this paper discusses, understand-ing and implementing test require-ments and methods for testing ofGPRS terminals is imperative formanufacturing success.

1.1 NOPs and NEMs conflict of interest

Network operators (NOPs) are facingconsiderable challenges as the MSmarket matures. Issues of increasedcosts (3G licenses) and lower revenuesmean that a new revenue streamand greater utilization of existingbandwidth are becoming an urgentrequirement. GPRS could offer thisopportunity provided the content isuseful and the technology reliable.

Network equipment manufacturers(NEMs) are also facing an increas-ingly challenging market place andhave cost considerations of theirown. With the price of MS falling,NEMs must take measures to reducecosts and maintain margins. Onepotential way to do this is to mini-mize test in the manufacturingprocess. The resulting increase inthroughput and reduction in outlayfor manufacturing equipment areseen as big incentives. However, thisremoval of test is contrary to theinterests of the NOPs.

The performance of a GPRS networkis substantially degraded by MSswith poor RF performance. UnlikeGSM telephony, when a block of corrupted data arrives at the GPRShandset, it will request a retransmis-sion. If this happens regularly, theuser will experience a poor data rateand the network will become ‘tied up’with retransmission of data, slowingthe data rate for all other users inthe cell. Also, the user's battery lifewill be reduced by the extra trans-mission of data. As GPRS devicescan be multiple uplink capable, battery life is very important.

Corruption of data can happen formany reasons. An obvious causecould be a poor receiver in the handset failing to correctly captureeach burst. Other problems could becaused by poor transmitters in theMS failing to produce accuratebursts or creating interference forother users. The reasons why theseproblems are more likely in GPRSthan in GSM are discussed later inthis document.

Whatever the anticipated problem,NOPs will have a great interest inensuring that all the MSs they usehave been adequately tested to verifyoptimum RF performance. Thisrequirement seems to contradictcurrent trends by NEMs to minimizeGPRS test.

1.2 Progression from GSM to GPRS and the related difficulties

One industry view is that, as anextension of GSM, GPRS can be adequately tested using existingGSM test methods. More clearly stated, ‘it should be possible to confirm the GPRS performance of a handset by testing only its GSMperformance’. This view stems froman early consensus that to achieveGPRS functionality would requireonly software changes to a GSM MS.However, some manufacturers arenow finding that over 75% of thedesign of the device required modification to achieve GPRS functionality.

Some of the main difficultiesencountered are described in sections 1.2.1 through 1.2.4.

5

1.2.1 Timing of up/down link bursts

With traditional GSM, the timing ofthe uplink and downlink channels isoffset by three timeslots (see figure 1).This means that if the MS receiveson downlink timeslot 0, the MS doesnot transmit until the downlinkequivalent of timeslot 3. The result isthat a GSM MS always has two time-slots to switch between RX and TX.

GPRS does not afford the MS thisluxury, often giving a maximumswitching time between uplink anddownlink of one timeslot rather thantwo. This places a much greaterstrain on the MS's ability to re-tuneand lock on the different frequencies(it puts demand on the local oscillator(LO) and its phase lock loop (PLL)performance). It also leaves a shortertime for the MS to retune to ‘listen’to surrounding cells for their BCH.

To cope with the added strain, GPRSMSs must have a more agile LO toretune in such short time periods.Verifying the increased agility of theMS receiver is an important test forGPRS. When the LO/PLL fails to settlequickly enough, increased frequencyerror occurs. Significant phase errorat the start of the burst can also beattributed to slow settling of theLO/PLL. As with other problems,these errors will result in corruptionof data.

Figure 1b: Burst offset arrangement in GPRS

Figure 1a. Burst offset arrangement in GSM

Uplink

Downlink

Time

Timeslots (TS)

Power level

TS5 TS6 TS7 TS0 TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS0 TS1 TS2TS0 TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS0 TS1 TS2 TS3 TS4 TS5

Uplink

Downlink

Time

Timeslots (TS)

Power level

TS0 TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS0 TS1 TS2 TS3 TS4 TS5TS3 TS4 TS5 TS6 TS7 TS0 TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS0

6

1.2.2 Receiving adjacent burstsat different power levels

In standard single slot GSM, eachframe contains only one burst ofdata per mobile on the downlink.Changes in the power level of thisburst are normally communicated tothe MS once every 26 frames on theSlow Associated Control Channel(SACCH) or on the Fast AssociatedControl Channel (FACCH) (by stealingspeech frames). With GPRS, the MScan receive data on multiple timeslots in the same frame, the powerof each burst differing by up to 10dB from the previous. To cope withthese adjacent transitions, the gaincontrol of the amplifier in the RXstage of the MS must be able toadjust and settle much faster thanwas required in GSM. Specifically,rather than managing the transitionover one frame period, the MS mustnow achieve the transition withinthe ramping periods of the adjacenttimeslots. This is only 34 µs asopposed to the 4 ms afforded inGSM (see figure 2).

If the amplifier cannot adjust fastenough then corruption of data islikely to occur. A commonlyobserved problem is where theamplifier fails to adjust to a highergain rapidly enough. If two burstsare transmitted with the second at amuch lower power than the first, thereceiver can become saturated,destroying the data and prompting aretransmission. This situationresults in increased network trafficand lower network efficiency.Problems have also been seen on theopposite transition, with the firstdownlink timeslot set to a lowerpower level than the second. Thisissue is due to the slew rate of theautomatic gain controller (AGC)being too slow.

Figure 2. Period between bursts is only 34 µs

Time

Power level

Range for Mode A

PRef

PRef-10dB

BCH

TS1 TS2 TS3

7

1.2.3 Transmitting adjacent bursts at different power levels

The existing ETSI defined PVT maskfor GSM has been modified to allow formultiple contiguous uplink timeslotsat different power levels. Similar tothe case of the adjacent RX, thisplaces additional strain on the TXstages. In the standard GSM PVTmask, the MS only has to cope withthe ramp up, overshoot, useful part,and ramp down characteristics of asingle burst. The multiple adjacentuplink bursts possible in GPRS meana MS has to transmit for much longerperiods while also contending withpower level transitions between bursts.

These uplink bursts can differ in powerby up to 30 dB. Achieving sufficientburst level accuracy on these uplinktransitions can prove very difficult.The resulting performance variesconsiderably depending on the vari-ation in burst power level, the natureof the transition (Hi/Lo or Lo/Hi) andthe method of burst transition thatis adopted (see section 1.2.4 OutputRF Spectrum due to switching). Theextent to which the effects are seencan be markedly different from MSto MS even with the same model.

As the GPRS transmitter may now becapable of transmitting on more thanone timeslot it is also necessary toconsider the thermal implications thatmight impact the TX performanceover long TX sequences.

1.2.4 Output RF Spectrum due to switching

Output RF Spectrum (ORFS) due toswitching measures the spread inspectral RF energy due to GSM/GPRSburst ramping. The results of this testare likely to be quite different for amulti-uplink GPRS MS compared to aGSM MS. In general, if the transmitterramps too quickly, unwanted spectralenergy spread will occur in thetransmission. By spreading intoother frequency channels, thiseffects the quality of service experi-enced by other users in the cell, thusincreasing their TX power (reducingbattery life) and increasing overallinterference. If the MS ramps up tooslowly, ORFS problems are avoidedbut you might see bit errors at thebeginning of bursts or the burst willbe outside of the specifications forpower versus time. Also, if the MS isramping much too slowly this interferes with users in adjacenttimeslots.

The ORFS due to switching is likely tobe quite different for a multi-uplinkGPRS MS than for a GSM MS. This is because of the difference in thetransition characteristics for adjacentbursts compared to a single burst.This transition must be within thePVT mask for the “guard period”between bursts.

A large consideration is whether thetransition is achieved by using twonormal GSM bursts, accepting what-ever transients are produced betweenthe normal ramp down/ramp uptransition, or by blurring the transi-tion and creating one double burst.Neither method has been specifiedin the standards so manufacturersmust decide on one method andthen ensure that it does not createan ORFS result that falls outside thedefined frequency mask.

Figure 3. Specification for uplink power vesus time mask

t

dB

+4+1-1-6

(**)

-30(***)

(*)

10 µs 10 µs8 µs

10 µs 10 µs8 µs(147 bits)

7056:13 (542.8) µs

(147 bits)

7056:13 (542.8) µs

34 µs(approximately)

8

1.3 GPRS without test

GPRS is the first extension of GSMto be defined without a requirementfor third-party type approval. Thisleaves the responsibility for designverification with the manufacturer.With many early MS designs failingto function adequately on the net-work, it can already be seen howfreedom to not fully adopt themandatory standards within GPRS, coupled with a lack of test,has disrupted Network Operatorprogress, i.e. not all mobiles work onall networks. This will affect roaming.

Testing offers an alternative tothird-party type approval by provid-ing consistency of standards acrossthe industry. Without manufacturersadopting a clear test policy for GPRSit is expected that a significant num-ber of MSs may display sub-optimalperformance. Consequently, furtherdifficulties can be predicted for theindustry.

Current estimates suggest that anNOP spends on average $700 toattract each new subscriber. In anindustry approaching some level ofsaturation, the NOP must work hardto retain this customer through newfeatures, benefits, and services. Withthe failure of WAP and other dataservices to take a serious hold onthe market, it is crucial that NOPsgenerate high expectations for whatGPRS will provide. They have todeliver on these expectations if theyare to allow the market to regainlost momentum.

These issues are important not only for GPRS but also to ensurethat W-CDMA will be given the bestpossible launch platform. GPRS willbe the first format designed primarilyfor data and will set the benchmarkfor user experience. NOPs hoping todispel the negativity around WAPmust use GPRS to demonstrate thevalue of MS data. The key will be toprovide services that are both usefuland enhanced by higher data rates.

If GPRS connections fail, are slow,or unreliable, it will only confirmthe market’s suspicions that MS datais not a technology that offers realbenefit. This is the concern of theNOPs but should also be consideredby the NEMs who rely on the NOPsto provide the services to sell theirproducts. In order to avoid disap-pointing users, the industry mustagree on a set of tested standardsfor the operation of MSs on GPRSnetworks.

BLER test mode

Uplink

PDTCHPDTCH

PACCH

9

2 Methods of testing GPRS

As GPRS is a packet-based, connec-tionless system you are faced with aproblem when determining a methodof test. In GSM it is easy to set theMS into continuous transmissionand reception (just by making a call).In GPRS, the MS only transmits andreceives when transferring data. Asa result, a method is needed bywhich the MS can be forced intocontinuous communication. Thereare four ways to approach GPRStest. These are described in the sections 2.1.1 through 2.1.3.

2.1 Test modes

2.1.1 BLER test mode

The BLER connection mode is anAgilent-proprietary (patent appliedfor) method that makes use of stan-dard signaling messages. This methoduses a packet ack/nack count fromthe mobile to compute the BLER. Tomake an acknowledgement the MSmust transmit. By keeping a count ofblocks that are received with, andwithout error, it is possible to assesshow well the MS receiver is copingwith incoming data.

This connection method sends GPRSMobility Management (GMM) infor-mation messages to the mobile withthe polling bit set. This forces themobile to respond with ack/nackmessages, which allows the test setto compute BLER. There are someadditional user configurable featuresin Agilent BLER mode that improvethe reliability of this connectionmethod for different manufacturers’mobiles. For example, by default amessage is sent to the mobile inevery block. This can cause somemobiles to be over-run. To counteractthis the polling rate can be sloweddown. Another situation that canoccur is that mobiles can respond tothe GMM information message indifferent ways. In some cases thiscauses the mobile to not respond inan appropriate manner. To overcomethis problem, the Logical Link Control(LLC) Frame Check Sequence (FCS)can be deliberately corrupted, whichforces retransmission of LLC blocksand in turn keeps the RLC/MAClayer transmitting.

Through this, BLER mode provides a method to test both the RX and TXperformances of the MS. As describedin section 1, both TX and RX meas-urements are very important forGPRS test.

A strong advantage of this mode isthat it does not require any specialfirmware in the MS. All other GPRStest modes require the MS to beeither controlled externally, to produce PRBS internally, or havethe ability to loop back data. This is unnecessary with Agilent BLERmode, allowing reliable testing of themobile early in its development. Infact, any GPRS MS should be able tobe tested using this mode.

As this mode of operation usesRadio Link Control/Medium AccessControl (RLC/MAC), signaling inwhat would appear to the MS asnormal communication, verifies that the lower layer protocols arefunctioning correctly.

Figure 4. Diagram of BLER connection

10

2.1.1.1 Using BLER mode to assessreceiver sensitivity

If the MS has a minimum sensitivityspecification of –x dBm, that is theminimal level at which the MS canreceive without error (or with aspecified error rate). The base station emulator can be set up in thefollowing configuration to obtain thebest results from the test set/MS.

By setting ‘burst 2’ of the downlinkclose to receiver reference sensitivitylevel and then setting ‘burst 1’ 10 dBhigher than this, you can both ensurethat the MS receives the signalinginformation and stress the receiverof the MS to the maximum. How? Inthis situation the MS has to cope witha very low-level signal and also witha maximum power level transitionspecified in the standards. By meas-uring its BLER you can assess theperformance of the receiver in thisextreme scenario. If this is set to thespecifications for MS sensitivity,then the MS should have a BLERresult of <=10%. If there is a valuegreater than this then there aresome issues with the MS's receiver.

This configuration can also bereversed by having the lower powerburst first. This assesses the slow rateof the receivers automatic gain control.

Another method for testing receiversensitivity is to gradually reduce thepower of the downlink to see whereBLER starts to increase showing thelimit of the MS's capabilities.

Figure 5. Setting up power offsets and reduction levels

Power

Time

-60 dBm

-85 dBm

-90 dBm

-x dBm

BCC

P0

PRL1 PRL2

PDTC

(burst 1)

PDTC

(burst 2)

11

2.1.1.2 What do the specifications say?

The Release 99 GSM 5.05 (now 3GPP45005) standard refers to the referencesensitivity of GPRS MS. The followingtable shows the reference sensitivityinput levels for GPRS MS in the vari-ous bands under static conditions.

The BLER rate in each case must beless than or equal to 10%.

GSM 400, GSM 700, Reference levelGSM 850, GSM 900, under staticDCS 1800, and conditions (dBm)PCS 1900PDTCH/CS-1 -104

PDTCH/CS-2 -104

PDTCH/CS-3 -104

PDTCH/CS-4 -101

The input levels given in the abovetable are referenced to normal GSM900 MS and have to be corrected bythe following values for other MSs:

GSM 400, GSM 700, GSM 850, and GSM 900 small MS +2 dB

DCS 1800 class 1 or 2 MS +4 dB

DCS 1800 class 3 and PCS 1900 class 1 or 2 MS +2 dB

PCS 1900 class 3 MS 0 dB

Further complication is added bythe requirements of the MS to beable to receive data when there ispower in adjacent timeslots. Thereare two situations.

Situation 1 – The MS has to receivedata in an adjacent timeslot intendedfor it, as well as the timeslot in ques-tion. In this case, the standardsspecify that the MS must be able to cope with a difference in powerlevels between the timeslots of up to 6 dB, without degradation of performance.

Situation 2 – The MS has to be ableto receive data in the presence ofpower in an adjacent timeslot notintended for the MS. In this case thedifference in power levels can be upto 20 dB.

12

2.1.2 ETSI test modes

Due to the requirement for testing of GPRS MSs, ETSI have defined testmodes specifically for type approval,these also offer a solution to themanufacture test. They work by forcing the MS to respond to givenstimuli over the air interface. Thesetest modes are defined in GSM 04.14version 6.2.0 Release 1997, Section5.4. and are referred to as ETSI testmode A and B. ETSI test mode Aonly allows transmitter testing whileB provides the ability to performtransmitter and receiver testing.Although mandatory in the standardsmany manufacturers have not yetimplemented either of the ETSIdefined test modes.

2.1.2.1 ETSI test mode A

Once the MS is camped on to the basestation signal and has performed aGPRS attach, a downlink TemporaryBlock Flow (TBF) is established anda “start ETSI test mode A” protocolmessage is sent over the air interface.(A TBF is a physical connection usedby the two Radio Resource (RR)entities to support the unidirectionaltransfer of Logical Link Control (LLC)Packet Data Units (PDUs) on packetdata physical channels.) This instructsthe MS to enter test mode A. Theconfiguration of the link (channel,timeslot(s), coding scheme, and such)is managed in accordance with thenormal rules for transmission, asdefined in GSM 04.60. The MS mustthen produce its own PRBS (Pseudo-Random Bit Sequence) to fill thedata blocks and continuously transmitin every frame.

Once the MS is in a transmittingstate, the test instrument can performtransmitter measurements on the MSto ensure that it is within tolerance.

A disadvantage of this mode is thatit requires the MS to have specialfirmware that enables the MS to produce a PRBS to populate thedata blocks. Implementing this functionality in the MS can increasedevelopment time. Additionally, thismode can only be used for transmittertesting. However, receiver testing isalso an important part of GPRSmanufacture.

Figure 6. Diagram of ETSI test mode A connection

ETSI test mode A

Uplink

PDTCHPDTCH

13

2.1.2.2 ETSI test mode B

Once the MS is camped on to the basestation signal and has performed aGPRS attach, a downlink TBF isestablished and a “start ETSI testmode B” protocol message sent over the air interface. This instructsthe MS to enter test mode B. Theconfiguration of the link (channel,timeslot(s), coding scheme, andsuch) is managed in accordancewith the normal rules for transmis-sion, as defined in ETSI GSM 04.60.

The test instrument sends datablocks on the downlink, containing apseudo-random data sequence. Forthe uplink, the data payload of thedata blocks contains the data payloadpreviously transmitted to the MS onthe downlink. This continues untilthe test instrument terminates thedownlink TBF.

As the MS is looping back the down-link data on the uplink, it is possibleto perform a comparison of the datato calculate the Bit Error Rate (BER).Since the MS is transmitting onevery frame, the instrument can alsomake transmitter measurements.For this reason ETSI test mode Boffers the greatest scope for testinga MS in GPRS mode. As a result, theindustry trend has to be to includeETSI test mode B as a priority overETSI test mode A. Although BLERalso offers receiver testing, BER is amore granular and precise measureof receiver performance. However,the specification for receiver testingis defined in BLER and not BER.

This mode has various advantagesand disadvantages. On the positiveside, it is an excellent test of receiverand transmitter capability. However,this does require firmware in the MSand can be very difficult to develop.

2.1.2.3 BER testing

When in a loop back situation, it ispossible to perform a BER test. Thisgives an accurate assessment of thecapability of the MS's receiver. AsGPRS offers asymmetric configura-tions it is possible that there will bemore downlink timeslots used thanuplink slots. In this situation it isnot possible to loop back all of thedownlink data. For a test solution tobe truly useful, it is necessary thatthe user is able to select which ofthe downlink timeslots to loop back.The reason for this is that the MSwill find some transitions more difficult than others. For example, ina multi-downlink configuration withtimeslots at different power levels,the MS may have to make differenttransitions in its automatic gain control – from high to low and low to high. Potentially this offers twosituations that the MS can find difficult. If the user is able to selectwhich timeslot(s) to loop back theywill also be selecting which of thetransitions to test (maybe both).

Figure 7 Diagram of ETSI test mode B connection

ETSI test mode B

Uplink

PDTCHPDTCH

PACCHPACCH

14

2.1.3 Manufacturers test mode

This mode of communication is verydifferent to the others. In this testmode the MS and test instrument arecontrolled completely independently,that is, there is no protocol controlbetween the instrument and the MS.For this to be possible, the manufac-turers/developers must have theirown proprietary communicationmethod to control their MS. If theMS is only capable of transmitting in this mode you can only maketransmitter measurements. However,if the MS can synchronize to thebase station signal (BCH) from the instrument and is capable ofreceiving/looping back the downlinkPDTCH, you will be able to performa BER test and assess the receiversability.

In GPRS, downlink power control isaided by information contained inthe RLC data block header that tellsthe MS approximately what power toexpect. This helps the MS to makeadjustments to its AGC. Normally inthis type of test mode, all of the datain the block is filled with PRBS. Thiscan cause a problem for the MS if it isusing the power reduction (PR) fieldsin the RLC header to prepare its inputcircuits. The PR field is two bits inlength (see table). In downlink powercontrol (DPC) mode A there are twooffsets defined. The P0, which is anoffset from the BCCH, and the PRoffset. These equate to:

Power reduction Power reductionbits (PR)00 0-2 dB less than

BCCH level - P001 4-6 dB less than

BCCH level - P010 8-10 dB less than

BCCH level - P011 Not useable

As can be seen, if this field is filledwith random data and the MS isdecoding the field to aid reception, the receiver will be disrupted by the values and the BER may beincreased.

This test mode is sometimes calledinstrument test mode, engineeringtest mode, manufacturers test mode,limited mode, or just test mode.

Test mode has certain advantagesand disadvantages. On the positiveside it can reduce test time as theMS can be controlled without use of the air interface, removing thenecessity for the MS to camp, GPRSattach, and such. However, thisrequires firmware in the MS, the MSmust also be controlled manually orremotely by a PC. This can greatlyincrease development time. Also byusing this type of test mode it is notpossible to fully stress the MS.

15

2.2 Why use a protocol basedtest mode

If a protocol based test mode is notused, the MS must be controlledindividually and any changes to theconnection must be applied to boththe MS and the test instrument sep-arately. This can greatly increase thedevelopment time, as it is necessaryto develop the command protocol andhave methods for controlling almostevery aspect of the MS. However, ifyou just look at manufacture testingand not development there are stillmany reasons why GPRS should betested using a protocol based solution.

• GPRS signaling overhead is far higher than that in GSM. GPRS has to deal with lost blocks, re-transmission, and associated signaling. This signaling overhead varies from MS to MS depending onthe exact receiver characteristics and performance. Using signaling gives a much more complete and realistic test of the GPRS MS by adding the stresses a GPRS MS will have to overcome in normal operation.

• Handover performance is critical for GPRS and cannot be tested adequately without signaling. In fact, without signaling it is not possible to perform a true handover. This would just be the MS and instrument re-tuning to a new frequency. More calls are

dropped in actual operation duringhandover than at any other time. This scenario is worse for GPRS data transmission than for voice. A dropped call does not just causeinconvenience. If the MS is in the process of downloading a large file to a PC, it may not be able to just call back and pick up the conversation at the point prior to the call being dropped. The file transfer has to be either partly or completely repeated.

• Using test mode does not allow adequate exercising of the LO/PLL. The MS will not perform the exercise of re-tuning between RX and TX to measure power in neighboring cells using the first entry in the BA table. This is important because the LO/PLL is one area of the GPRS MS that has changed most when compared to GSM MS.

• By using a protocol-based test mode you are verifying the correctfunctioning of the MS’s protocol stack, while this is not truly necessary in manufacture at the early stage of a MS’s lifecycle, thisis very important.

All of these issues are addressedthrough adoption of a protocol-basedsolution.

3 Conclusion

There are many reasons to implementtesting of GPRS functionality. Theonly reason not to test GPRS func-tionality is to save money. This however, is a short-sited policy, assaving money in the short term (bydecreasing overall test time) couldcost sales in the long term. Firstly,customers do not want a MS thatperforms badly since they have topay for data regardless of whether it contains errors or not. Secondly,NOPs do not want poor quality MSsoperating on their networks due tothe subsequent increase in trafficand degradation of performance.

There is only one preventative measure to be taken here and that is a solid GPRS testing strategy. Inmany cases a GPRS testing can be agood indication of the MS perform-ance in GSM mode, the reverse ofthis is not true. A good GPRS testingstrategy should be able to test themajority of GSM functionality anddoes not have to result in a largeincrease in test time. It is hoped thatthis document has some influenceon industries’ choice of policy thusleading to the successful introductionof high performance devices thatwill fuel a growth in the age ofmobile data communications.

Agilent Technologies Test andMeasurement Support, Services, and AssistanceAgilent Technologies aims to maximize the value you receive, while minimizingyour risk and problems. We strive to ensurethat you get the test and measurementcapabilities you paid for and obtain thesupport you need. Our extensive supportresources and services can help youchoose the right Agilent products for yourapplications and apply them successfully.Every instrument and system we sell has aglobal warranty. Support is available for atleast five years beyond the production lifeof the product. Two concepts underlieAgilent's overall support policy: "OurPromise" and "Your Advantage."

Our PromiseOur Promise means your Agilent test andmeasurement equipment will meet itsadvertized performance and functionality.When you are choosing new equipment,we will help you with product information,including realistic performance specifica-tions and practical recommendations fromexperienced test engineers. When you useAgilent equipment, we can verify that itworks properly, help with product operation,and provide basic measurement assistancefor the use of specified capabilities, at noextra cost upon request. Many self-helptools are available.

Your AdvantageYour Advantage means that Agilent offers a wide range of additional experttest and measurement services, which youcan purchase according to your uniquetechnical and business needs. Solve problems efficiently and gain a competitiveedge by contracting with us for calibration,extra-cost upgrades, out-of-warranty repairs,and on-site education and training, as wellas design, system integration, project man-agement, and other professional engineeringservices. Experienced Agilent engineers andtechnicians worldwide can help you maximizeyour productivity, optimize the return oninvestment of your Agilent instrumentsand systems, and obtain dependablemeasurement accuracy for the life of those products.

By internet, phone, or fax, get assistancewith all your test and measurement needs

Online assistance: www.agilent.com/find/assist

Phone or FaxUnited States: (tel) 1-800-452-4844

Canada: (tel) 1-877-894-4414(fax) (905)-282-6495

China: (tel) 800-810-0189(fax) 1-0800-650-0121

Europe: (tel) (31 20)-547-2323(fax) (31 20)-547-2390

Japan: (tel) (81)-426-56-7832(fax) (81)-426-56-7840

Korea: (tel) (82-2)-2004-5004 (fax) (82-2)-2004-5115

Latin America: (tel) (305)-269-7500(fax) (305)-269-7599

Taiwan: (tel) 080-004-7866 (fax) (886-2)-2545-6723

Other Asia Pacific Countries: (tel) (65)-375-8100

(fax) (65)-836-0252Email: tm_asia@agilent.com

Product specifications and descriptions in this document subject to change without notice.

© Agilent Technologies, Inc. 2001 Printed in USA January 11, 20025988-4509EN