mob

Cellular Mobile Test

Application Notes

2935 Series

• GPRS Mobile Phones - An Overview For Test Professionals

• How Does the "Phase" of a Mobile Effect Its Testing Strategy

• Maximizing GSM Mobile Repair Center Throughput

• Testing GAIT Phase One Mobiles

• Testing GSM-R Terminals with the IFR 2935 Test Set

• Testing Mobiles Faster with the 2935

• Using the IFR 2935 GSM Tester with GPRS Test Option 01

• Why Does the 2935 Have Two Manual Modes

For the very latest specifications visit www.aeroflex.com

ApplicationNote

Understanding General Packet Radio Service mobile

technology is critical for test professionals. The wireless data

revolution has arrived. Are you ready for the next generation

of data enabled GSM/GPRS phones?

GPRS Mobile Phones - An Overview for

Test Professionals

I. Introduction

In this application note we will cover many aspects about

GPRS. The objective is to provide you with a compre-

hensive view of how GPRS works, and what is required to

test GPRS mobile devices. This application note is not

designed to take the place of the standard, as there is

much more to the operation of GPRS than can be cov-

ered in one application note.

Since GPRS is similar to GSM in many ways, much of this

application note will focus on the differences between

GPRS and GSM. It will also give you a good understand-

ing of how GPRS operates and how Aeroflex has taken a

lot of the mystery out of GPRS testing with the 2935 GPRS

test option.

II. What Is GPRS?

GPRS stands for General Packet Radio Service. GPRS is a tech-

nology where mobile phone users can transfer data using "pack-

ets" of information rather than conventional "circuit switched"

communications. Packet communication systems have an inher-

ent advantage over circuit switched systems due to the fact that

packet based wireless communications networks can provide

data services "on demand" to the subscriber, without being tied

down to a dedicated connection.

Within cellular standards, the Global System for Mobile

Communications or GSM family includes the GPRS standard and

both are Time Division Multiple Access or TDMA technologies.

Within TDMA technology, the user is allocated a particular radio

channel and is then assigned to one particular timeslot within that

channel. With GSM and GPRS, there are 8 users that are assigned

time slots 0 through 7, on that particular radio channel.

Until GPRS was introduced, voice and digital data were trans-

ferred over conventional GSM networks using "circuit switched"

technology where the user was allocated a particular time slot

whether data was present on the slot or not. With GPRS, howev-

er, that restriction is lifted and data is now "packet routed" and

sent as system resources allow. This is a big improvement in effi-

ciencies in the cellular network and a primary reason that opera-

tors are moving towards packet routed systems.

As we look at the transmission of data within a cellular network, it

is important to understand the relationship between information

that is dependent upon a constant time interval, such as voice

communications or live video data, and data transmissions that

are not timing dependent. Digitized voice and real time digitized

video data require a data connection that is based on a “fixed”

rate of transmission that allows the transmission of voice or video

data and the subsequent recovery of that data without significant

delays or "latency".

Any variation in the transmission of more than a few milliseconds

would cause unacceptable information dropouts and time shifts

of the information, making seamless communication difficult at

best.

Non-voice or non-real-time video data, on the other hand, does

not care if the data received or transmitted is a few milliseconds

late. This is where packet routed networks achieve a huge effi-

ciency advantage over circuit switched networks. With packet

data services, packets of information do not require a dedicated

circuit path since they are sent as requirements and system

capacity dictate. With packets, the transmitting and receiving enti-

ties are assigned IP addresses and the packet routing network fig-

ures out how to "route" packets to the appropriate device.

Figure 1.0 shows how a circuit switched network operates. Within

a circuit switched network, there is a physical requirement for a

connection from Point "A" to Point "E" before information can flow

between the two users. The switches at each switch point (B, C

and D) must be connected to the next segment or the circuit is

considered to be "open" and no information (voice or data) can

flow.

Figure 1.0 Circuit Switched Network

Figure 2.0, on the other hand, shows a simplified version of how

a "packet routed" network operates. A packet network operates

in a similar manner to the way that your "mail" or postal system

operates. Just like a letter that you send in the "mail", information

is placed into a packet (envelope) and then sent on its way. The

packet of information then goes through a number of different

routing centers (nodes) before being delivered to the final end

user. Just like a letter sent through the postal system, "To:" is the

address of the final destination of the letter, and "From:" indicates

the origination point of the letter.

Once placed into the "system" it is then left up to the "operations"

department (the network) to decide what is the most efficient path

to transmit that information from point "A" to point "E". Notice

that the data in "information packet A" takes a different route than

the data in "information packet B", however they both arrive at the

same destination because the network determines the best route

for the packet, based on capacity and utilization.

Figure 2.0 Packet Routed Network

III. Why Do Users Want Or Need GPRS?

The predominant use for GPRS services is for data services where

users place a high value on the ability to transmit data in a mobile

environment. This can be as diverse as a "chat session" for a

teenager to "e-mail" and "stock quotes" for a businessman.

Examples of data services that are available or that are being pre-

pared for GPRS include:

TTeexxtt aanndd vviissuuaall iinnffoorrmmaattiioonn such as share prices, sports scores,

weather, flight information, news headlines, prayer reminders, lot-

tery results, jokes, horoscopes, traffic, location sensitive services

and so on. This information need not necessarily be textual - it

may be maps, graphs or other types of visual information.

SSttiillll iimmaaggeess such as photographs, pictures, postcards, greeting

cards, presentations and static web pages can also be sent and

received over the mobile network as they are across fixed tele-

phone networks.

MMoovviinngg IImmaaggeess. For mobile communication to continue its

aggressive growth, much of the content will be less textual and


more visual. Moving images in a mobile environment have sev-

eral applications including monitoring sites for intruders and

remote monitoring of elderly or medical patients.

Videoconferencing applications, in which sales people can have

a regular sales meeting without having to go to a particular phys-

ical location, are another application for moving images.

Other applications such as web browsing, document sharing and

audio downloads show the diverse nature of the uses for GPRS.

IV. By The Way - What Is WAP?

WAP stands for Wireless Application Protocol. It is not specifical-

ly a GPRS function, and can be used by a number of wireless

technologies as the intermediate layer between the application

and the mobile device. Mobile devices do not enjoy the graphi-

cal display or memory overhead associated with conventional

"fixed" computer resources or laptops. WAP was designed to pro-

vide a global standard for delivering Internet features to GPRS

devices simply and efficiently.

V. When Will GPRS Happen?

As of 2002, 96 operators in 45 countries have operational GPRS

networks. GPRS enabled handsets will grow from a scant 10 mil-

lion in 2001 to over 280 million in 2005. Figure 3.0 shows the

anticipated growth of GPRS enabled devices over the next three

years.

Figure 3.0 Growth of GPRS Enabled Handsets

VI. GPRS Network Operation

GPRS deployment involves overlaying a packet-based air interface

onto the existing circuit switched GSM network. This gives the

user an option to use a packet-based data service or convention-

al voice services. To supplement a circuit switched network archi-

tecture with packet routing is a significant upgrade. Adding GPRS

to an existing GSM network requires the network operator to add

new infrastructure nodes and add software upgrades to existing

network elements.

A GPRS network operates in a mode called "Always On". This

doesn't mean that the GPRS mobile is constantly transmitting to

the network and is chewing up battery reserve along the way. The

GPRS mobile device initiates a "GPRS attach sequence" upon

power-up or GPRS mode operation. During the GPRS attach

sequence, the mobile establishes its location to the network and

its corresponding "address" through the use of the Temporary

Logical Link Identifier orTLLI.

After the GPRS attach is completed, the mobile can then send and

receive packets. The TLLI serves as the mobile's identifier to route

packets to and from the GPRS network.

The GPRS network controls the flow of the packets to and from

the mobile. This is accomplished by the use of the Packet Control

Unit or PCU, the Serving GPRS Support Node or SGSN and the

Gateway GPRS Support Node or GGSN.

The PCU provides the interface between the Base Station

Subsystem or BSS and the rest of the network by converting pack-

et traffic to PCU frames that have the same configuration as the

Transcoder Rate Adaptor Unit or TRAU frames used by GSM net-

works for transferring circuit switched data. Thus, the BSS can

now route both traffic types, packet and circuit switched informa-

tion. The PCU controls such things as cell change orders, paging

the mobile, packet powercontrol and timing advance, packet time

slot configuration and frequency hopping parameters.

The SGSN performs the mobility management function. It con-

trols routing of the packets between the PCU and the GGSN and

buffers many megabytes of packet data. It controls and routes

unacknowledged packets to the SGSN during a cell change of the

mobile station and provides session management between the

mobile and the GGSN. Compression and charging information is

also handled at the SGSN.

The GGSN is the "Gateway" between the GPRS network and the

Internet. While the mobile is roaming throughout the GPRS net-

work, the GGSN is the fixed point for packet data transmission. By

configuring the "packets" for the transmission to the Internet the

GGSN acts as the interface between the GPRS network and the

rest of the IP world using X.25, IP or PPP based networks.

The GSM BSS comprising the Base Tranceiver Station or BTS and

Base Station Control or BSC remains virtually unchanged when

GPRS is added to the network. Since GPRS uses the same mod-

ulation scheme (GMSK) as GSM, there is little to change from a RF

hardware perspective. Software needs to be updated to handle

the packet data traffic and accompanying interfaces to the PCU,

SGSN and GGSN.

Figure 4.0 shows a typical GSM network with the GPRS packet

control units and SGSN and GGSN added.

VII. How Does a GPRS Mobile Transfer Packet Data?

1. The GPRS Attach Sequence

The GPRS mobile establishes communication with the network

through a process called a GPRS attach. The GPRS attach session

includes a dialog of packets sent between the mobile and the

base station. The mobile, upon power up or being placed into a

GPRS mode, will automatically initiate an attach sequence.

There are two different types of GPRS attach sequences that can

occur. One is called simply, GPRS Attach. This applies to all class-

es of mobiles, either A, B or C. (See section VIII, 2. Mobile Station

Class). The GPRS attach allows the mobile to attach to the net-

work and then, if needed, begin packet data traffic.

The other sequence is called a Combined Attach. This applies to

Mobile Station Class A or B. (See also section VIII, 2. Mobile

Station Class). In this instance the mobile informs the network that

it wishes to be both IMSI-attached for circuit switched operation

as well as GPRS-attached for packet data services. This is similar

to a GSM registration process where the mobile has performed a

location area update and IMSI attach, along with the GPRS attach.

Now dealing with packets of data, the network controls when

those packets are sent. It can also control the quantity of packets

sent, depending on the capabilities of the mobile. It also allows

for the GPRS mobile to continuously be "attached" to the network,

although being "attached" doesn't mean that data is flowing.

To attach to the network, the mobile starts with an "attach

request" message to the network, which includes the mobiles

International Mobile Subscriber Identifier or IMSI, which is then

processed by the network into a Packet-Temporary Mobile Sub-

scriber Identity or P-TMSI and resent back to the mobile. The P-

TMSI forms the basis of what is called a Temporary Logical Link

IdentifierorTLLI , used to track the mobile from the SGSN through

the network. As part of the Combined Attach procedure, the

mobile is authenticated with the mobile's Home Location Register

or HLR. After authentication, the SGSN does an update of the

GPRS location and sends an "Attach Accept" message to the

mobile, and the mobile responds with an "Attach Complete". The

mobile is now "attached" to the network, and due to the TLLI,

packets can be routed to and from the network.

Figure 5.0 shows the Mobility Management messaging between

the mobile and the SGSN for a GPRS attach sequence.

Figure 5.0 A GPRS Data Message Sequence Chart

2. Sending and Receiving Packets

Since the mobile has now been assigned a TLLI and is attached

to the network, packet data can be sent. With a mobile originat-

ed data session, the mobile issues a channel request. The net-

work sees the channel request and issues an immediate assign-

ment. The immediate assignment points the mobile to a Packet

Data Traffic Channel or PDTCH and data blocks are then sent to

and from the mobile.

The network, knowing the location and TLLI of the mobile can also

initiate a data session by simply issuing an immediate assignment.

This is part of the Radio Resource messaging sent between the

BSS and the mobile.

VIII. Types of GPRS Mobiles

1. Type 1 MS versus Type 2 MS

The "Types" of MS are the same as in GSM. A Type 1 MS is not

required to transmit and receive at the same time.

Type 2 MS are required to be able to

transmit and receive at the same time.

Almost all GSM/GPRS mobiles are a

type 1 MS. This means that, although

they operate in a frequency offset

mode, (45 MHz - GSM/GPRS

900 MHz Band, 95 MHz - GSM/GPRS

1800 MHz Band,

80 MHz - GSM/GPRS 1900 MHz

Band) they are not technically in a true

duplex mode. This is because they do

not transmit and receive data at the

same time.

A good example of a type 2 MS is an

analog phone, where the phone has a

diplexer and is capable of transmitting

and receiving concurrently.

2. Mobile Station Class

The GPRS mobile is going to fall into

one of three categories called Mobile

Station Class. As we go through these

illustrations, we will refer to circuit

switched operation as GSM and pack-

et routed operation as GPRS. Figure 4.0 A GPRS Network Overlay on a GSM Network


Mobile Station Class A mobiles supports both GSM and GPRS

operation and monitoring concurrently, allowing the user to be in

a GSM call or GPRS data session and then monitor the applicable

paging channel of the other service and be able to react upon it

appropriately.

Mobile Station Class B mobiles supports GSM or GPRS operation

- but not concurrently. Once the user is in either mode, it is not

required to monitor the paging channel for the other service and

therefore, is not required to respond if a page is sent to that

mobile from the network.

Mobile Station Class C mobiles support only GPRS or GSM data.

These types of mobiles are typically modems. No provision is

made for the mobile to monitor or react to pages from either

mode. By default, GSM circuit switched data mobiles are Mobile

Station Class C mobiles.

IX. Channel Coding Schemes

Channel coding schemes are an important part of GPRS opera-

tion. Channel coding in a wireless cellularnetwork is how the dig-

ital data (either voice from the vocoder or data for non-voice appli-

cations) from the mobile or base station is formatted to deal with

the inherent problem of transmitting information across a radio

channel. Channel coding includes parity generation, convolu-

tional coding, puncturing and interleaving. These processes are

structured so that either the mobile or the base station can

receive a stream of data bits with corrupted values and still have

a high likelihood of decoding the bit stream correctly.

A good analogy is packing a fragile gift prior to shipment in the

post. You add layer upon layer of protective covering, paper, bub-

ble wrap or whatever protective medium you choose to protect

the integrity of the gift. Much as we would "protect" the gift and

the recipient "unpacks" the gift, channel coding "protects" the

data that is being transmitted to and from the mobile and base

station where it is "unpacked".

In GPRS we have four different channel coding schemes. The

reason for having multiple schemes is to provide for varying

degrees of data rates, dependent upon the Quality Of Service or

QOS requirements. QOS takes into account the mobile user's

required data rates (graphics or text) and the channel conditions

to and from the mobile. The coding used can be extensive, (CS1)

where we get the most protection for the data and correspond-

ingly lower data rates - to very little (CS4) where the data rates are

much higher, since fewer of the transmitted "bits" are used for

error correction.

Again, using our "fragile gift" analogy, the channel coding can vary

from no packaging, because the delivery address is right next

door to our home, to extensive packaging, since the package is

going around the world to some exotic location located miles

from any paved road.

GPRS channel coding schemes with their appropriate data rates

are shown in table 1.0:

Slot CombinationsChannel Coding Scheme 1 Slot 4 Slots 8 SlotsCS1 9.2 kbps 36.8 kbps 73.6 kbpsCS2 13.55 kbps 54.2 kbps 108.4 kbpsCS3 15.75 kbps 63 kbps 126 kbpsCS4 21.55 kbps 86.2 kbps 172.4 kbps

Table 1.0 GPRS Channel Coding Schemes and Associated Data Rates

X. Multi-Slot Classes

Recall from GSM that the mobile is sending uplink bursts to the

base station during its allocated time slot every frame. Each frame

contains 8 time slots, therefore allowing up to 8 users for that par-

ticular radio channel. The burst time is 577 microseconds, which

equates to a frame length of 4.615 milliseconds (8 times 577

microseconds). These frames are part of a larger multi-frame,

which in GSM is equal to 26 frames. Bursts are sent in a duplex

mode where one burst position (time slot) is assigned for mobile

communication in the downlink and uplink per frame. The uplink

is offset from the downlink by 3 time slots. Only one time slot is

used for voice communications, since the sampling rates are set

to occur every 4.615 milliseconds. See figure 6.0 for an example

of a GSM frame.

Figure 6.0 A GSM Frame

1. Downlink and Uplink Slots

With GPRS no longer tied down to using only one slot, the GPRS

mobile can handle both multiple downlink slots as well as send

multiple uplink slots. The burst structure and frame structure

remains the same as GSM, however the multiframe has been

expanded to 52 frames.

In GPRS mobile terminology, the term Multi-Slot Class refers to the

number of downlink and uplink slots the mobile can appropriate-

ly handle. For example, Multi-Slot Class 1 means that the mobile

can handle one slot on the downlink and transmit one slot on the

uplink per frame. Multi-Slot Class 10, on the other hand, provides

4 slots on the downlink and 2 on the uplink. Table 2.0 shows the

different Multi-Slot class configurations.

Where:

Rx: Rx describes the maximum number of receive times-lots that the Multi-Slot or MS can use per GSM TDMAframe.

Tx: Tx describes the maximum number of transmit times-lots that the MS can use per GSM TDMA frame.

Sum: Sum is the total number of uplink and downlink TSthat can actually be used by the MS per TDMA frame.

Multi-Slot class Maximum number of slots Minimum number of slotsRx Tx Sum Tta Ttb Tra Trb

1 1 1 2 3 2 4 22 2 1 3 3 2 3 13 2 2 3 3 2 3 14 3 1 4 3 1 3 15 2 2 4 3 1 3 16 3 2 4 3 1 3 17 3 3 4 3 1 3 18 4 1 5 3 1 2 19 3 2 5 3 1 2 110 4 2 5 3 1 2 111 4 3 5 3 1 2 112 4 4 5 2 1 2 1

Table 2.0 Multi-Slot Class Configurations

More information can be obtained by going to the standard 3GPP

TS 05.02.

2. Relationship of Downlink to Uplink Slots

When looking at table 2.0, under the heading "minimum number

of slots" the terms Tta, Ttb, Tra and Trb are seen. These terms refer

to timing relationship between the downlink and uplink slots. This

is because the mobile is periodically required to monitor channel

conditions and then report them back to the network to deter-

mine if a cell change is required, what channel coding scheme

would be best utilized, power levels, etc. The mobile needs to per-

form channel quality measurements during an idle slot and the

utilization of transmit to receive offsets allows the mobile sufficient

time to accomplish this task.

Where:

Tta: Tta relates to the time needed for the MS to performadjacent cell signal level measurement and get readyto transmit.

For type 1 MS it is the minimum number of timeslotsthat will be allowed between the end of the previoustransmit or receive TS and the next transmit TS whenadjacent cell signal level measurements are to beperformed between. It should be noted that, in prac-tice, the minimum time allowed may be reduced byamount of timing advance.

Ttb: Ttb relates to the time needed for the MS to get readyto transmit. This minimum requirement will only beused when adjacent cell power measurements arenot required by the service selected.

For type 1 MS it is the minimum number of timeslotsthat will be allowed between the end of the last pre-vious receive TS and the first next transmit TS orbetween the previous transmit TS and the next trans-mit TS when the frequency is changed in-between. Itshould be noted that, in practice, the minimum timeallowed may be reduced by the amount of the timingadvance.

Tra: Tra relates to the time needed for the MS to performadjacent cell signal level measurement and get readyto receive.

For type 1 MS it is the minimum number of timeslotsthat will be allowed between the previous transmit or

receive TS and the next receive TS when measure-ment is to be performed between.

Trb: Trb relates to the time needed for the MS to get readyto receive. This minimum requirement will only beused when adjacent cell power measurements arenot required by the service selected.

For type 1 MS it is the minimum number of timeslotsthat will be allowed between the previous transmit TSand the next receive TS or between the previousreceive TS and the next receive TS when the fre-quency is changed in-between.

XI. Testing a GPRS Mobile

1. What is the Difference Between GSM and GPRS

Tests?

GSM and GPRS utilize the same bandwidth, bit rates and modula-

tion formats. The digital modulation format is called Gaussian

Minimum Shift Keying orGMSK. The primary difference is that with

GPRS, packet data is sent on multiple slots in either the uplink or

downlink direction using different channel coding schemes as dis-

cussed earlier.

Recall from past GSM experience that sensitivity was tested using

Bit Error Rate or BER tests based on the class of bits (more on this

later). GPRS does not assign weighting to those different classes

of bits; rather it uses a similarmethod of checking sensitivity called

block error rate or BLER.

2. Transmitter Tests

Transmitters are the most tested part of either a mobile or base

station. This is due to the inherent inter-modulation products and

the spectral efficiency issues that accompany modern digital

modulation techniques and the corresponding requirement to

"co-exist" with other mobile transmitters in a given spectrum.

Here are the most common tests for a GSM digital mobile RF

transmitter:

Spurious Emissions and Spectral Measurements: Spurious sig-

nals can be defined as inter-modulation products in and out of

channel or band, and are usually caused by non-linear amplifiers,

I/Q mixer balance and spectral re-growth. These emissions can

be tracked with a good quality spectrum analyzer by looking at in-

band and out-of-band emissions.

PPoowweerr vveerrssuuss TTiimmee -- TThhee GGMMSSKK BBuurrsstt SSttrruuccttuurree:: Recall from sec-

tion IX that the GPRS mobile can send multiple bursts within one

frame on the uplink depending on its multi-slot classification. If a

mobile is capable of this, then the test equipment manufacturer

can devise tests that allow for power profile analysis of the GMSK

burst in each of the slots utilized by the mobile. Keep in mind that

the mobile will only send multiple bursts per frame on a consec-

utive basis (i.e. slot 1, slot 2, slot 3, etc…) and will not send them

on an alternate slot basis (i.e. slot 1, slot 3, slot 7, etc…).

Begin by looking at a normal GMSK burst utilized in either a GSM

or GPRS function. Figure 7.0 shows a Powerversus Time template

for a normal burst format.


Compare this to a GPRS uplink

that has two consecutive bursts

side by side as shown in figure

8.0. This would be consistent

with a mobile that conforms to

Multi-Slot Classes 3, 5, 6, 9, 10,

19, 24 per table 2.0. Notice that

the time between the two

bursts allows for variation in the

power. The mobile can either

power down, or continue trans-

mitting random data. It is up to

the designer to make this deci-

sion but it must conform to the

power versus time template to

ensure that it does not cause

interfering emissions. Most

mobiles today only support one slot uplink.

RRMMSS aanndd PPeeaakk PPhhaassee EErrrroorr:: For GMSK modulation, this checks

the modulation accuracy of the transmitter in the Mobile Under

Test. It is made on the useful part (information ordata) of the burst

and is usually averaged over a predetermined number of bursts.

Peak is the worst case measurement of the burst phase error.

FFrreeqquueennccyy EErrrroorr:: A test of the stability of the mobile's transmitter

to keep on frequency regardless of modulation format.

BBiitt TTiimmiinngg:: This test checks the accuracy of the mobile's trans-

mission timing. To accommodate for near-far effect, the network

can request that the mobile advance its burst by a predetermined

number of bits.

In GSM, this is accomplished by the BTS measuring the bit shift of

the Training Sequence Code orTSC that is found on all uplink nor-

mal bursts from the mobile to the BTS. (Remember that with GSM,

the mobile is always sending uplink bursts during a voice call).

The BTS then sends timing adjustments back to the mobile on the

Slow Associated Control Channel or SACCH message where the

mobile then adjusts the timing advance. Since the SACCH is

mapped over 102 GSM TDMA frames, the mobile receives an

update every 480 ms.

With GPRS, this process will not work, since the continuous trans-

mission of uplink bursts is no longer provided. GPRS uses a dif-

ferent method called the Continuous Timing Advance Update

Procedure. The continuous timing update procedure relies on

the Packet Timing Control Channel or PTCCH for both the uplink

and downlink. Here, the mobile is assigned to one of 16 logical

sub-channels that repeat every 416 frames on either the 12th or

38th burst. The mobile sends an access burst to the base station

and the BTS can then estimate the distance to the respective

mobile and then return the appropriate timing advance message.

Optionally, the BTS may track the shift of the incoming access

bursts and normal bursts on the PDTCH from the mobile and

determine the distance to the mobile station. Using a method

called polling, the PCU sends a polling message to the mobile and

the mobile responds with four identically formatted access bursts

during a particular uplink radio block (a block is equal to four

bursts) assigned by the PCU. The BTS meas-

ures the timing advance and forwards this to

the PCU, which responds with a correspon-

ding message to the mobile.

3. Receiver Tests:

The receiver tests are used to determine the

sensitivity of the receiver, or how low a signal

the mobile can process before it causes

excessive errors. In analog days, this was a

Signal + Noise and Distortion test or SINAD.

For digital modulation techniques, however,

we need to look at an alternative method to

test sensitivity.

Recall from past GSM experience that the bits

within the voice data are assigned different

"class" codes to determine their relative

importance in transmission and were afford-

ed more extensive convolutional coding or

"protection" based on their importance. In

GSM, we therefore tested the various BER in each of those class-

es of bits. This gave us BER1, BER2, RBER1b and RBER2 bit error

measurements. GSM also tracked Frame Erasure Rates or FER,

where the whole frame was lost.

These combined tests, along with a GSM Single-slot TCH loop (A)

test allow test equipment manufacturers to accurately test the

mobile's sensitivity.

BER compares a known good data stream (bits) transmitted to the

mobile from the test system. Typically, the mobile then "loops

back" the data stream to the test system, where the data sent is

compared to the data received. Those "bits" that do not come

back correctly are flagged as errors, hence the name "Bit Error

Rate". The measurement is expressed as a ratio of the received

"error bits" to the known good bits transmitted. This test is used

to determine the sensitivity of the mobile's receiver, where the

sensitivity is determined by reducing the power of the received

signal until a predetermined BER is reached. Faster tests will set

a predetermined pass/fail limit BER and set a particular output

level (typically around -104 dBm) and simply check the mobile

against the pass/fail parameters.

GPRS utilizes a different method to determine sensitivity of the

mobile. This is called Block Error Rate or BLER. A block is defined

as four consecutive bursts from the base station to the mobile.

GPRS mobiles that support BLER support GPRS test modes A, B

and C. BLER means the test system sends random RLC blocks of

data to the mobile which replies with ACK/NACK (acknowl-

edge/not acknowledge) responses to the test system. The ratio

of NACKs to ACKs determines the block error rate.

A GPRS BER test can also be performed if the GPRS mobile is

capable of doing a GPRS Test Mode "B" Loopback test. If the

mobile cannot support Test Mode "B", then it must be able to sup-

port GPRS Test Mode "A". Mobiles that support both modes are

known as supporting GPRS Test Mode "C".

Conclusion - GPRS Now and In The Future

This application note has introduced information about GPRS

operation. It has explained various differences between GSM cir-

cuit switched operation and GPRS packet operation. GPRS is a

fundamentally different way of handling data communications

over the air, and is the first step in packet data functionality that

sets the basis for packet data operation for next generation tech-

nologies including EGPRS (EDGE) and 3G WCDMA.

GPRS will continue to evolve. Currently manufacturers offer a lim-

ited range of mobile products that have been carefully matched

to the networks that they operate. There are a number of issues

that make GPRS mobile and network inter-operation challenging.

This requires comprehensive solutions that take the guesswork

out of GPRS mobile performance testing.

As we gain experience, many of these issues will fade away. New

and different problems will undoubtedly develop to challenge us

as data rates and features expand.

Aeroflex will continue to enhance product performance and will

be releasing more application notes in the future as GPRS tech-

nology evolves. Please check our Website for additional informa-

tion. www.aeroflex.com

Part No. 46891/907, Issue 2, 07/05

CHINA Beijing

Tel: [+86] (10) 64672716

Fax: [+86] (10) 6467 2821

CHINA Shanghai

Tel: [+86] (21) 6282 8001

Fax: [+86] (21) 62828 8002

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601

Freephone: 0800 282388

USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352

w w w . a e r o f l e x . c o m

i n f o - t e s t @ a e r o f l e x . c o m

As we are always seeking to improve our products,

the information in this document gives only a general

indication of the product capacity, performance and

suitability, none of which shall form part of any con-

tract. We reserve the right to make design changes

without notice. All trademarks are acknowledged.

Parent company Aeroflex, Inc. ©Aeroflex 2005.

Our passion for performance is defined by three

attributes represented by these three icons:

solution-minded, performance-driven and customer-focused.


ApplicationNote

During the evolution of GSM mobile there have been 2 phas-

es of implementation.

The phase of the mobile defines the RF capability of the

mobile and hence the way it is tested.

How does the “Phase” of a mobile

effect its testing strategy

Introduction

During the evolution of GSM mobile there have been 2

phases of implementation Phase 1 and Phase 2. The

phase of the mobile defines the RF capability of the

mobile and hence the way it is tested. The phase of the

mobile its transmitted to the network or test equipment as

part of the class mark.

The class mark is the information that is passed from the

mobile during the registration. This is used by the network

(or test equipment) to identify the capability of the mobile.

The rest of this article explains the differences in RF per-

formance and features between the phases of mobiles.

The following provides the details of this information cov-

ering the frequency allocations, power classes of mobiles,

power levels.

Frequency Allocations

Phase 1

Phase 1 was the initial phase of mobile and supported

P-GSM, GSM 1800 (often known as DSC) and GSM 1900 (often

known as PCS). The following outlines the frequency plans for

each of the bands

GSM Band Channel Mobile transmit and in MHz Mobile Receive and Number Basestation Receive Frequency Basestation transmitRange in MHz Frequency in MHz

P-GSM 900 1 to 124 890 + 0.2 x channel number Mobile Tx + 45 MHz(Primary Band)GSM 1800 512 to 885 1710.2+0.2 x (channel number -512) Mobile Tx + 95(DCS)GSM 1900 512 to 810 1850.2+0.2 x (channel number -512) Mobile Tx + 80(PCS)

Phase 1 frequency Allocation

Phase 2

Phase 2 added the capability for additional channels in the GSM

900 band, known as E-GSM.

GSM Band Channel Mobile transmit and in MHz Mobile Receive and Number Basestation Receive Frequency Basestation transmitRange in MHz Frequency in MHz

E- GSM 0 to 124 890 + 0.2 x channel number Mobile Tx + 45 MHz(Extended Band)

975 to 1023 890 + 0.2 x channel number-1024) Mobile Tx + 45 MHz

Additional Phase 2 - Frequency Allocation

Power Class

The power class of the mobile defines the maximum output

power level for the mobile in each of the GSM bands are shown

below :-

GSM 900

Power Class Maximum Power Level Maximum Output Power1 ---- ----2 PL2 39 dBm, 8 W3 PL3 37 dBm, 5 W4 PL4 33 dBm, 2 W5 PL5 29 dBm, 800 mW

GSM 900 Power Classes

GSM 1800

Power Class Maximum Power Level Maximum Output Power1 PL0 30 dBm, 1 W2 PL3 24 dBm, 250 mW3 PL29 36 dBm, 4 W


GSM 1900

Power Class Maximum Power Level Maximum Output Power1 PL0 30 dBm, 1 W2 PL3 24 dBm, 250 mW3 PL30 33 dBm, 2 W


Power Levels

The base station controls the mobile output power level by send-

ing a power level (as a number) which the mobile then transmits.

This is used to ensure that the optimum power level is received

by the base station and maximises the battery life. The following

are the power levels for each of the GSM bands and the standard

test specification for each of the power levels.

GSM 900

Power Level Nominal Output power Normal Specification2 39 dBm ± 23 37 dBm ± 34 35 dBm ± 35 33 dBm ± 36 31 dBm ± 37 29 dBm ± 38 27 dBm ± 39 25 dBm ± 310 23 dBm ± 311 21 dBm ± 312 19 dBm ± 313 17 dBm ± 314 15 dBm ± 315 13 dBm ± 316 11 dBm ± 5 Phase 2 only17 9 dBm ± 5 Phase 2 only18 7 dBm ± 5 Phase 2 only19 5 dBm ± 5 Phase 2 only

GSM 900 Power Levels

GSM 1800

Power Level Nominal Output power Normal Specification29 36 dBm ± 2 Phase 2 Only30 34 dBm ± 3 Phase 2 Only31 32 dBm ± 3 Phase 2 Only0 30 dBm ± 31 28 dBm ± 22 26 dBm ± 33 24 dBm ± 34 22 dBm ± 35 20 dBm ± 36 18 dBm ± 37 16 dBm ± 38 14 dBm ± 39 12 dBm ± 410 10 dBm ± 411 8 dBm ± 412 6 dBm ± 413 4 dBm ± 414 2 dBm ± 5 Phase 2 Only15 0 dBm ± 5 Phase 2 Only



GSM 1900

Power Level Nominal Output power Normal Specification30 33 dBm ±2 dB 31 32 dBm ±2 dB0 30 dBm ±3 dB 1 28 dBm ±3 dB 2 26 dBm ±3 dB 3 24 dBm ±3 dB 4 22 dBm ±3 dB 5 20 dBm ±3 dB 6 18 dBm ±3 dB 7 16 dBm ±3 dB 8 14 dBm ±3 dB 9 12 dBm ±4 dB 10 10 dBm ±4 dB11 8 dBm ±4 dB 12 6 dBm ±4 dB 13 4 dBm ±4 dB 14 2 dBm ±5 dB 15 0 dBm ±5 dB


Burst Power Profile

The power profile test ensures that the GSM burst lies within a

predefine power / time template. The power profile is not static

with changes in power level. The following shows how the major

points of the power profile changes for low power levels

GSM 900

Power burst profile definition for low power levels

Power burst profile definition for low power levels

Point A is defined as :--- 6 dBc for Power Levels 15 and higher-4 dBc for Power Level 16;-2 dBc for power level 17;-1 dBc for power level controls levels 18 and 19

Point B is defined as :--30 dBc (dB with respect to the carrier) or -17 dBm(an absolute value in dB with respect to 1 mW),whichever is the higher.

The impact of this is that the lower limit of the mask can change

dependent upon the actual power measured . The following are 2

examples :

Example 1

A mobile is set to PL19 (5 dBm) and has no error, the absolute

measured power will be +5 dBm hence the -30 dBc power level

equivalent to an absolute power -25 dBm

(+5 - 30), and the absolute value of -17 dBm would be equivalent

to -22 dBc (+5 - 17). Therefore the as -22 dBc is greater than -30

dBc the absolute

Example of power burst calculation for power level 19

Power Level 19 with No error then the -17 dBm limit is used

Example 2

If the mobile is set to PL 16 (i.e. +11 dBm) and has a error of + 5

dB (ie transmitting +16 dBm), then the -30 dBc value would be

equivalent to -14 dBm, and the -17 dBm value would be equiva-

lent to -33 dBc. Therefore the -30 dBc limit will be used.

Example of power burst calculation for power level 16

Power Level 16 with a +5 dB error then the -30 dBc limit is used

GSM 1800/ 1900

The Power time template for GSM 1800 and 1900

The Power time template for GSM 1800 and 1900

The specification for GSM 1800 and 1900

Point A is defined as :--4 dBc for power level 11,

-2 dBc for power level 12,-1 dBc for power levels 13,14 and 15

Point B is defined as-30 dBc or -20 dBm, whichever is the higher.

Part No. 46891/855, Issue 2, 07/05

CHINA Beijing

Tel: [+86] (10) 64672716

Fax: [+86] (10) 6467 2821

CHINA Shanghai

Tel: [+86] (21) 6282 8001

Fax: [+86] (21) 62828 8002

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601


USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352














ApplicationNote

Service center throughput is affected by more than just test

times - it involves the entire workshop process.

Administration, data management, de-skilling of repetitive

tasks, test setup times, test equipment user interfaces, test

software management as well as test times on equipment can

all significantly affect workshop efficiency.

Maximizing GSM Mobile Repair

Centre Throughtput

Maximizing GSM Mobile Repair Center through-

put

Service center throughput is affected by more than just

test times - it involves the entire workshop process.

Administration, data management, de-skilling of repetitive

tasks, test setup times, test equipment user interfaces,

test software management as well as test times on equip-

ment can all significantly affect workshop efficiency.

This application note outlines and quantifies the impact of

the IFR 2935 and PhoneTest system on the complete

workshop.

Issues surrounding the repair center process

Let us briefly review the repair center process.

Booking in

At book-in, an operator enters any details associated with the job,

such as Customer Name, IMEI, the customer-reported fault, and

Warranty Status into the center's business system.

Pre-screen

At pre-screen, the operator sorts mobiles into those with and

without electrical problems. In making the decision, the operator

refers to the job ticket for the customer-reported faults. However,

in a busy workshop, there is a time penalty associated with

accessing any paper-based information.

If the mobile has an electrical fault, then the pre-screen operator

carries out an initial test run to assist in diagnosis. The mobile is

connected to a test set either via a simple RF cable, coupler or

dedicated jig. The operator selects the test program (usually a

"brief" one), and specifies the allowance that the test set is to

make for the RF loss within the connection mechanism.

He then runs the test and waits for it to complete.

If pre-screen and repair are carried out as separate functions, the

operatorwaits for a results printout to send, with the mobile, to the

technician who will diagnose the fault and implement the repair.

Repair & Alignment

To accelerate diagnosis, a technician makes use of every available

piece of information. These include the customer-reported fault,

the results of previous tests, past experience of similar symptoms

on similar mobiles and any 'Top 10' fault checklist that the work-

shop has created for itself. It is simply not practical to present all

this information to a technician - even if it could fit on a busy repair

bench, gathering it together takes too long.

Having made a diagnosis, the technician executes the repair.

However, repair centers and/or mobile manufacturers require the

repair to be documented. This involves noting, in a standardized

format, symptoms, faults found, action taken and components

used.

If a mobile's RF stages have been disturbed by the repair then it

has to be re-aligned. During alignment, the mobile has an RFcon-

nection to the test set, but is usually under the control of the man-

ufacturer's software. The problem is that each of the many types

of mobile has a different 10 to 15 step procedure, with each step

requiring the setting of 4 or 5 test set parameters. Although the

process is straightforward, the settings - which have to be precise

- are repetitive, error-prone, and slow, resulting in a typical align-

ment taking several minutes. Mistakes are inevitable but, as some

mistakes will not necessarily be detected during alignment. This

can allow the phone to be misaligned, a circumstance which, at

best, will not be detected until final test.

Final test

After the mobile has been repaired and aligned, most workshops

subject it to a final test.

As with the pre-screening test, the operator has to select the test

program and RF insertion loss, start the test and wait for it to com-

plete. Usually, final test differs from pre-screen in that the nature

of the final test is often dictated by the customer - centers that

have contracts with large retail organizations are familiar with this.

Another difference is that a final test tends to be long.

A third difference is that some repair centers are contracted to

have an element of traceability. In this case, there is a need to

have test results printed and filed.

Booking out

After the workshop stage, the mobile is booked out and shipped.

The center's business system raises the warranty claim, for which

it requires the 'documentation' that the repairer completed earli-

er on. If the documentation was done on paper, then the details

need to be transcribed into the system.

The Aeroflex product

Aeroflex's solution to the issues discussed above lies not simply

in a fast and stable GSM test set, but also in the software system

that accompanies it. The 2935 test set is fast, stable and accurate


but it is the PhoneTest software that gives the efficiency improve-

ments that maximize throughput.

The PhoneTest system combines:

• PhoneTest itself, in effect the user interface for the 2935, tailored

to suit the role in hand.

• A workshop management option, comprising:

• PhoneTest-Manager, Aeroflex's system for recording and archiv-

ing the entire test and repair history of every mobile.

• PhoneTest-Exchange, providing configurable links into the user's

own databases, whether for analyzing repairs, or for transferring

commercial information.

• PhoneTest-Repair, which can be configured to simplify capture

of repair data.

The impact of 2935 & PhoneTest on the repair center

process

To best understand the effect of PhoneTest and the 2935, let us

examine how it works in a repair center. As the discussion devel-

ops, an estimate is made of the amount of time that can be saved

during each process.

Pre-screen

The classification into electrical and non-electrical faults (or,

indeed no fault found) rests, in part, on knowing key information

such as the customer-reported fault. Rather than the operator

accessing information from job tickets, PhoneTest speeds-up the

process by automatically accessing the fault information and war-

ranty status, directly from the business system.

Estimated time saving: 15 sec.

Regarding initial test, the time taken by the operator to select the

right test program and RF insertion loss can be removed by con-

figuring PhoneTest to automate the process. This reduces opera-

tor involvement as it’s simple one-button operation.


The 2935 is arguably the fastest test set aimed at repair center

use. Although test speed is significant, just as important to maxi-

mizing throughput is the ability to create and fully customize test

sequences so that they test only the relevant parameters. This is

particularly important since pre-screen testing needs to be kept

very short.


The time taken to print pre-screen test results can be completely

removed simply by logging them to PhoneTest-Manager instead.

Estimated time saving: 1 min.

Repair and alignment

There are many ways that PhoneTest can help in diagnosing the

repair. The system can:

• Present the fault, as reported by the customer, direct from the

business system.

• Access the results of the pre-screen test. Incidentally, the oper-

ator can also see the full test-and-repair results of any other vis-

its that the mobile has made to the Repair Center.

• Compare the pre-screen results with the knowledge base that

PhoneTest creates of previous tests and repairs on mobiles of

the same type, and then automatically present the technician

with probable causes based, literally, on past experience.

• Display any product-dependent guidance checklist that the

Repair Center, or the manufacturer, wishes the repairer to have

access to. This may be as simple as displaying the manufactur-

er's "Top 10" fault list. At the other extreme, PhoneTest may dis-

play the manufacturer's repair documentation, circuits and parts

lists.


To assist with post-repair 'paperwork', PhoneTest-Repair can be

configured to allow the technician to pick, from simple product-

or manufacturer-dependent lists, details such as symptoms, fault,

action taken and components replaced. Given this information,

combined with the Job Details on the center's business system,

PhoneTest can either create a warranty claim directly or, if pre-

ferred, it can pass the details up to the business system for it to

process.


At the alignment stage, PhoneTest has a special feature - 'TCH

Test Mode stores' - that can be configured to reduce the literally

hundreds of possible test set settings to a simple two-step

process: first select the type of mobile in question, and then pick

the next step in the manufacturer's process. Not only speeding-

up the process, it reduces the risk of errors (some of which would

be uncovered later on at final test).

Estimated time saving: 2 min 30 sec.

Final test.

To be effective, final test needs to be comprehensive, but to

reduce costs, it also needs to be simple, fast and error-free. As at

pre-screen, PhoneTest can be set to automatically load the cor-

rect test sequence and insertion losses for the mobile in question.


In addition to the 2935 being fast, users can configure tests to

match their own, or their customers', requirements. With multiple

2935s at one final test workstation, users can increase through-

put by testing two or more mobiles simultaneously from a single

PC.

Estimated time saving: 1 min 50 sec.

Where traceability is required, as opposed to waiting for results to

be printed (and then having to file them), users automatically log

results centrally to the PhoneTest-Manager database.


In order to see a mobile's full test-and-repair history, it is only nec-

essary to identify the mobile to PhoneTest. This can be done by

registering it on a 2935, or by entering (through the PC keyboard

or using a bar code reader) the mobile's IMEI. The latter does not

require a 2935 to be present; it can be done on the supervisor's

PC, the QC manager's PC or any PC anywhere on the center's net-

work.

In addition, when a mobile passes final test, PhoneTest can auto-

matically print a customized certificate for returning to the end-

user. This is seen as a way of building customer satisfaction.

Book out

At booking out, the dispatch operator closes the job on the

Center's business system. PhoneTest system can assist by

already having:

• The business system that the mobile had already passed final

test (and thus ensuring that mobiles cannot be dispatched inap-

propriately).

• Sent up the repair details (symptoms, faults, actions, parts, etc)

so that the business system can make the warranty claim auto-

matically.

Helping the workshop supervisor

The workshop test supervisor has a key role. Any problems, what-

ever the magnitude, with any mobiles, equipment or methods are

referred to him for 'instant resolution'.

Included in his responsibilities is the creation of tests. As well as

PhoneTest being supplied with a range of ready-to-run sequences

for various depths of test and band combinations, PhoneTest's

script wizard guides you through the creation of custom

sequences to meet particular needs. And, as the scripts are

saved as simple text files, they can be further fine-tuned with any

editor.

Once scripts, stores, or any other aspect of a testset's settings are

ready for release to the workshop, it wastes time to spend a typi-

cal 5 minutes per workstation in, waiting for the operator to finish

their current job and installing the data. It is far quicker to simply

put the information on a central server, and have every worksta-

tion access it from there. As PhoneTest is Windows based, work-

stations can be networked with virtually any technology that the

userwishes. In addition, if a centerhas multiple sites, and the sites

are already networked, centralized distribution becomes even

more cost effective.

Incidentally, because 2935s are controlled from their PCs, central

distribution of scripts, stores and settings is applicable even when

a test set is away for calibration. When the 2935 returns, there is

no risk of forgetting to update it.

As the supervisor is treated as the all-round expert for every prob-

lem imaginable, he needs as much help as he can get. In addi-

tion to conventional training and telephone support, Aeroflex can,

if required, provide dial-in support direct to a specific workstation.

Conversely, the supervisor can make use of simple inter-PC net-

working to access workstations at remote sites.

Throughput implications

The above estimates yield a total savings of a little over 10 min-

utes per mobile. For a repair center with a throughput of 500

mobiles per day a 10 min saving per mobile is equivalent to over

400 operator hours per week, thus giving significant scope for

increasing capacity.

For a repair center with 20 test sets, whenever the supervisor

updates (for example) a script, a saving of 5 min per test set

means a saving on his time of 1.5 hours.

There are additional benefits:

• Optimizing the use of skilled technicians.

• Reducing errors, thus increasing both throughput and customer

satisfaction.

• Traceability, without the overheads.

• Extra customer satisfaction through automatically printing cus-

tom "Pass certificates".

• If required, dial-in support.

Implementation scheme

PhoneTest does far more than simply provide a user interface to

the 2935 test set. However, it is unnecessary to implement all the

features discussed above, nor is it necessary to implement the

final system from "day one". Consider the following typical (but by

no means mandatory) scenario, any stages of which can be

implemented or omitted as appropriate:

1. Start with the 2935s controlled by PhoneTest. Immediately, the

user gets the benefit of a fast, stable test solution, with a sim-

ple-to-use PC-based user interface, and full flexibility in the cov-

erage and flow of test sequences.


2. Progressively configure PhoneTest to yield the benefits of auto-

matic test sequence and insertion loss selection, faster align-

ment throughput via the TCH Test Mode stores, and certificate

printing. In addition, the PCs can be networked for immediate

distribution of test sequences, stores and settings and, if

required, use readily-available PC packages for remote access.

3. Use PhoneTest-Manager to give virtually cost-free traceability.

4. Configure PhoneTest-Exchange for accessing and, if required,

updating information held on the center's business system.

5. Finally, configure PhoneTest-Repair to capture repair details

(and virtually any other operator-entered information that the

center requires). This information can, if necessary, be passed

on via PhoneTest-Exchange to the business system for prepa-

ration of the warranty claim.

Some real examples

1. A 2935 user with multiple repair centers in several countries,

operating hundreds of 2935s.

• The main requirement from the 2935 is for a fast and reliable

test set. However, most benefit comes from PhoneTest's unique

TCH Test Mode stores facility which, he reports, "doubles align-

ment throughput".

• Workstations are networked for rapid distribution of stores and

setups.

2. A repair center with one workshop, operating about fifty 2935s.

• Organized into separate repair/align and final test stages.

• Final test has 2 test sets per operator. If it passes, automatically

prints a certificate for returning with the mobile to the end user.

• Workstations are networked for instant, centralized update of

scripts, stores and settings.

• Uses PhoneTest-Manager to meet contractual commitment to a

major retail chain to maintain traceability. Currently have three

years of records on-line. If a mobile is returned, the complete

history is retrieved automatically.

3. A repair center with several workshops, operating about twen-

ty-five 2935s.

• Existing inter-site networking means that scripts, stores and set-

tings are immediately available everywhere within the organiza-

tion.

• All workstations log results to central PhoneTest-Manager data-

base - provides full traceability.

• Have configured PhoneTest-Repair for their own particular repair

detail entry requirements.

• Use PhoneTest-Exchange to automatically pass, on a selective

basis, repair details and job status update to their business sys-

tem database.

Appendix

IT issues within a PhoneTest system

The main text of this application note describes how PhoneTest

systems range in complexity from:

• one PC running PhoneTest simply as the user interface for a

2935 testset, through to:

• tens of workstations, across multiple sites, with central result log-

ging, full traceability, and real-time access to the corporate busi-

ness system, all managed by a single supervisor from a single

desk.

This appendix provides a brief overview of the IT (Information

Technology) issues applicable to a PhoneTest system.

Stand-alone workstations

A single workstation comprises a 2935, a PC, and PhoneTest soft-

ware. The 2935 connects to one of the PC's COM ports, using

the RS232 cable supplied with the testset.

Multiple 2935s can be connected to the same PC, so long as the

latter has sufficient COM ports with non-overlapping IRQs.

Interconnected workstations

Instant distribution of test sequences, stores, settings, etc, is

gained simply by networking together the workstations PCs, the

supervisor's PC and, optionally, a server.

PhoneTest can use any networking scheme that is supported by

the PC's WindowsTM operating system. PhoneTest accesses test

sequences, stores, settings, etc as simple files, whose paths are

configurable within PhoneTest. Therefore, it is necessary only to

set-up a master directory structure physically on one PC (such as

the supervisor's or, ideally, an "always on" server) and, on each

workstation PC, create links to the structure via virtual drives.

Central result logging

Test and repair results can be logged either:

• to Aeroflex's unique PPhhoonneeTTeesstt--MMaannaaggeerr system, or

• via PPhhoonneeTTeesstt--EExxcchhaannggee, to the user's own conventional rreellaa--

ttiioonnaall ddaattaabbaassee ssyysstteemm,

or to both.

PPhhoonneeTTeesstt--MMaannaaggeerr does not use a relational database. It is a

hierarchical database, optimized for efficient storage and retrieval

of multiple variable-length records indexed by a single key, as is

required when dealing with test and repair histories of mobile

phones (where the key is the mobile's IMEI). One distinct benefit

of using PhoneTest-Manager is that, other than running a simple

Windows-based server process on one of the PCs (preferably the

database's own server), it requires no user configuration.

Conversely, for PhoneTest to log results to the user's own rreellaa--

ttiioonnaall ddaattaabbaassee ssyysstteemm, provision of the database, configuration of

its tables, fields, reporting and data presentation facilities are all

under the control of the user. Although requiring greater effort in

terms of initial system set-up, as everything is under the user's

direct control, maximum flexibility is gained in terms of report

generation and data presentation.

Logging to PhoneTest-Manager

On the technical side, PhoneTest logs results to PhoneTest-

Manager via a simple server process that, typically, runs on the

central PC that hosts the database. The communication mecha-

nism uses TCP/IP. In case of failure of either the central server or

the network, PhoneTest workstations automatically buffer results

locally and, when the system is re-established, log buffered

results to the server.

Exchanging information with the user's own databases

Assuming that the user's own relational databases have 32-bit

ODBC drivers, then PhoneTest can exchange information with

them. The exchange of data (test and repair results, job details,

etc) is controlled by a simple SQL (Structured Query Language)

subset, using statements such as SELECT, UPDATE and INSERT.

As the user will typically be using different databases for com-

mercial and technical data, PhoneTest can use different DSNs

(Data Source Names) for the various business system and results

transactions.

To help with getting a PhoneTest-Exchange system "up and run-

ning", PhoneTest includes:

• two sample Microsoft Access databases (one for results, one for

job details,

• an Access 32-bit ODBC driver, and an

• SQL file

all pre-configured for:

• retrieving job details (customername, warranty status, customer-

reported fault, etc), and

• storing and retrieving full test-and-repair histories to and from a

results database.

Part No. 46891/903, Issue 2, 07/05

CHINA Beijing

Tel: [+86] (10) 64672716

Fax: [+86] (10) 6467 2821

CHINA Shanghai

Tel: [+86] (21) 6282 8001

Fax: [+86] (21) 62828 8002

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601


USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352














ApplicationNote

With the recent explosion of digital cellular service and the

acquisitions and mergers of wireless cellular operators, there

has been a sudden resurgence in the

concept of the "world phone."

Testing GAIT Phase One Mobiles

Nothing is ever static in the world of digital cellular and

wireless communications. With the recent explosion of

digital cellular service and the acquisitions and mergers of

wireless cellular operators, there has been a sudden

resurgence in the concept of the "world phone." Since

450 million subscribers worldwide are using GSM (Global

System for Mobility) technology, you can count on GSM to

be the cornerstone of any "world phone" development.

Accordingly, the GAIT (GSM - ANSI-136 Interoperability

Team) phone features GSM as part of its core technology,

along with ANSI-136 (American National Standards

Institute) or TDMA (Time Division Multiple Access) tech-

nology.

In the US there are significant driving factors that are

pushing the development of GAIT mobiles. Cingular, the

company formed by the combination of the Southwestern

Bell and Bell South wireless groups, and AT&T are the two

predominant TDMA carriers in the US. These operators either

own or are moving to GSM technology and want to promote the

ability to use one mobile on both GSM and TDMA networks while

also allowing for overseas roaming.

Cingular's network features both GSM and TDMA technologies

with the core technology being TDMA. They have also stated that

they are going to deploy GPRS (General Packet Radio Service) and

EDGE (Enhanced Data for GSM Evolution) data services. With the

two operations coming together, there has been a driving force to

combine the two technologies to allow the operator to provide

"seamless" coverage for their customers without forcing them to

a competitor’s network.

AT&T Wireless has a similar, although slightly different situation.

While Cingular is faced with the tasks of integrating existing sys-

tems and future GSM based data services, AT&T is developing an

entirely new GSM network to support their new direction into

GSM/GPRS/EDGE technologies. Their requirement is similar,

however, in that they need to provide "seamless" coverage for

their customer base as well. By capturing the customer on their

networks, the operators retain customer revenue and boost cus-

tomer satisfaction thereby reducing customer turn over.

What is GAIT Phase One?

GAIT Phase One is the integration of both TDMA and GSM voice

and data technology into one handset with the ability to select and

operate on either a TDMA or GSM network. TDMA capabilities

offer the ability to roam on AMPS (Advanced Mobile Phone

Service) networks, since this is an integral part of the ANSI-136

feature set.

GAIT technology is being developed by the GSM ANSI-136

Interoperability Team headed by Cameron Coursey of Cingular.

This group operates under the banner of the TDMA users’ group

known as the Universal Wireless Communications Consortium

and the GSM Alliance.

The GAIT Phase One phone will provide the ability for the user to

select and then use eithera GSM orTDMA network as determined

by the operator. In this first rendition, GAIT mobiles will also allow

roaming on other networks as defined by agreements between

the user's operator and other wireless service providers. GAIT

mobiles will not allow intra-system hand-offs or hand-overs.

Simply stated, the mobile will know who its primary operator is

and based off an internal look up table, called the NSDB (Network

Selection Database), the mobile will know whether to go to either

a TDMA or GSM network. This will be based on a number of

parameters programmed into the mobiles NSDB which resides

on the revised SIM (Subscriber Identity Module) card adopted for

the GAIT mobile.

An additional feature of GAIT mobiles will allow the user to send

and receive SMS messages using a feature called GHOST (GSM

Hosted SMS Teleservices). More on that feature later.

So what do we test on a GAIT mobile?

Testing a GAIT mobile is really like testing separate GSM and

TDMA handsets. There is some additional GAIT related testing

that is required, but that is limited to just a few specific tests. Let's

look at the RF (Radio Frequency) performance side first.

RF Parametric Testing of GAIT Mobiles

As with any mobile phone, the requirements for testing the RF

parametrics are still valid. However, with the GAIT phone, you are

required to test for both the GSM and TDMA RF performance,

depending on how in-depth your testing requirements are. For

review, let's look at the two technologies and the typical associat-

ed RF performance test parameters.

GSM Mobile RF Test Parameters

Transmitter Parameters:

PPoowweerr LLeevveell:: This test measures the mobile transmitter power out-

put level while the mobile is in conversation on the TCH (Traffic

Channel (GSM)). The pass limits are different for different power

levels. The limits for each test are determined by the different

power levels and classes for GSM 900, 1800 and 1900 perform-

ance.

RRMMSS PPhhaassee EErrrroorr:: This test checks the accuracy of the phase

modulation of the transmitter in the Mobile UnderTest. It is made

on the useful part (information or data) of the burst and is usually

averaged over a predetermined number of bursts.

PPeeaakk PPhhaassee EErrrroorr:: This test again checks the modulation accura-

cy of the transmitter in the mobile under test. It is made on the

useful part of the burst over a predetermined number of bursts.

This is the worst case measurement of the phase error.


to keep on frequency.

BBiitt TTiimmiinngg:: This test checks the accuracy of the mobile's trans-

mission timing.

PPoowweerr PPrrooffiillee CCoonnffoorrmmaannccee:: The ability of the mobile to control

ramping power, power over time and shut down power within a

defined power-time profile mask. See figure 1.0 for example.

Figure 1.0 - GSM Power Burst Profile - Normal Burst

Receiver Parameters:

BBEERR11 ((BBiitt EErrrroorr RRaattee)):: This test checks the mobile receiver's bit

error rate for Class 1 bits (which are subject to error correction by

the mobile).

BBEERR22:: This test checks the mobile receiver's bit error rate for Class

2 bits (which are not error corrected).


RRBBEERR11bb ((RReessiidduuaall BBiitt EErrrroorr RRaattee)):: This test checks the mobile

receiver's residual bit error rate for Class 1b bits, that is, the bit

error rate in those frames that have not been erased by the

mobile.

RRBBEERR22:: This test checks the mobile receiver's residual bit error

rate for Class 2 bits (which are not error corrected).

FFEERR ((FFrraammee EErraassuurree RRaattee)):: This is the measurement of the per-

centage of speech frames that are so badly corrupted that they

fail their Class 1 parity check and are erased.

ANSI-136 (TDMA) Mobile RF Test Parameters

Transmitter Parameters:

PPoowweerr LLeevveell:: This test measures the mobile transmitter power out-

put level while the mobile is in conversation on the DTC (Digital

Traffic Channel). The pass limits are different for different power

levels. The limits for each test are determined by the different

power levels and classes forTDMA 800 and 1900 performance.

EErrrroorr VVeeccttoorr MMaaggnniittuuddee:: This test checks the modulation accura-

cy of the transmitter in the mobile under test. It is derived from

the plot of the symbol on an I/Q quadrant map to determine the

magnitude of the error vector. See figure 2.0.

Figure 2.0 - EVM (Error Vector Magnitude) where vector "A" is

the optimal path and "B" is the actual path. The EVM (vector

"C") is measured from the ideal point (1) to the actual landing

point (2) of the symbol and is expressed as the ratio of C to A.


to keep on frequency.

AAddjjaacceenntt CChhaannnneell PPoowweerr:: This tests the mobile transmitters spec-

tral performance by measuring transmitted power in the adjacent,

alternate and second alternate channels (offset 30 kHz, 60 kHz

and 90 kHz from the center frequency). See figure 3.0.

Figure 3.0 - Adjacent Channel Power Mask for TDMA ANSI-136

30 kHz channels for mobile phone digital modulation

Receiver Parameters:

BBEERR:: This test checks the mobile receiver's bit error rate using a

pseudo-random bit stream looped back through the mobile's

transceiver or, as an alternative method, using the actual reported

BER from the phone at a low signal level, typically -110 dBm.

Depending on the type of testing required the testing require-

ments and accuracy will vary. In an engineering operation, you

will do both GSM and TDMA full RF parametric analysis. In a serv-

ice environment, that could change to just a simplified test to trou-

bleshoot the RF circuitry. The level of functionality and accuracy

is dependent upon your specific need.

Protocol Testing - what's changed?

Protocol testing is an important part of the mobile phone test

process, especially if you are in the verification and validation busi-

ness for the testing of new mobile phone types. GSM and TDMA

protocol test systems have been around for quite a while and for

the purposes of this review, we won't go into the vast amounts of

protocol testing required.

But, what's changed for GAIT mobile testing? In testing a GAIT

mobile you need to add two different tests to the conventional

TDMA or GSM protocol tests. These tests are:

1. Test of the Network Selection Database or NSDB. The network

selection algorithm determines the mobile's ability to locate and

select the proper network for that phones primary operator. For

example, if you are a Cingular customer, your GAIT phone would

want to stay on a Cingular network. This is good for Cingular since

they control the quality of the call and it is good for you since you

won't have to pay additional roaming charges by using some

other operator’s network. As a test professional, you will want to

test the network selection algorithms for proper functioning.

2. GSM Hosted SMS Teleservice. Also known as GHOST. The

GAIT mobiles use GSM SMS for messaging and over-the-air pro-

gramming in both GSM and TDMA modes. In TDMA mode, the

GSM SMS PDU (Protocol Data Unit) as defined in GSM 03.40 is

tunneled across the ANSI-136 air interface. Therefore, the mobile

has to be able to process a standard GSM SMS message while

operating in both types of networks.

GAIT Modes

Before going forward, let's define the different modes in which a

GAIT mobile operates. These modes are simply operating states

that the mobile may operate in, depending upon the type of home

network. There are four different modes for the GAIT mobile as

defined below:

GGSSMM NNaattiivvee MMooddee:: This is where the mobile is homed to a GSM

network and is operating on a GSM network. In this mode, the

mobile is able to perform standard GSM type functions including

SMS and Data services (circuit switched at 9600 or 14400 bps

and packet switched service if it is offered).

AANNSSII--113366 NNaattiivvee MMooddee:: This is where the mobile is homed to an

ANSI-136 network and is operating on an ANSI-136 network. In

this mode, the mobile is able to perform standard ANSI-136 func-

tions including text messaging using GHOST SMS and Data serv-

ices (circuit switched at 9600 bps).

GGSSMM FFoorreeiiggnn MMooddee:: This is where the mobile is homed to an

ANSI-136 network but is operating on a GSM network. At this

time, the GSM portion of the phone is operational and the net-

work's Interworking and Interoperability Function (IIF) handles the

call delivery and SMS routing from the ANSI-136 network to the

serving GSM network.

AANNSSII--113366 FFoorreeiiggnn MMooddee: This is where a GSM native subscriber

is accessing an ANSI-136 network. Again, routing of calls and

SMS is handled by the network's Interworking and Interoperability

Function.

Testing the NSDB - what's involved?

So what is involved in testing the Network Selection Database?

Network selection is the mobile's ability to acquire the correct

network and to obtain the best service in a given geographic area.

The mobile looks at the stored information on the SIM card, as

defined by the home service provider, for the information to

determine which networks the phone can access. This can

include GSM, ANSI-136 and AMPS networks. For GAIT mobiles,

the network selection algorithm is a combination of the ANSI-136

Intelligent Roaming Algorithm and GSM's network selection algo-

rithm.

Tables 1 and 2 show the service provider categories for ANSI-

136/AMPS and GSM networks, respectively. In ANSI-136, the

Home service provider is defined by the Home SID (System

Identity) or SOC (System Operator Code). Partner, Favored and

Forbidden service providers are defined by SID and SOC lists that

are part of the Intelligent Roaming Database. If the SID or SOC of

a network is not included in any of these databases, then it is con-

sidered a Neutral service provider. In GSM, the Home service

provider is defined by the MCC (Mobile Country Code) and MNC

(Mobile Network Code) that are part of the mobile station's IMSI

(International Mobile Subscriber Identity). A Preferred GSM serv-

ice provider is defined as one whose MCC/MNC combinations

are stored in a Preferred PLMN (Public Land Mobile Network) List.

Similarly, a Forbidden GSM service provider is defined as one

whose MCC/MNC combinations are stored in a Forbidden PLMN

List. A Neutral GSM service provider is one whose MCC/MNC is

not part of the mobile's IMSI or one of the PLMN Lists.

Service Provider Category Action Taken

Home Select Immediately, no background triggered scan-ningPartner No background triggered scanningFavored Select Best Available, use background scanning for

selection of better Service ProviderNeutral Use background scanning for selection of a better

Service ProviderForbidden Emergency Calls only

Table 1 - ANSI-136 Intelligent Roaming Service Provider

Categories

Service Provider Category Action Taken

Home Select ImmediatelyPreferred Select Best Available, use background scanning for

selection of better Service ProviderForbidden Emergency Calls only

Table 2 - GSM Network Selection Service Provider Categories

Table 3 shows the priority used by the GAIT mobile station to

obtain service on different networks, based on a PPI (Protocol

Priority Indicator) setting. For example, if the PPI indicates ANSI-

136 Preferred then the mobile station follows the priority shown

in the left-hand column when searching for service. Suppose that

the GAIT mobile station scans for service and determines that

there are two networks available for selection, an ANSI-136

Neutral system and a GSM Neutral system. If the PPI is set to

ANSI-136 Preferred, then the mobile selects the ANSI-136 Neutral

system. Conversely, if the PPI is set to GSM Preferred, then the

mobile selects the GSM Neutral system.

ANSI-136/GSM GSM/ANSI-136(ANSI-136 Preferred) (GSM Preferred)

ANSI-136 Home GSM HomeANSI-136 Partner ANSI-136 HomeGSM Home ANSI-136 Partner

ANSI-136 Favored GSM PreferredGSM Preferred ANSI-136 FavoredANSI-136 Neutral GSM NeutralGSM Neutral ANSI-136 Neutral

ANSI-136 Forbidden GSM ForbiddenGSM Forbidden ANSI-136 Forbidden

Table 3 - Combined GAIT Network Selection Process

For AMPS and ANSI-136 operation, the mobile looks at the SID

and the SOC broadcast on the network's DCCH (Digital Control

Channel) (Analog Control Channel (ACC) forAMPS networks). For

GSM operation, the mobile looks at the MCC and MNC found on

the GSM network's BCCH (Broadcast Control Channel).

Depending on how the NSDB is configured and the correspon-

ding signal quality of the base station's DCCH or BCCH, the

mobile will determine what is the best network to access.

A test solution for GAIT should offer 800 MHz and 1900 MHz

operation for ANSI-136 and 800 MHz, 900 MHz, 1800 MHz and

1900 MHZ GSM operation. (800 MHz will be required with the

upcoming GSM deployments in the US.) It should offer the abili-

ty to select all channels within these bands and set the channel at

a particular power level. It should also provide the ability to select

and set SID and SOC information for ANSI-136 operation and

MCC and MNC (3 digit for North American operation) for GSM.


Putting together the Test System

The test system will require the use of the IFR 2935 GSM Test

Head and the IFR 1900-5 CSA TDMA test system along with a

computer system. The computer system will require two com

ports, one to control each of the instruments. In addition, the sys-

tem will need to be running Windows 98, 2000 or NT 4.0.

Connect the 2935 using a RS-232 Null Modem cable and the

1900-5 using a standard RS-232 cable.

Figure 3.1 shows the system configuration

Setting Up a GSM Test Process

Figures 4.0 through 4.3 show how the IFR 2935 GSM test system

can be set up to test the various parameters we've discussed so

far. Figure 4.0 shows the initial setups for power measurements,

frequency error, timing error and phase error measurements.

Figure 4.1shows band selection, control and traffic channel setup,

timeslot designation, power level settings, timing advanced set-

tings and RF generator level. In addition, the ability to set RF Cable

Gain and Loss is very important for accurate power measure-

ments. Additional test parameters are shown for registration time

out and protocol time out (loss of communications).

Figure 4.0 - GSM Parametric Test Setup

Figure 4.1 - Additional GSM Test Setup

Figure 4.2 - GSM Receiver BER/FER Test SetUp

Figure 4.3 - GSM Network Selection SetUp using

MCC and MNC

Figure 4.2 shows the setup limits and samples for the various BER

tests for GSM testing, along with the FER test parameters. The

RxLEV is the mobile's measurement of signal level it is receiving

and RxQUAL is the phone’s evaluation of the signal received

expressed in a numeric response that gives a relative indication

of the quality of the signal.

Figure 4.3 shows the network setup parameters for the mobile

country code, the mobile network code and additional network

parameters. Also, the MNC should be able to handle the 3 digit

MNC that is used on the North American GSM networks, not just

the two digit MNC used in the rest of the world. This capability is

critical to testing GAIT mobiles.

Figure 4.4 - GSM Full Power Profile

Figure 4.4 shows the burst profile for measuring the GSM power

burst to ensure that it meets the specification. Markers are nice

to allow for timing measurements within the burst profile. Figure

4.5 shows how these capabilities are displayed for easy reference

of the mobile’s performance.

Figure 4.5 - IFR 2935 GSM Test Display

Setting Up an ANSI-136 Test Process

Setting up an ANSI-136 test is very similar to the GSM test forpara-

metric tests. Figure 5.0 shows the GAIT Phone Test parametric

display for ANSI-136 phones. As you can see, you can test EVM,

Power, Frequency Error and Origin Offset, along with different

time slots, power levels and timing advance. You also have the

ability to perform multiple hand-off tests, both in band and across

band. Figure 5.1 shows the Adjacent Channel Power Mask for

ANSI-136 digital power measurements.


Figure 5.0 - Digital ANSI-136 parametric measurements

Figure 5.1 - Adjacent Channel Power Measurements (ACPM)

Figures 5.2 and 5.3 show the setup parameters for both the ana-

log and digital settings. The SID and SOC parameters are what

the NSDB uses for selection of the proper network. In addition,

you have the ability to select Analog Control Channel number

(ACC) and Digital Control Channel number (DCCH). In the digital

mode, you have the ability to select either the 800 MHz cellular

band or the 1900 MHz PCS band. RF levels can also be set.

Figure 5.2 - Analog Set Up Configuration

Figure 5.3 - ANSI-136 Digital SetUp Configuration

Testing GHOST

Testing GHOST functionality in a mobile requires the ability to sim-

ulate network functions. The GSM Hosted SMS Teleservice func-

tion is standardized on the GSM SMS Protocol Data Unit (PDU) as

specified in GSM 03.40. Handling this PDU is the primary con-

cern of testing the SMS functionality of the mobile.

Figure 6.0 shows how a mobile terminated GHOST message is

generated from the network IIF through the ANSI-41D network

(the mobile network that supports TDMA) and onto the ANSI-136

air interface for delivery from the base station to the GAIT mobile.

Of course, this is for a mobile that is operating in an ANSI-136

Native Mode orANSI-136 Foreign Mode. A mobile operating in a

GSM Native Mode or GSM Foreign Mode would just use the stan-

dard GSM SMS delivery procedures.

As you can see, the network has built a SMDPP (Short Message

Delivery Point to Point) message that incorporates the GSM SMS

PDU in the form of SMS Bearer Data. The source of this bearer

data can be another GAIT mobile, a GSM mobile or an ANSI-136

mobile.

This information is mapped onto a R-Data (Relay Data) message

where the mandatory (M) information elements are set and

optional (O) information elements are enabled as the network

operator chooses. The mandatory information elements are

required for minimum functionality to deliver the R-Data Unit.

Optional information elements can be enabled that allow for addi-

tional information to be presented such as the originating user's

address.

The R-Data message payload is the R-Data Unit, which includes

the higher-layer-protocol identifier (HLPI) and the HLPDU (Higher-

layer-protocol data unit) which is the GMS SMS PDU.

The higher layer protocol identifier contains information about the

R-Data Unit, in this case showing that the Teleservice Type

Indicator is set to one designating it as a carrier specific teleser-

vice (versus a standardized teleservice). The TSAR (Teleservice

Segmentation and Reassembly) bit is enabled for either a single R-

Data message or multiple R-Data messages (0 or 1). In addition,

the message is indicated as a GHOST message through the use

of the Teleservice Protocol Identifier Subfield which is set to 00

0011.

Figure 6.0 - GHOST Message Mapping for a Mobile Terminated GHOSTSMS message (courtesy of the GAIT Standards Group)


For mobile terminated GAIT GHOST testing, a comprehensive test

system should allow you to select the ANSI-136 R-Data message

information elements and set them to the proper values for GAIT

support. It should also allow you to monitor a mobile originated

call and log the information coming from the mobile. Figure 7.0

shows a test system screen for handling GAIT GHOST SMS mes-

saging.

The system allows the ability to receive and send GHOST mes-

sages through the R-Data message on both the DCCH and DTC.

The system will automatically determine if the mobile is camped

on a DCCH or on a DTC and handle the messages appropriately.

Interoperability with the 2935 Phone Test system allows the user

to send and capture GSM messages easily from the 2935.

Figure 7.0 - GAIT GHOST Testing Example

Figure 7.1 shows some of the advanced features for GAIT testing.

To properly set the various information elements and parameters

associated with a GHOST message, the user needs to be able to

select and manipulate each of the Higher Layer Protocol

Identifiers and the GSM SMS PDU. The R-Data Unit contains the

HLP Identifier data from the HPLI (Higher Layer Protocol Identifier)

elements and has a specific length in octets that is automatically

calculated. This system then builds the R-Data Unit based on the

selected parameters.

The message-type mandatory information element is definable as

are the Optional R-Data elements that can be configured directly

by selecting a pop-up window that allows additional parameters

to be set.

After the GSM SMS PDU, HLPI and R-Data mandatory and option-

al information elements are defined, the system then builds an

entire R-Data message for transport to the phone. Conversely, the

system will also decode an R-Data message and display the

appropriate parameters.

This particular test system offers the user three ways to build a

HLPDU (GSM SMS PDU).

1. By loading a predetermined GSM SMS PDU formatted file.

2. By using a default PDU based on a SMS message received

from the 2935.

3. By creating a new PDU based on your input.

Figure 7.1 - Setting the GHOST HLPI and R-Data Message

Parameters

Creating a new GSM SMS PDU

To create a new GSM SMS PDU, select the "Create New PDU"

button (See Figure 7.2). The next screen that comes up is the

screen used in formatting the SMS PDU. Formatting the GSM

SMS PDU is important in that the GAIT test specification requires

the testing of different PDU contents. For more information, see

the GAIT document, GAIT-T-H-1-1-4-0.doc.

The GSM SMS PDU is formatted according to GSM 03.40. The

main parameters of the PDU are as follows:

SMS PDU Value DescriptionInformation Element

TP-MTI User Defined Message Type Indicator - SMS Deliver "00" isdefault for GAIT.

TP-MMS User Defined More Messages to Send - GAIT Default is "0" -More messages are waiting. "1" means no moremessages are waiting.

TP-RP User Defined Reply Path - GAIT Default is "0" - Parameter isnot set in this SMS Deliver. "1" means the returnpath parameter is set in this SMS Deliver.

TP-UDHI User Defined User Data Header Indicator - GAIT Default is "0".The TP-UD field contains only the short message."1" means the beginning of the TP-UD field con-tains a header in addition to the short message.

TP-SRI User Defined Status Report Indicator - "0" means a status reportwill not be returned to the SME. "1" means a sta-tus report will be returned to the SME.

TP-OA User Defined Originating Address, International Number, E.164format.

TP-PID User Defined Protocol Identifier - "0000 0000" Default. Usingthe PID of "0111 1111" enables sim data down-load, in conjunction with the DCS being set to 8bit data.

TP-DCS User Defined Data Coding Scheme:a. Default Alphabet, Class 0 (1111 0000) or

240b. Default Alphabet, Class 1 (1111 0001) or

241c. Default Alphabet, Class 2 (1111 0010) or

242d. 8 Bit Data, Class 2 (1111 0110) or 246

TP-SCTS User Defined Service Center Time Stamp.TP-UDL Auto Calculation User Data Length - Auto Calculated based on

message length.TP-UD User Defined User Data .

Defining a GSM SMS PDU

The GAIT Phone Test Suite will map this file into a HLPDU data

field and then send it out as a GHOST message to a mobile oper-

ating in the ANSI-136 Native or Foreign mode when the Send

GHOST button is selected on the main GAIT Phone Test tab.

Figure 7.2 - Setting the GSM SMS PDU Parameters

Other GAITTests, emergency calls

Other tests associated with the GAIT mobile should include the

ability to make an emergency call internationally without requiring

registration of the mobile on the network, especially when the

phone is on a forbidden network.

Opening the GATE to the future of GAIT - Phase One

+ and beyond

Since the world of wireless communication continues to turn,

there will undoubtedly be more features added to the GAIT

mobile. Phase One + will add GPRS capability to the mobile for

faster data rates.

Cameron Coursey, head of the GAIT Standards Group says it well,

"The TDMA and GSM networks offer complimentary performance

features that will play an important part in the delivery of quality

wireless services to our customers. The GAIT mobile will act as a

bridge to bring these two world class technologies together to

provide true international world phone capabilities and its bene-

fits to operators and users alike."

Where we go from here is only limited to our imagination as we

take another step closer to a true "world phone."

Part No. 46891/900, Issue 2, 07/05

CHINA Beijing

Tel: [+86] (10) 64672716

Fax: [+86] (10) 6467 2821

CHINA Shanghai

Tel: [+86] (21) 6282 8001

Fax: [+86] (21) 62828 8002

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601


USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352














ApplicationNote

Testing GSM-R terminals with the

IFR 2935 test set

The IFR 2935 test set software has been enhanced to cover the

GSM-R frequency bands, therefore giving service companies an

economical way of testing the new digital radio system for

European Railways.

The "Union Internationale des Chemins de Fer" (UIC) has devel-

oped the Project EIRENE, that specifies the functional require-

ments for a digital radio standard for the European railways. This

standard must satisfy the mobile communications needs of the

European railways. It encompasses ground-train voice and data

communications, together with the ground-based mobile com-

munications needs of trackside workers, station and depot staff

and railway administrative and managerial personnel.

This project must provide interoperability for trains and staff cross-

ing national or other borders and offer an appropriate standard for

future replacement of national radio systems operating on both

important internal routes and low to medium traffic rural areas.

To meet the functionality and performance requirements of

EIRENE, the following system services are required:

VVooiiccee sseerrvviicceess

· Point-to-point voice calls

· Public emergency calls

· Broadcast voice calls

· Group voice calls

· Multi-party voice calls

DDaattaa SSeerrvviicceess

· Text message bearer service

· Bearer service for general applications

· Bearer service for automatic fax

· Bearer service for train control applications

Direct mode for local set-to-set operation without

network infrastructure

RRaaiillwwaayy ssppeecciiffiicc aapppplliiccaattiioonnss

Three distinct mobile radio types are required, based on the type

of role they will perform and the environment in which they will

operate:

CCaabb rraaddiioo - for use by the driver of the train

GGeenneerraall ppuurrppoossee rraaddiioo - for general use by railway personnel

OOppeerraattiioonnaall rraaddiioo - for use by railway personnel involved in train

operations such as shunting and trackside maintenance.

To reduce the development and deployment costs of such a sys-

tem, UIC has based its standard on GSM technology , 900 MHz

band and modified it to satisfy the railway’s requirements while

keeping the capacity of connection to GSM public networks.

Therefore the radios will be allowed to use channels 1 to 124 and

955 to 1023 of the GSM 900 MHz frequency plan, as well as

channels 512 to 885 of the GSM 1800 MHz frequency plan.

This capacity to connect to public networks makes testing EIRENE

radio sets using an IFR 2935 test set possible.

IFR 2935 test set description

The IFR 2935 test set is a PC based system that is both easy to

use and flexible. The product core is a GSM test head (2935) driv-

en by dedicated software (Phonetest) through an RS232 interface.

This system is "Windows" based and accesses all modern PC

resources. It has several measurement modes, and can be used

as a network simulator to test the mobile in its usual environment,

or as a simple test instrument to align mobiles without any proto-

col emulation.

When used as a network simulator, the "Manual Mode" allows for

a mobile to register on a control channel , therefore measuring

the access burst parameters to establish a call initiated either from

the mobile or from the simulator. The test set displays the stan-

dard transmitter measurements ( power level, modulation accura-

cy, etc.) and receiver measurements (sensitivity, etc.). It is also

possible to check the capacity of changing the traffic channel

while in conversation (handover), the terminal's response to sim-

ulator's instructions and the quality of the voice path. SMS trans-

mission and reception capacity can be tested as well.

In automatic mode, the same measurements can be performed

under the control of test routines or "Scripts". A script is dedicat-

ed to test on several channels and several power levels and to

generate a test report, the evidence that the terminal is working

well. Therefore a terminal can be checked quickly and returned

back to users or routed for repair in case of problems. The test

report is useful for the repair shop. Building a new test script is

possible starting from an existing script edited as a text file. A

script wizard is also provided with the system, allowing for fast and

easy generation of new scripts.

The basic way of using the 2935 and Phonetest as a simple ter-

minal test system can be extended as the PC offers storage and

communication facilities when used with the Phonetest Manager

software option. This is a comprehensive database storing all test

results and providing a data analysis facility for better quality man-

agement.

Operation

AA.. MMaannuuaall TTeessttiinngg

The terminal to test is fitted with a dedicated test SIM card

allowing the terminal to register within the test network whose

parameters are:

MMoobbiillee CCoouunnttrryy CCooddee MMCCCC 000011


MMoobbiillee NNeettwwoorrkk CCooddee MMNNCC 0011

Also configure the different test set parameters (menu

Test/*Settings/Test Settings).

Select the frequency band (990000 oorr 990000//11880000), the traffic channel

within the GSM-R specific band, cable gains or losses according

to the frequency range and the connection type (direct connec-

tion or antenna coupler). Connect the terminal to the test set and

run "MMaannuuaall MMooddee". The lowerpart of the screen displays the con-

trols, test parameters and test results.

Then proceed as follows:1. Switch the terminal "On". After a few seconds, the terminal reg-

isters and terminal information is displayed in a separate window.

Please notice that the test set is able to capture and measure the

access burst (short burst) of the terminal.

2. Make a call from the terminal or from the test set using the key

" PPaaggee MMoobbiillee".

3. As soon as the call is complete, the screen displays repetitive

measurements.

aa)) TTrraannssmmiitttteerr mmeeaassuurreemmeennttss

PPoowweerr pprrooffiillee : Checking the burst shape

PPoowweerr lleevveell : The value must match the level requested for the

mobile and display this on the screen (PL9)

RRMMSS pphhaassee eerrrroorr : Modulation measurement quantified in RMS

PPeeaakk pphhaassee eerrrroorr : Modulation measurement quantified in Peak

FFrreeqquueennccyy eerrrroorr: Instantaneous value of the frequency difference

between the transmitter frequency and the expected value. This

result can change but must remain within limits.

TTiimmiinngg aaddvvaannccee:The result should equate to the timing advance

requested to the phone. This value (TA) is shown at the bottom left

of the Phonetest window under the settings information being dis-

played.

The "TTrraaccee" window allows for graphical measurements:

FFuullll ppoowweerr pprrooffiillee: Graphical display of the burst profile,

useful to check former faulty bursts

UUsseeffuull ppaarrtt//RRaammppss : Extends the profile analysis capability

PPhhaassee pprrooffiillee : Displays how phase varies during the bursts

IInn--cchhaannnneell ssppeeccttrruumm: Displays the RF spectrum generated by the

mobile

IIQQ aaddjjuussttmmeenntt: Displays the lines relevant for IQ modulator adjust-

ment

bb)) RReecceeiivveerr mmeeaassuurreemmeennttss

RRxx LLEEVV : Displays the RF level measured by the receiver. The

value must match the RF generator level, taking into

account the connection losses.

RRxxQQUUAALL : Displays the quality of reception measured by the

receiver. This value is calculated by the phone against

known bits within the received frames.

Measuring receiver sensitivity requires other measurement tech-

niques : BER measurement is performed by the test initiating RF

loopback within the mobile, i.e. the bits received by the mobiles

are transmitted back to the test set and the BER measurement is

performed on these. This measurement requires the mobile to be

fitted with a test SIM. To select these test functions, click on the

"UUppddaattee" key. This opens a window named "MMaannuuaall MMooddee

SSeettttiinnggss". Within this window, it is possible to re-configure all the

test parameters, introduce cable compensations and select the

different test modes for the receiver.

-- TTaallkkbbaacckk : The user speaks into the mobile's microphone and

this speech is then transmitted to the test set by the mobile . The

test set re-transmits the speech to the mobile with a slight delay,

providing a simple way of testing both microphone and earphone.

If the speech check is unsuccessful then to aid diagnosis "Test

tone" can be selected. The test set will send a continuous tone to

the earphone only.

-- BBEERR : The test set puts the mobile into RF loopback and performs

the BER measurement on class 1 & class 2 bits (BER1 & BER2).

-- RRBBEERR : The test set puts the mobile into RF loopback and per-

forms the Residual BER measurement on class 1b & class 2 bits

(RBER1b & RBER2) together with the frame erasure rate (FFEERR).

RBER2 is the most sensitive measurement and is used as the ref-

erence measurement mode for sensitivity.

It is therefore possible to check that the terminal sensitivity is bet-

ter than the minimum required, while setting the RF generator

level at the reference value. The bit error rate should be below the

specified limit.

Note : Test limits and test conditions are defined inside the

"TTeesstt//* SSeettttiinnggss//TTeesstt LLiimmiittss((TTxx)),,TTeesstt LLiimmiittss((RRxx))" menus.

Automatic testingThe 2935 test set is fitted with an automatic system able to run

complex routines (scripts) including some manual operations.The

test engineer will have to switch the terminal on, make or answer

a call, check the audio quality or clear the call, following the test

routine. Other operations will be fully automatic.

To run an automatic test, select the relevant script using

"FFiillee//LLooaadd SSccrriipptt". Then start using "RRuunn SSccrriipptt"and follow the

instructions appearing on the screen.

At the end of the script, close the session using the "DDoonnee" key.

This opens a window allowing for data recording (if data base

options like Phonetest Manager or Phonetest Exchange are avail-

able) or/and printing. It is also possible to come back to manual

test or re-run the script or another one before closing the session.

The 2935 test set is delivered with a wizard helping to easily cre-

ate new test scripts. Scripts can be edited and modified, adding,

changing or deleting lines and values in order to obtain a test rou-

tine perfectly suited for the terminal.

Other testsGSM-R terminals may also have GPRS packet data transmission

mode. The 2935 test set allows to check this mode if fitted with

the GPRS test option. SMS transmitted and received can also be

tested.

ConclusionThanks to the frequency range used by the GSM-R standard, the

2935 test set is perfectly suited for testing GSM-R terminals to

ensure terminal operation. Its capability in both manual and auto-

matic test modes gives service centers a solution with adaptabili-

ty, a simple man/machine interface and good communication

capacity with data bases. Therefore the 2935 Phonetest is the

perfect tool for GSM-R terminals servicing.

Part No. 46891/948, Issue 1, 12/05

CHINA Beijing

Tel: [+86] (10) 6539 1166

Fax: [+86] (10) 6539 1778

CHINA Shanghai

Tel: [+86] (21) 5109 5128

Fax: [+86] (21) 5150 6112

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Cambridge

Tel: [+44] (0) 1763 262277

Fax: [+44] (0) 1763 285353

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601


USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352











ApplicationNote

Testing mobiles faster with the 2935

by Peter Jennings

In the GSM repair field, there is an ever-present need to improveefficiency. This is not just so as to improve the turn-around ofmobiles, but also to be able to maximize the return on capitalequipment. One way in which efficiency can be improved is tominimize the time taken in testing a mobile phone.

There are three ways in which the user of an Aeroflex 2935 GSMtest set can achieve faster test times, with no degradation in testaccuracy:

1. Use test scripts that have been optimized for speed.

2. Use the new High Speed Test option, 02.

3. Use both.

This Application Note concentrates on the first of these. It pro-vides simple guidelines on how best to optimize test sequencesfor speed. It shows typical time savings that can be made by fol-lowing the guidelines, and also the additional improvements thatcan be made through using the High Speed Test option, 02.

Guidelines for speed-optimized scripts

The guidelines for increasing test speed do not involve changingthe accuracy (and hence the validity) of the test; they simplyinvolve re-ordering the sequence.

1. Minimizing the effect of changing RF generator level

Changes to the 2935's RF generator level have a delaying effecton the results of some tests, proportional to the magnitude of thelevel change. The implications of this are:

(a) If testing the mobile over a range of RF generator levels, thenminimize the size of each level change. This can be done byincreasing (or decreasing) the level progressively, as opposed toarbitrarily jumping from a low level to a high one and back again.

(b) Assuming that the effective running of the script is unaffected,a change in RF generator level should be followed by setup state-ments or protocol changes, rather than parametric tests. In par-ticular, if using the High Speed Test option, then RxLEV andRxQUAL tests should be left as long as possible after a change ofgenerator level.

The following segments from 'standard' and speed-optimized('HS') scripts illustrate the latter point:

2. Minimizing the effect of changing test settings

Non-optimized scripts, including the standard ones that have traditionally been included with PhoneTest (e.g. 'GSM 900+1800Brief test.seq'), tend to be structured as a simple sequence oftests, repeated several times over as the traffic channel changes.Clearly, it would be preferable to start the next sequence of testsusing the settings that were inherited at the end of the previoussequence. For example, if the mobile's power level is progres-sively decreased during the first test sequence, then it should beprogressively increased during the next sequence, and so-on,with the sequences alternating:

:Place_Call 1 1Clear_From_MSPage_Mobile 1 1Set_MS_Power Highest:Measure_PowerMeasure_Power_ProfileMeasure_TimingMeasure_Tx_FreqMeasure_RMS_PhaseMeasure_Peak_PhaseSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_Power:HandOff 63 1Set_MS_Power Highest:Measure_PowerMeasure_Power_ProfileMeasure_Tx_FreqMeasure_RMS_PhaseMeasure_Peak_PhaseSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_Power:HandOff 124 1Set_MS_Power Highest:Measure_PowerMeasure_Power_ProfileMeasure_Tx_FreqMeasure_RMS_PhaseMeasure_Peak_PhaseSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_Power:

‘Standard’

Tests on eachtraffic channelcarried out in thesame order. In thiscode there is atotal of ninesettings of mobilepower, with tworelatively longdelays as itchanges all theway from “Lowest”to “Highest”.

:Set_RF_Gen_Level -80.0:Page_Mobile 1 1Set_MS_Power HighestTest_Speech_QualitySet_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerMeasure_Power_ProfileMeasure_TimingMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_PowerMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack:

‘HS’

‘Measure_RxLev’called long afterRF Gen Levelchange.Hardware alreadysettled, so resultavailableimmediately.

:Set_RF_Gen_Level -80.0:Page_Mobile 1 1Set_MS_Power HighestTest_Speech_QualitySet_TCH_Mode RBERSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerMeasure_Power_ProfileMeasure_TimingMeasure_Tx_FreqMeasure_RMS_PhaseMeasure_Peak_PhaseSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_PowerMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Measure_RxLevSet_MS_Power 9Set_TCH_Mode TalkBack:

‘Standard’

‘Measure_RxLev’calledimmediately afterRF Gen Levelchange. Testhas to wait forhardware tosettle.


PhoneTest V2.2-onwards includes, for continuity, the traditionalscripts such as 'GSM 900+1800 Brief test.seq' and, in addition, an'HS' version of the 'Brief' and 'Comprehensive' scripts that havebeen speed-optimized using the guidelines above. The very sim-ple nature of the 'Call + RF test' and 'Call Processing only' scriptsmeans that no significant gain can be made from an 'HS' version.

Savings in test time

The savings in test time that can be gained through script opti-mization and using the High Speed Test option, depends upon thenature of the test sequence being run. To indicate the typicalspread, the table below lists the savings made on the four test categories supplied as standard with PhoneTest:

As an example of optimizing a custom script, one large 2935 userhas developed a final test script for use with Nokia mobiles. Thesavings that could be made on that test were:

Reducing burst counts

Although the above tests were all carried out on a "like-for-like"basis, with no change to any test parameters, we have also inves-tigated the effect of reducing the number of bursts used by cer-tain tests, and have found that further time savings can be madewith minimal effect on accuracy.

By modifying the following setup statements at the start of thescripts such that:

Set_RmsPhase_Bursts 20 becomes:Set_RmsPhase_Bursts 2

Set_PeakPhase_Bursts 20 becomes:Set_PeakPhase_Bursts 4

Set_TxFreq_Bursts 20 becomes:Set_TxFreq_Bursts 6

a further reduction in test time can be gained. See right-hand column below:

'HS' dual band script

All the standard and 'HS' PhoneTest scripts are supplied withPhoneTest, the latest version of which can be found in the 2935downloads section at www.aeroflex.com. So that they correctlyreflect the tests in the standard version, the supplied 'HS' do notinclude any burst count reduction. To summarize all the points inthis application note, here is the full listing of the 'HS Optim bursts'version of the standard GSM 900+1800 Brief test.seq:

Note: All the above tests include a number of factors that are beyond the control of the test set, i.e. thetime taken for a mobile to register, for the user to answer a call, close a call, and conduct a speechquality test. The percentage gains shown above include the following allowances for these activities:

• ‘Brief’, ‘Comprehensive’, ‘Call + RF Test’, ‘Call Processing only’: 30 sec.• ‘Typical Nokia Final Test’ (dual band): 25 sec.• ‘Typical Nokia Final Test’ (tri band): 40 sec.

All tests, both “Normal” and “HS” were run using a 38400 Baud rate between the PC and the 2935 Testhead.

Time saved by using:‘HS’ script, ‘HS’ script, ‘HS’, Optim Bursts,

Opt 2 disabled Opt 2 enabled Opt 2 enabledSingle band 5 sec (6%) 22 sec (27%) 29 sec (35%)‘Brief’Dual band 11 sec (8%) 40 sec (29%) 55 sec (40%)

Single band 8 sec (6%) 30 sec (21%) 38 sec (27%)‘Comprehensive’Dual band 8 sec (3%) 51 sec (20%) 67 sec (27%)

‘Typical Nokia Dual band 5 sec (6%) 25 sec (30%) 30 sec (36%)Final Test’ Tri band 11 sec (9%) 39 sec (30%) 46 sec (36%)

Time saved by using:‘HS’ script, ‘HS’ script,

Opt 2 disabled Opt 2 enabledDual band 5 sec (6%) 25 sec (27%)‘Typical Nokia Final

Test’ Tri band 11 sec (10%) 39 sec (34%)

Time saved by using:‘HS’ script, ‘HS’ script,

Opt 2 disabled Opt 2 enabledSingle band 5 sec (6%) 22 sec (27%)‘Brief’Dual band 11 sec (8%) 40 sec (29%)

Single band 8 sec (6%) 30 sec (21%)‘Comprehensive’Dual band 8 sec (3%) 51 sec (20%)

‘Call + RF test’ Single band N/A 4 sec (8%)Dual band N/A 8 sec (11%)

‘Call Processing only’ N/A N/A N/A

:Place_Call 1 1Clear_From_MSPage_Mobile 1 1Set_MS_Power Highest:Measure_PowerMeasure_Power_ProfileMeasure_TimingMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_Power:HandOff 63 1:Measure_PowerSet_MS_Power MiddleMeasure_PowerSet_MS_Power HighestMeasure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_Phase:

HandOff 124 1:Measure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_Power:

‘HS’

Tests on one trafficchannel mirror theorder that theywere done on thepreceding channel.In this code thereare only sevensettings of mobilepower, with eachchange reflectingonly half of themobile’s range.

Note that, althoughtests are carriedout in a slightlydifferent order, thecoverage isidentical.

Part No. 46891/922, Issue 1, 10/03

CHINATel: [+86] (21) 6282 8001Fax: [+86] (21) 6282 8002

EUROPETel: [+44] (0) 1438 742200Fax: [+44] (0) 1438 727601

FRANCETel: [+33] 1 60 79 96 00Fax: [+33] 1 60 77 69 22

HONG KONGTel: [+852] 2832 7988Fax: [+852] 2834 5364

SCANDINAVIATel: [+45] 9614 0045Fax: [+45] 9614 0047

SPAINTel: [+34] (91) 640 11 34Fax: [+34] (91) 640 06 40

UNITED KINGDOMTel: [+44] (0) 1438 742200Toll Free: [+44] (0800) 282 388 (UK only)Fax: [+44] (0) 1438 727601

USATel: [+1] (316) 522 4981Toll Free: [+1] (800) 835 2352 (US only)Fax: [+1] (316) 522 1360



As we are always seeking to improve our products,the information in this document gives only a generalindication of the product capacity, performance andsuitability, none of which shall form part of any con-tract. We reserve the right to make design changeswithout notice. All trademarks are acknowledged. Parent company Aeroflex, Inc. ©Aeroflex 2003.

Print_HeaderSet_RF_Gen_Level -80.0ResetSystem_Type GSM900/1800Set_BCCH_Arfcn 62Set_MS_Power 9Set_MS_Timing 0Set_Authentication_Check ONSet_IMSI_Attach AllowedSet_Registration_TimeOut 30Set_Protocol_TimeOut 20Set_TxPower_Limits HIGH -2.0 2.0Set_TxPower_Limits NORMAL -3.0 3.0Set_TxPower_Limits MID -4.0 4.0Set_TxPower_Limits LOW -5.0 5.0Set_TxTiming_limit 1.000000Set_TxFreq_Limit 0.100000Set_RMSPhase_Limit 5.000000Set_PeakPhase_Limit 20.000000Set_TxFreq_Bursts 6Set_RmsPhase_Bursts 2Set_PeakPhase_Bursts 4Set_TxMeasure_Mode AverageSet_RxLev_Limits 4 6Set_RxQual_Limit 4Set_RxBER1_Samples 45500Set_Ber1_Limit 0.410000Set_RxBER2_Samples 8200Set_Ber2_Limit 2.439000Set_RxRBER1b_Samples 33000Set_RBer1b_Limit 0.410000Set_RxRBER2_Samples 8200Set_RBer2_Limit 2.439000Set_RxFer_Samples 500Set_Fer_Limit 0.200000Set_TCH_Mode TalkBack

RegistrationPlace_Call 62 1Clear_From_MSPage_Mobile 1 1Set_MS_Power HighestTest_Speech_QualitySet_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerMeasure_Power_ProfileMeasure_TimingMeasure_RMS_Phase

Measure_PowerSet_MS_Power MiddleMeasure_PowerSet_MS_Power HighestMeasure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack

HandOff 699 1Set_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_PowerMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack

HandOff 885 1Set_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerSet_MS_Power MiddleMeasure_PowerSet_MS_Power HighestMeasure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack

Clear_From_BSPrint_Summary

Measure_Peak_PhaseMeasure_Tx_FreqSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_PowerMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack

HandOff 63 1Set_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerSet_MS_Power MiddleMeasure_PowerSet_MS_Power HighestMeasure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack

HandOff 124 1Set_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Measure_PowerMeasure_Power_ProfileMeasure_RMS_PhaseMeasure_Peak_PhaseMeasure_Tx_FreqSet_MS_Power MiddleMeasure_PowerSet_MS_Power LowestMeasure_PowerMeasure_Rx_RBER2Set_RF_Gen_Level -80.0Set_TCH_Mode TalkBack

HandOff 512 1Set_BCCH_Arfcn 699Set_TCH_Mode RBERMeasure_RxLevSet_RF_Gen_Level_By_Class -104.0 -102.0Set_MS_Power Lowest


ApplicationNote

The GPRS Option 01 for the 2935 provides new features that

allow efficient and flexible testing of GPRS capable mobiles.

Using the IFR 2935 GSM Tester with

GPRS Option 01

Introduction

Circulation of GPRS capable phones is now widespread

and growing. All major phone manufacturers have intro-

duced GPRS capability to their latest designs. Tools in

support of GPRS phone testing have emerged to satisfy

the needs of design, manufacturing and service. This

application note provides an overview of IFR 2935 GPRS

Test option 01 which has been designed to meet the

essential testing needs of engineers and technicians alike.

Testing a GPRS capable terminal

2935 GPRS Test option 01 can be requested either as an

upgrade to existing installed 2935's or as a factory fitted

option with new equipment. The option makes it possible

to test the functional capabilities of GPRS phones in both

voice call mode using circuit switched connection or in

data call using packet mode connection. Measurement

of transmitter and receiver performance can be per-

formed in either mode, especially important as some

measurement parameters such as Tx power profile, power con-

trol, Tx Timing and Rx sensitivity are defined or determined differ-

ently for GPRS. In test environments where time is an important

consideration, utilising the GPRS capabilities of phones as an

alternative to normal voice call testing can offer savings. Using the

GPRS mode of operation can also help to simplify testing.

The text that follows describes the implementation for GPRS test-

ing offered in the IFR 2935 with option 01.

Screen 1 shows the revised PhoneTest set-up screen including

GPRS configuration controls. All control parameters are con-

tained in a single screen to simplify set-up.

Screen 1 GPRS test setting screen

Screen 2 shows PhoneTest operated in manual mode and illus-

trates the addition of GPRS test mode and call processing fea-

tures together with GPRS specific receiver BER / BLER test param-

eters. Those familiar with 2935 PhoneTest will recognize that little

has apparently changed to the general presentation of measure-

ment information thus making the transition to GPRS testing pain-

less for the technician. Test progress bars associated with receiv-

er BER & BLER tests inform the user when the measurement is

complete. The measurement time is determined by the number

of samples specified in the GPRS test setting screen 1, e.g. using

a sample length of 400 results in a BLER measurement time of 8

seconds. Test time can be further reduced by lowering the sam-

ple length but at the expense of degrading measurement uncer-

tainty.

Screen 2 Manual Mode Operation showing additional GPRS

features

In manual test mode GPRS testing is initiated by first selecting the

GPRS mode key. Before a GPRS attach can be performed the ter-

minal must first be registered. When the GPRS attach is complet-

ed, GPRS class mark information is updated in the terminal infor-

mation window as shown in screen 3 and the manual mode dis-

play title bar is updated to indicate GPRS Attached. Once

attached, the terminal remains so until switched off or until regis-

tration is repeated. Under normal circumstances registration and

attach both occur automatically simply by turning the phone on.

Test parameters associated with the GPRS attach access burst are

frozen and displayed until Test Mode is activated.

Screen 3 The terminal information screen with

GPRS information

The terminal is now ready to receive or send packet data. The

operator commands the terminal into a Test Mode operation

(GPRS Option 01 supports both Test Modes A and B). In either

Test Mode, the terminal receives and sends data from which the

tester measures the transmitter and receiver performance. Unlike

GSM voice mode testing, it is not necessary to dial a number or

receive a call using the terminal keypad, connection is completed

automatically. As well as being simpler, this also helps to make

testing faster. Further speed improvements are provided through

the specific use of the more efficient BER Test Mode B. To estab-

lish with confidence the receiver's sensitivity performance, GPRS


terminals that are able to operate in Test Mode B require fewer

frames than would be the case in a corresponding GSM voice call.

From the manual mode screen, the user can easily initiate hand-

off commands that tell the terminal to switch to a different PDTCH

(Packet Data Traffic Channel) and different TN, (time slot number).

Alternatively the same action can be achieved by selecting the

Update key in which case the user can also control PL (Power

Level) and TA (Timing Advance) parameters too. Screen 4 shows

the both the handoff and the Update windows.

Screen 4 Initiating a Handoff

Using The 2935 Script Wizard to Build GPRS Test Cases

2935 with GPRS Option 01 supports expanded script wizard func-

tions that provide the user the added flexibility to easily define a

range of simple or comprehensive test sequences that will exer-

cise a phone in GSM or GPRS or both modes of operation.

Aeroflex supplies a variety of standard test scripts incorporating

GPRS test cases that make it possible to commence testing imme-

diately. Screen 5 shows the initial setup for defining a GSM or

GPRS test script. Selecting Registration with GPRS attach causes

the presentation of additional GPRS pages as the wizard pro-

gresses.

Screen 5 GPRS PhoneTest script wizard

As the user progresses through the script wizard, they can set up

the test parameter and test limits independently for each of GSM

and GPRS modes. Any voice call mode tests specified are per-

formed first, followed by any defined GPRS tests. The power lev-

els and traffic channel numbers are common to both modes

although in later screens the user can further refine the selection

making it possible to test GPRS on different power level and chan-

nel combinations to those used for voice call mode. In this way

very efficient test sequences can be designed that avoid repeat-

ing tests unnecessarily.

Screen 6 shows the Tx testing setup screen. The transmit param-

eters are common to those measured in voice call mode howev-

er, there are some subtle differences in the way GPRS power con-

trol and timing are defined and therefore they can merit meas-

urement in both modes.

Screen 6 GPRS Tx test setup

Screen 7 shows the Tx test limit setup screen. Different limits can

be set for different power level test cases as well as the number

of measurement averages that are used.

Screen 7 GPRS Tx test limit setup

Screen 8 GPRS Rx test setup

Screen 8 shows the Rx testing setup for BER or BLER.

Tx and Rx test parameter control is provided independently to

permit greater flexibility in designing scripts. BER measurement

requires the mobile to support Test Mode B, whereas BLER meas-

urement is common to Test Modes A and B. Either measurement

can be specified for coding schemes CS1 or CS4 independently.

However, RxQual measurements are only relevant when using

coding scheme CS1.

Full flexibility in configuring the Rx test pass/fail limits allows the

user to define as comprehensive or as quick a test as required.

Screen 9 shows the Rx Limits setup through the script wizard.

Screen 9 Rx Limits setup screen

Conclusion

Currently phone manufacturers offer a limited range of GPRS

capability. In time, the range of specified GPRS classes available

will lead to a ever wider variety of different phone types and lev-

els of capability. This evolution will increasingly mean that flexible

GPRS modes of testing will take precedence over standard voice

call testing. This application note has introduced GPRS option 01

for IFR 2935 and considered the various benefits of using the

GPRS mode of operation. Aeroflex will continue to enhance prod-

uct performance in line with the advances in phone capability to

ensure affordable and competent test capability exists. Visit the

Aeroflex website at

http://www.aeroflex.com/products/commtest/cellparamet-

ric/2935.cfm to keep up to date on Aeroflex GPRS test capabili-

ties.

Part No. 46891/906, Issue 2, 07/05

CHINA Beijing

Tel: [+86] (10) 64672716

Fax: [+86] (10) 6467 2821

CHINA Shanghai

Tel: [+86] (21) 6282 8001

Fax: [+86] (21) 62828 8002

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601


USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352














ApplicationNote

Within the 2935 and PhoneTest there are 2 manual modesof operation, designed to simplify the test, repair and align-ment of GSM Mobiles.

Why does the 2935 have 2 Manual

Modes?

Introduction

Why are there 2 different manual modes in

PhoneTest?

Within the 2935 and PhoneTest there are 2 manual

modes of operation, designed to simplify the test and

repair of a mobile. Firstly there is a synchronous mode

called BCCH mode (in call testing) and a unsynchronised

mode called TCH mode (used for alignment).

BCCH Manual Mode

This mode controls the mobile as if it were operating on

a network and is used to verify autotest failures and assist in the

repair of the mobile. BCCH (Broadcast Control CHannel) Manual

mode gives the user control over the important network functions

of the mobile such as:

Registration

Placing a call to the mobile

Placing a call from the mobile

Clear down from the mobile

Clear down from the basestation

Handoffs between traffic channels

Handoffs between frequency bands

Setting of control channel, traffic channel, time slot, power level,

timing advance and signal generator level

In addition a wide range of parametric functions are measured to

the ETSI 300 607-1 specification. These are measured for both

normal conversation bursts and also the access burst (RACH

burst).

The transmitter measurements are :-

Mobile power,

Power ramp profile,

Frequency error,

Peak and RMS phase and

Timing advance.

Graphical displays of:-

Full Power Profile,

Useful Part,

Ramps, and

In-channel spectrum.

The receiver can be evaluated by measuring bit error rates and

residual bits error rates on class 1b and class 2 bits, as well as the

frame erasure rate (FER).

Information reported to the network is also displayed such as the

RSSI (RX LEV and RX QUAL) and the reported timing advance and

power level.

Figure 1: BCCH Manual Mode

The flexibility and ease of use of the 2935 allows the user to

quickly assess the operation of a mobile under a wide range of

operating conditions.

TCH Test Mode

TCH test mode is used in the alignment and fault finding of a

mobile when the mobile have been placed in a test mode by

either the use of external propriety software or by a sequence of

"hidden" key presses.

Figure 2: TCH Manual Mode

TCH mode provides an unsynchronised mode where the trans-

mitter parameter can be measured without the need for a call to

be established. The measurements available are mobile power,

power ramp profile, frequency error, peak and RMS phase, timing

advance and output spectrum. As the training sequence is used

as a reference for frequency, phase and timing measurements, if

a training sequence is not present then only the transmit power

and output spectrum can be measured.

In addition to the transmitter measurements an unbursted GMSK

signal or a fixed carrier signal can be generated at any frequency

in the 900, 1800 and 1900 bands. The output level can be set

within the range -40 to -120 dBm in steps of 0.1 dB.

The graphical TCH mode display of the mobile is simple and effi-

cient. With the TCH mode the user has the ability to save the test

configurations for allowing routine calibration steps to be easily

recalled.

Types of alignment

Each manufacturer has a different method of mobile alignment.

The following identifies the parts of the mobile that can be aligned

and how the alignment is carried out.


Figure 3: A very simplistic block diagram of a 900 MHz GSM

Mobile. The grey areas represent the areas of primary align-

ment.

Mobiles transmit power alignment

The mobile is placed in a test mode, where it is set up to transmit

a signal of predefined frequency and power. The test equipment

is set up to measure the transmitted power from the mobile and

the transmitted power is adjusted for a desired level.

RSSI (Received Signal Strength Indicator)

The test equipment is set up to generate a signal of a known leveland frequency which is injected into the mobile. The signalstrength measured via the test mode on the mobile. Then a sec-ond signal generator level is then measured, from this theabsolute RSSI and the RSSI linearity is determined.

AFC (Automatic Frequency Control)

A signal is injected into the mobile at a known offset from the

channel and the mobile then measures the frequency to identify

if it within its AFC capture range.

IQ alignment

The separate gains of the IQ channels in the mobile are adjusted

until they are balanced. This is usually carried out with the mobile

generating a GMSK (Gausian Minimum Shift Keying) signal carry-

ing all 1’s or all 0’s data. This produces a characteristic spectrum

where the maximum signal is offset by 67.71 kHz from the nomi-

nal carrier and distinctive spurs are produced as shown below.

Automatic testing

When the mobile has been repaired and aligned a sequence of

predefined test are carried out. To verify the mobile is suitable to

be used on the GSM network. The 2935 and PhoneTest provide

a highly flexible and high speed approach to GSM mobile repair

and test.

RX I

RX Q

TX Q

TX I

AGC

PA

VCOFast Switching

890 -915MHz

BW =200kHz

90°

90°

935 -960MHz

BW =200kHz

PhaseDetect

DividersLoopFilter

XTAL13MHz

DuplexFilter ANT

IQ Alignment TX power Alignment

Receiver Alignment

Part No. 46891/856, Issue 2, 12/05

CHINA Beijing

Tel: [+86] (10) 6539 1166

Fax: [+86] (10) 6539 1778

CHINA Shanghai

Tel: [+86] (21) 5109 5128

Fax: [+86] (21) 5150 6112

FINLAND

Tel: [+358] (9) 2709 5541

Fax: [+358] (9) 804 2441

FRANCE

Tel: [+33] 1 60 79 96 00

Fax: [+33] 1 60 77 69 22

GERMANY

Tel: [+49] 8131 2926-0

Fax: [+49] 8131 2926-130

HONG KONG

Tel: [+852] 2832 7988

Fax: [+852] 2834 5364

INDIA

Tel: [+91] 80 5115 4501

Fax: [+91] 80 5115 4502

KOREA

Tel: [+82] (2) 3424 2719

Fax: [+82] (2) 3424 8620

SCANDINAVIA

Tel: [+45] 9614 0045

Fax: [+45] 9614 0047

SPAIN

Tel: [+34] (91) 640 11 34

Fax: [+34] (91) 640 06 40

UK Burnham

Tel: [+44] (0) 1628 604455

Fax: [+44] (0) 1628 662017

UK Cambridge

Tel: [+44] (0) 1763 262277

Fax: [+44] (0) 1763 285353

UK Stevenage

Tel: [+44] (0) 1438 742200

Fax: [+44] (0) 1438 727601


USA

Tel: [+1] (316) 522 4981

Fax: [+1] (316) 522 1360

Toll Free: 800 835 2352










Racal 6103Digital Radio Test Set

•• Battery Life TestingApplication Notes – rev. 1.1For use in conjunction with software releaseversion 3.03.

• Introduction:

These notes are provided to enable the user tobecome familiar with the Battery Life TestSequences. These Test Sequences follow aprocedure proposed by The EuropeanManufacturers Association - Task Group forMobile Stations (ECTEL-TWG). The Battery LifeTest procedure is designed to providemanufacturers with a consistent and accuratemethod to measure both the Standby and Talktimes of a mobile handset. The Battery Life TestSequences are available from RacalInstruments as part of the RIBASIC ExampleSequence Library, Part number 64-0047.

Note: The Battery Life Test procedure is undercontinuous improvement to enhance productperformance. The method detailed in thisapplication note follows the currentrecommendations. Please check with RacalInstruments to ascertain if any enhancementshave been made to the recommended testmethod.

•• Equipment required:

• Racal Instruments 6103 with firmware version3.03 or higher (check this by selecting "Selftests/System" from the main menu)

• Example Sequence Library P/N (64-0047)• RF test lead• Racal Instruments Test SIM• GSM, DCS or PCS Compatible phone and

battery• Epson-compatible printer.

•• Equipment set up

1. Connect the printer to the 6103 via the parallelprinter port. Ensure that the “Parallel” port isselected under the “Results” menu. Fullycharge the mobile battery. Install test SIM.

2. Connect the mobile under test to the 6103using a suitable RF lead. (an electricalconnection is preferable to an inductive).

3. If required enter the offset for the RF cable.

•• RIBASIC Sequence Library

The example sequence library disk contains a setof 15 RIBASIC example program files designedfor use with the Racal Instruments 6103 DigitalRadio Test Set. RIBASIC is a simpleprogramming language, based on BASIC, a toolwhich can be used to control the range of testfunctions available on the 6103 and manipulatethe results.

Only the Battery Life Test Sequences ‘STANDBYand TALK.SEQ’ are considered in thisdocument. To obtain more details regarding theTest Sequences and RIBASIC please refer tothe “WHERE TO GET MORE INFORMATION”section at the end of this document.

•• How to load the test sequence files

• Copy sequence files to a PCMCIA memory card• Insert the card into a slot in the 6103 Test Set• Select "Test Sequences" from the main menu• Select "Run File"• Select required file i.e. ‘STANDBY.SEQ’Note: These files can be used to test GSM900,

DCS1800 and PCS1900 phones. Differentparameter lists for all 3 systems are built intothe sequences. The appropriate parameters areloaded automatically by selecting theappropriate radio system.

RACAL INSTRUMENTSRACAL INSTRUMENTS

Battery life testing sequences•• Measurement of Standby Time

•• Description:

Measures a mobile's battery life in idle mode byrepeatedly running the Synchronization/Registration test until the mobile fails to performlocation updating. The periodic update intervalis selectable as a number of minutes from 1 to1530. The selected value is rounded up to thenearest 6 minutes, the default is 30.

Note: The Standby Time test is performed with thepaging multi-frames parameter set to 5.

Standby battery life test procedure:• Run the test sequence:

• select "Test Sequences" in main menu• select "Run File"• select the file “STANDBY.SEQ” from the

PCMCIA card and press ‘Accept’• Choose either to keep the default parameter

settings (shown in Table 1.0) or to enter yourown values. The default values are as per theECTEL- TWG recommendations, which areselected to give accurate, repeatable andrepresentative results. If you choose to set yourown parameters, the test set will ask for thefollowing:

• Current band, select 0, 2 or 3 (GSM,DCS OR PCS)

• Value of paging multi-frames (2-9)• Discontinuous Transmission DTX 0,1, 2• Periodic Update Interval Time (0-1530

minutes, each value is rounded up to thenearest 6 minutes, i.e. 1=6 etc)

• Value of page mode (0-3)

The Periodic Update Interval Time (PUIT) specifieshow often the mobile will perform locationupdating. The recommended standard setting is30 minutes, but any value down to a minimum of 6can be used. The test will run until the 6103 setdoes not receive a location update from the mobilewhen expected (i.e. the battery has run down), Aresults sheet showing start time, stop time andtotal standby time (shown in Figure 1.0) will thenbe printed.

•• Standby battery life test sequence details

♦ Automatic results printing is turned off, andprinting is directed to the printer, via theparallel port.

♦ The recommended ECTEL values (shown inTable 1.0) can be selected, or user can enterown values.

♦ The DSINF parameter is used to set the cellallocation table and the neighbor cell table (see"GSM Parameter Configuration commands",Section 6, Volume 2 of the 6103 operator'smanual).

♦ The test time out is set so it is slightly longerthan the Periodic Update Interval

♦ The Start time is recorded and then theSynchronization/Registration test is runrepeatedly until the mobile fails to respond.

♦ The stop time and stop date is then recorded.♦ Finally the total Standby time is calculated, and

the result printed.

Table 1.0 Standby Mode Parameter Settings

Parameters Settings: GSM / DCSSIM Phase 2 with clock stopRx Level -82 dBmCell re-selection NoT3212 (PUIT) 30 mins (5 deci hours)BA list 16 frequencies as follows

GSM:1,9,17,26,34,42,50,58,67,75,83,91,99,108,116,124

DCS:512,530,560,580,610,640,670, 700, 720, 740, 760, 790,810, 840, 860, 885

SMS CB OFFBS_PA_Mfrms (DRX) 5 multi frames (2 – 9)DTX – Discontinuous

Transmission.OFF (2)

Control Chan. Freq. ARFCN 62 / 70Page_Mode 0 (0 – 3)Handover No

Figure 1.0 Sample Standby test results sheet.

+-------------------------------------------+| 6103 Test Set (GSM) – Mobile Battery Life || in idle mode (standby time) |+-------------------------------------------+| Mobile Details || IMSI: ************ IMEI: *********** || Revision: Phase 2 Power class: 4 || Encryption: A5/1 A5/2 |+-------------------------------------------+| Test Results || || Parameters used: || Periodic update interval: 30 mins || Page Mode: 0 || Paging Multiframes: 5 || Discontinuous Transmission: 2 || || Test started at 09:30:36 on 20 APR 1998 || Communication lost 19:11:38 on 20 APR 1998 || Total standby time 9 Hours, 41 minutes |+-------------------------------------------+

Measurement of Talk Time

•• Description:

Measures a mobile's battery life in dedicated mode(i.e. talk time) by continuously running the voiceloopback test until the mobile fails to respond.The power level and channel are selectable atthe start of the sequence and these parametersare then used when the voice loopback test isrun.

Note: The Talk time test is performed withdiscontinuous transmission (DTX) switched off.DTX may be enabled by the addition of the“CONIFG DTX” to the Battery Life testSequences or by directly editing the value ofthe DTX parameter in a parameter file. Formore details regarding DTX, please refer to the“WHERE TO GET MORE INFORMATION”section at the end of this document.

•• Talk time battery life test procedure:

• Run the Talk test sequence as per Standbytest, but choose TALK.SEQ instead

• Choose whether to keep the default parametersettings (shown in Table 2.0) or enter your ownvalues. The default values are as per theECTEL-TWG recommendations, and havebeen selected to give accurate, repeatable andrepresentative results. If you choose to set yourown parameters, the test set will ask for thefollowing:

• Current band select 0, 2, or 3 (GSM,DCS or PCS)

• Value of paging multi-frames (2-9)• Discontinuous Transmission DTX 0, 1, 2• Periodic Update Interval Time (0-1530)• Value of page mode (0-3)• Select whether to hop on a pre-defined

series of channels (-shown in table 1.0-)or run the test on one particular channel.If you choose to run the test on onechannel, enter the channel number inthe text box.

The mobile is put into voice loopback mode untilthe call is lost, indicating the battery is discharged.A results sheet (shown in Figure 2.0) is thenprinted showing the start time, stop time and totalTalk Time.

•• Talk time Battery Life Test Sequence

♦ The recommended ECTEL values (shown inTable 2.0) can be selected, or user may entertheir own values.

♦ It is recommended to run the test with channelhopping on, as defined in Table 2.0. If it is setto off the user must choose a channel to runthe test on. The start time of the battery test isthen recorded.

♦ The DSINF parameter is the same as that usedin the standby time test

♦ The Start time is recorded, the Voice Loopbacktest is then run repeatedly until the mobile failsto respond.

♦ The stop time and stop date is then recorded.♦ Finally the total Talk time is calculated, and the

results printed.

Note: Both the standby and talk time testsequences are limited to testing an assumedmaximum battery life of one year.

Table 2.0 Dedicated Mode Parameter Settings

Parameter SettingsRx Level -82 dBmHopping ON, 5 frequencies

GSM: 1, 30 ,62, 93,124DCS: 512, 600, 690, 780, 885

Handover NoMS-Tx-Lev(Power)

5 (29 dBm) GSM1 (28 dBm) DCS

BA List 16 frequencies as followsGSM:19,17,26,34,42,50,58,6,7.75,83,91,99,108,116,124DCS:512,530,560,580,610,640,670,700,720,740,760,790,8 1 0,840,860,885

Uplink DTX 2 (OFF)Call Continuous

Figure 2.0 Sample Talk time test results sheet.

+---------------------------------------------+| 6103 Test Set (GSM) – Mobile Battery Life || in dedicated mode (Talk Time) |+---------------------------------------------+| Mobile Details || IMSI: ************ IMEI: *********** || Revision: ** Power class: 4 Encryption: On |+---------------------------------------------+| Test Results || || Parameters used || TCH ARFCN: (hopping) || Mobile power level: 7 || Periodic U/date interval: 30 mins || Page Mode: 0 || Paging Multiframes: 5 || Discontinuous Transmission: 2 || || Test started at 09:34:54 on 21 APR 1998 || Communication lost at 11:48:52, 21 APR 1998 || Total talk time: 2 Hours 14 minutes |+---------------------------------------------+

WHERE TO GET MOREINFORMATION

•• How to obtain the sequence library and theBattery life test procedure

To obtain a copy of the RIBASIC sequencelibrary and the Battery Life test procedureplease contact your local Racal Representativeor the Customer Help Desk on +44 (0)1753741010.

•• About these sequences

See explanatory comments built into thesequences: view them using any standard texteditor program such as Microsoft Notepad

•• About running sequences

See the 6103 Operator's Manual Volume 2,"Test Sequences" (Section 4) and "TestSequences Menu" (Section 5)

•• About programming in RIBASIC

See the 6103 Operator's Manual Volume 2,"Programming Test Sequences" (Section 5Annex A)

•• About parameters to use when running testswith RIBASIC

For running tests with the RIBASIC "TEST"command, see the 6103 Operator's ManualVolume 2, "Single Test Commands" (Section 6)

For defining parameters with the RIBASIC"CONFIG" command, see the 6103 Operator'sManual Volume 2, "GSM ParameterConfiguration Commands" (Section 6)

•• About the Debug facility.

The debug facility is an engineering feature ofthe 6103 designed for use by Racal engineers.This port can provide access to additionalinformation and parameters not normallyaccessible via the instruments front panel.Further information may be obtained bycontacting Racal Instruments.

•• Parameter Settings for Battery life testing

The sequence file can be edited and saved tomemory card to control the discontinuoustransmission parameter DTX. This is different tothe previous Battery Life Test sequnces forversion 2.02B, where it was necessary to usethe Debug facility to enable DTX.

The setting for the DTX parameter are 0, 1 and 2.The values have the following meaning:

0 : MS may use DTX if possible1 : MS is commanded to use DTX (DTX On)2 : MS is commanded Not to use DTX (DTX Off)

•• TroubleshootingNOTE: when using these files, you mayoccasionally get the error message "RIBASICError: Range of values for ......." - this is not anerror in the sequence files, but is caused by thefollowing:

1. If you have enabled offsets under the "Offsets"menu, then some of the parameters used in thetests may be out of range. Either alter the offsetor modify the values used in the sequence.

Disclaimer: The RIBASIC Battery Life testingsequence files have been tested on a RacalInstruments 6103 Digital Radio Test Set with alimited selection of mobile handsets. It is theresponsibility of the user to determine thesuitability of all commands and parametersused in this sequence for their equipment.Racal Instruments Limited take no responsibilityfor any damage caused by the use/misuse ofthis sequence.

In case of problems please contact your localRacal Representative or the Customer HelpDesk on +44 (0)1753 741010.

©Copyright Racal Instruments Limited 1998

Racal Instruments Inc. 4 Goodyear Street, Irvine CA 92618, USA. Tel: +1 714 859 8999 Fax: +1 714 859 7139Racal Instruments Ltd. 480 Bath Road, Slough, Berkshire SL1 6BE, United Kingdom. Tel: +44 (0) 1628 604455 Fax: +44 (0) 1628 662017 Racal Systems Electronique S.A. 18 Avenue Dutartre, 78150 Le Chesnay, France. Tel: +33 (1) 3955 8888 Telex: 697215 Fax: +33 (1) 3955 6735Racal Systems Elettronica srl Strada 2 - Palazzo C4, 20090 Milanofiori Assago MI, Italy. Tel: + 39 (2) 5750 1796 Fax: + 39 (2) 5750 1828Racal Elektronik System Gmbh Frankenforster Strasse 21, 51427 Bergisch Gladbach , Germany. Tel: + 49 2204 9222-0 Fax: + 49 2204 21491Racal Australia Pty. Ltd. 3 Powells Road, Brookvale, NSW 2100, Australia. Tel: + 61 2 9936 7000 Fax: + 61 2 9936 7036Racal Electronics Pte. Ltd. 26 Ayer Rajah Crescent, 04-06/07, Ayer Rajah Industrial Estate, Singapore 0513 Tel: + 65 7792200 Fax: + 65 7785400Racal Instruments Ltd. Sun House 13th Floor, 181 Des Voeux Road, Central, Hong Kong. Tel: + 852 281 58633 Fax: + 852 281 58141

RACAL INSTRUMENTSRACAL INSTRUMENTS

Automated Testing ofGSM Base Stations

by Steve Gledhill, Racal Instruments

Cell phones rely on

remote base stations,

which need

sophisticated test

equipment and

procedures for

proper installation

and maintenance.

Testing a Remote BTS

○

○

○

○

○

○

○

○

○

○

○

○

○

○

○

○

○

T he growing population of cell phones—today estimated at nearly300,000,000 in service worldwide—has increased the focus on mo-bile-phone base stations. Without efficiently functioning stations,

nobody talks. To avoid such problems, testing these stations is increasingly important.Essentially, a mobile-phone cellular radio base station is a sophisticated transmitter

and receiver that communicate with all of the subscribers’ mobile phones in closeproximity. The area that the base station covers is referred to as the cell. Signals to andfrom the base station communicate with the fixed telephone network through T124-channel links.

Testing such a base station should be a straightfor-ward job of applying a set of RF parametric measure-ments, for example, measuring output frequency, power,

and modulation from the trans-mitter and determining sensi-tivity, tuned frequency, anddistortion on the receiver. Inreality, the testing is signifi-cantly more complex. Manytransmitter and receiver chan-nels and associated antennason a base station and in digitalsystems, such as a global sys-tem for mobiles (GSM), codedivision multiple access (CDMA), digital advanced mobile phoneservice (D-AMPS), and parameters such as bit error rate (BER),must be measured as well as handover procedures and otherhousekeeping matters.

Numerous vital tests are carried out, especially during instal-lation and commissioning, and they must be accurate and re-peatable. The quality and scope of the tests performed at thisstage determine the efficiency and, ultimately, the revenueearned from a base station.

GSM Base StationsThe correct title for a GSM/personal communications service

(PCS) base station is a base transceiver station (BTS). It hasevolved into a sophisticated assembly with embedded supplier-specific software used to control the functions of the base station.The software has a fundamental impact on testing because thetester cannot directly control the BTS to carry out tests; the BTSmust be controlled through embedded software interfaces.

○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○

WIRELESS TEST

A BTS should not be considered inisolation to the other components thatmake up a network. Figure 1 shows asimplified GSM/PCS mobile phone net-work block diagram that focuses onthree important functional blocks: themobile switching center (MSC), the basestation controller (BSC), and the BTS.

The MSC is closest to the corenetwork infrastructure and responsiblefor switching and routing the traffic.Each MSC interfaces to a number ofBSCs in the network over a 1.5-Mbit/slink using the A interface. The BSCsare themselves interconnected to a numberof different BTSs that they control.

BTS TestingTesting is an essential part of a base station’s

installation and commissioning process. Even though theconstituent parts of a BTS will have been thoroughlytested by the manufacturer, they are disassembled fortransport and reassembled on site.

This can give rise to some faults. More significantly,other faults inevitably will arise during the complexinstallation process. For example, wiring errors willoccur even in the best-managed working environment,connectors will not always be correctly crimped, backplaneswitches can be incorrectly set, and RF connectors andcables can easily introduce faults.

When contemplating test strategies for cellular radiotesting and maintenance, consider the time taken for testingas time that could be used to generate revenue. Even onebad voice channel can affect numerous connections duringthe course of a day and produce large numbers of faultreports. A rapid, automated, high-integrity tester that iseasy to operate could increase revenue significantly.

Figure 2. Test Set Connected to the A-bis and RF Portsof a GSM/PCS Base Station

GSM/PCSBase Station

A-bis InA-bis Out RF Out

RF In

RF In

RF Out

A-bis In

A-bis Out

MSC

BSC

BSC BSC

BSCBTS BTS

BTS

A Interface

A-bis Interface BTS AntennasFigure 1. GSM/PCS MobilePhone NetworkConceptualBlock Diagram

Portable, Automated BTS TesterThe BTS is controlled by the BSC through a

1.5-Mbit/s interface, known as the A-bis interface. A1.5-Mbit/s link consists of 24 64-kbit/s time slots. In theA-bis interface, each of the 64-kbit/s slots is subdividedinto four 16-kbit/s speech subslots. Each GSM carrieraccommodates eight speech channels.

As a result, two 64-kbit/s time slots are needed foreach carrier, meaning that, in theory, 12 carriers canbe controlled and duplex speech connected for each1.5-Mbit/s link. In practice, some time slots are neededfor signaling, so only about 10 carriers can be servicedby one link.

Although the system from the BSC outward to thenetwork is standardized, the A-bis interface differs foreach manufacturer. This is a major challenge for thetest-equipment supplier. Not only must vendor-specific

software be developed for eachmanufacturer’s equipment, but on-going development also is requiredto keep up with software enhance-

ments.As a result, it is necessary

for the tester to interfacewith the A-bis connections tocontrol it and carry out auto-mated testing. The interfacemust communicate in bothdirections to read the mes-

sages returned from thebase station. Interfacesalso are required at theantenna connection portsof the BTS.

Figure 2 shows howthe test set connects tothe A-bis and RF ports

of a BTS. With this arrangement, the tester can switchthe BTS to different RF channels and check the controlsignals sent to the mobiles. The tester also shouldfeature programmable go, no-go test limits to minimizeoperator errors, printing capabilities to support futuretrend analysis and maintenance tracking, embeddedcontrol software to eliminate the need for a PC, and adesign conducive to portable applications.

About the AuthorSteve Gledhill is the marketing manager for radio com-

munications test at Racal Instruments. He has nearly 30years of experience in RF test and measurement, initiallywith Marconi Instruments and more recently with IFR.Racal Instruments Ltd., 480 Bath Rd., Slough, BerkshireSL1 6BE, United Kingdom, 011 44 1628 604455, e-mail:[email protected].

○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○

WIRELESS TEST

Reprinted from EE-Evaluation Engineering , April 2001Copyright © 2001 by Nelson Publishing Inc. • www.evaluationengineering.com

mob

Documents