GSM KPI MOS

Product Name "Confidentiality Level " "
"GSM BSS "INTERNAL "
"Product Version "Total 39 pages "
" V00R01 " "




GSM BSS Network KPI (MOS) Optimization Manual





For internal use only





"Prepared by "GSM&UMTS Network "Dong "Date "2008-2-21 "
" "Performance Research"Xuan " " "
" "Department " " " "
"Reviewed by " " "Date "yyyy-mm-dd "
"Reviewed by " " "Date "yyyy-mm-dd "
"Granted by " " "Date "yyyy-mm-dd "







Huawei Technologies Co., Ltd.


All rights reserved



Revision Record

"Date "Revisio"Change Description "Author "
" "n " " "
" "Version" " "
"2008-1-21 "0.9 "Draft completed. "Dong Xuan "
"2008-3-20 "1.0 "The document is modified "Wang Fei "
" " "according to review comments." "
" " " " "
" " " " "
" " " " "
" " " " "
" " " " "
" " " " "
" " " " "
" " " " "






GSM BSS Network KPI (MOS) Optimization Manual




Key words: MOS, interference, BER, C/I, power control, DTX, frequency
hopping, PESQ, PSQM /PSQM+, PAMS

Abstract: With the development of the radio network, mobile operators
become more focused on end users' experience instead of key
performance indicators (KPIs). The improvement of the end users'
experience and the improvement of the network capacity are regarded
as KPIs. Therefore, Huawei must pay close attention to the
improvement of the soft capability of the network quality as well as
the fulfillment of KPIs. At present, there are three methods of
evaluating the speech quality: subjective evaluation, objective
evaluation, and estimation. Among the three methods, objective
evaluation is the most accurate. The PESQ algorithm defined by the
ITU can objectively evaluate the speech quality of the communication
network. This document uses the mean opinion score (MOS) to label the
speech quality after objective evaluation.

This document describes the factors of MOS, the impact of each factor
on the MOS, and the methods of improving the network QoS and then the
speech quality. It also describes the attention points during the
test of speech quality of the existing network and the device
capability value of the lab test. In addition, this document
introduces the differences between the speech test tools. The methods
and principles of using the test tools are omitted. This document
serves as a reference to the acceptance of network KPIs and the
marketing bidding.




References: ITU-T P.800\ ITU-T P.830\ ITU-T P.861\ ITU-T P.862\ITU-T
P.853




List of acronyms:

"Acronym "Expansion "
"MOS "Mean Opinion Score "
"PESQ "Perceptual Evaluation of Speech Quality "
"PSQM "Perceptual Speech Quality Measurement "
"PAMS "Perceptual Analyse Measurement Sytem "
Contents

1 Basic Principles of MOS 3
1.1 Subjective Speech Quality Evaluation 3
1.2 Objective Speech Quality Evaluation 3
1.2.1 PSQM (P.861) Recommendation or Algorithm 3
1.2.2 PESQ (P.862) Recommendation or Algorithm 3
1.2.3 P862.1 Recommendation (Mapping Function for Transforming) 3
1.2.4 P.563 Recommendation 3
1.3 Speech Processing of Involved NEs 3
1.3.2 MS 3
1.3.3 BTS 3
1.3.4 BSC 3
1.3.5 UMG 3
2 Factors That Affect the MOS in GSM 3
2.1 Introduction to GSM Speech Acoustic Principles 3
2.2 Impact of Field Intensity and C/I on the Speech Quality 3
2.3 Impact of Handover on the Speech Quality 3
2.4 Impact of DTX on the Speech Quality 3
2.5 Impact of Speed (Frequency Deviation) on the Speech Quality 3
2.6 Impact of Speech Coding Rate on the Speech Quality 3
2.7 Impact of Transmission Quality on the Speech Quality 3
3 Method of Analyzing the Problem of Low MOS 3
3.1 Process of Analyzing the Problem of Low MOS 3
3.2 Method of Solving the Problem of Low MOS 3
3.2.1 Consistency Check and Sample Check 3
3.2.2 Um Interface Check 3
3.2.3 BTS Check 3
3.2.4 Abis Transmission Check 3
3.2.5 BSC Check 3
3.2.6 A Interface Transmission Check 3
3.2.7 MGW Check 3
3.2.8 Miscellaneous (Comparison of MOS Before and After Network
Replacement) 3
4 Test Methods and Suggestions 3
4.1 Test Tool Selection and Test Suggestions 3
4.2 Suggestions on the Test of the Existing Network 3
5 MOS Cases 3
5.1 Differences Between Speech Signal Process and Signaling Process 3
5.1.1 GSM Speech Signal Process 3
5.1.2 Signaling Process 3
5.2 Identified MOS Problems 3
6 Feedback on MOS or Speech Problems 3
6.1 Test Requirements 3
6.2 Requirements for Configuration Data in Existing Network 3

Tables

Table 1 Relations between the quality grade, score, and listening effect
scale 3
Table 2 Impact of DTX on the speech quality 3
Table 3 Mapping between the speech coding scheme and the MOS value 3
Table 4 Mapping between speech sample and MOS 3
Table 5 Impact of TFO on the improvement of speech quality (GSM Rec. 06.85)
3
Table 6 Identified MOS problems 3
Table 7 Network configuration parameters to be provided 3








Figures

Figure 1 PESQ process 3
Figure 2 Mapping between P862 and P862.1 3
Figure 3 Overall speech quality prediction of P.563 3
Figure 4 Typical MOS test process 3
Figure 5 Speech processing on the MS side 3
Figure 6 Speech processing on the BTS side 3
Figure 7 Handling process in the GTCS 3
Figure 8 Codec cascading 3
Figure 9 Fault location flow 3
Figure 10 Speech data transmission on the Um interface (schematic drawing)
3
Figure 11 BSC6000 speech process 3









Basic Principles of MOS


1 Subjective Speech Quality Evaluation

ITU-T Rec. P.830 defines a subjective evaluation method toward speech
quality, that is, MOS. In this method, different persons subjectively
compare the original speech materials and the system-processed speech
materials and then obtain an opinion score. The MOS is obtained
through the division of the total opinion scores by the number of
persons. The MOS reflects the opinion of a person about the speech
quality, so the MOS method is widely used. The MOS method uses an
evaluation system of five quality grades, each quality grade mapping
to a score. In the MOS method, dozens of persons are invited to
listen in the same channel environment and to give a score. Then, a
mean score is obtained through statistical treatment. The scores vary
largely from listener to listener. Therefore, abundant listeners and
speech materials and a fixed test environment are required to obtain
an accurate result.

Note that the opinion of a listener about the speech quality is
generally related to the listening effect of the listener. Therefore,
the listening effect scale is introduced in this method. Table 1
describes the relations between the quality grade, score, and
listening effect scale.


Relations between the quality grade, score, and listening effect scale
"Quality "Score"Listening Effect Scale "
"Grade " " "
"Very good "5 "The listener can be "
" " "totally relaxed without "
" " "paying attention. "
"Good "4 "The listener should pay "
" " "some attention. "
"Average "3 "The listener should pay "
" " "close attention. "
"Poor "2 "The listener should pay "
" " "very close attention. "
"Very poor "1 "The listener cannot "
" " "understand even with very "
" " "close attention. "



Although the formal subjective listening test is the most reliable
evaluation method and the network performance and any coding/decoding
algorithm can be evaluated, the test result varies from listener to
listener. In addition, the factors such as the listening environment,
listeners, and speech materials should be strictly controlled during
the test. As a result, this method consumes a lot of time and money.
Therefore, several objective evaluation methods, such as PSQM, PESQ,
and P862.1, are introduced. For details about the objective
evaluation methods, see the next section.





2 Objective Speech Quality Evaluation


1 PSQM (P.861) Recommendation or Algorithm

The perceptual speech quality measurement (PSQM) recommendation or
algorithm introduces the system of five quality grades, with each
grade further classified in the form of percentages through the %PoW
(Percent Poor or Worse) and %GoB (Percent Good or Better) scales.
Although the PSQM involves subclassification, it is still one of the
subjective evaluation methods. At present, someone uses a computer to
generate a wave file. Through the changes in the wave file before and
after network transmission, the quality grade is obtained to evaluate
the speech quality. In 1996, the PSQM was accepted as Recommendation
P.861 by the ITU-T. In 1998, an optional system based on measuring
normalizing blocks (MNBs) was added to P.861 as an attachment.


2 PESQ (P.862) Recommendation or Algorithm

Jointly developed by British Telecom and KPN, the Perceptual
Evaluation of Speech Quality (PESQ) was accepted as ITU-T
Recommendation P.862 in 2001. The PESQ compares an original signal
with a degraded signal and then provides an MOS. The MOS is similar
to the result of a subjective listening test. The PESQ is an
intrusive test algorithm. The algorithm is powerful enough to test
both the performance of a network element (NE) such as decoder and
end-to-end speech quality. In addition, the algorithm can give test
results by degradation causes, such as codec distortion, error,
packet loss, delay, jitter, and filtering. The PESQ is the industry's
best standard algorithm that has been commercially used.

Figure 1 shows the PESQ process.



PESQ process

For both the PSQM and the PAMS, a speech reference signal should be
transmitted on the telephone network. At the other end of the
network, the sample signal and the received signal should be compared
through the use of digit signal processing so that the speech quality
of the network can be estimated. The PESQ incorporates the advantages
of both the PSQM and the PAMS. It improves the VoIP and hybrid end-to-
end applications and modifies the MOS and MOS-LQ calculation methods.
Initially, these methods are used to measure the coding algorithm.
Afterwards, they are also used to measure the VoIP network system.


3 P862.1 Recommendation (Mapping Function for Transforming)

The perceptual evaluation of speech quality (PESQ) is a method of
objectively evaluating the speech quality of the communication
network. It is developed on the basis of the PSQM+ and PAMS. In
February 2001, the PESQ was accepted as ITU-T Recommendation P.862.
Afterwards, P.862.1 (mapping function for transforming) was added.
Not an independent protocol, P.862.1 is only the mapping of P862.
P.862.1 simulates the human ear's perception of speech more exactly
than P.862. Therefore, P.862.1 is more comparable to a subjective
listening test than P.862. The high scores obtained according to
P.862.1 are higher than those obtained according to P.862. The low
scores obtained according to P.862.1 are lower than those obtained
according to P.862. The watershed is at the score of 3.4. Therefore,
according to P.862.1, the percentage of MOSs above 3.4 should be
increased to enhance end users' experience.

The following is the formula to translate P.862 scores into P.862.1
scores:





Mapping between P862 and P862.1


4 P.563 Recommendation

The P.563 Recommendation was prepared by the ITU in May 2004. As a
single-end objective measurement algorithm, P.563 can process only
the received audio streams. The MOSs obtained according to P.563 are
spread more widely than those obtained according to P.862. For an
accurate result, several measurements should be performed and the
scores should be averaged. This method is not applicable to
individual calls. If it is used to measure the QoS of several calls,
a reliable result can be obtained.

Figure 3 shows the overall speech quality prediction of P.563.



Overall speech quality prediction of P.563


3 Speech Processing of Involved NEs

This section introduces the speech processing of all the involved
network elements (NEs): MS, BTS, BSC, and UMG. Faulty speech
processing of any one of the NEs will affect the speech quality.

Accordingly, four transmission procedures are involved in the
transmission of speech signals. The transmission procedures are Um-
interface transmission, Abis-interface transmission, Ater-interface
transmission, and A-interface transmission. Faults in any one of the
transmission procedures will lead to bit errors. Therefore, if a
speech-related problem occurs, the four NEs and the four transmission
procedures should be troubleshoot.

If the problem occurs on the Um interface, the transmission quality
on the Um interface should be optimized. If the problem occurs on the
other interfaces, the fault should be located on the basis of the bit
error rate (BER). The BSC6000 can perform BER detection.

Figure 4 takes the DSLA as an example to illustrate a typical MOS
test process.



Typical MOS test process





1 MS

Figure 5 shows the speech processing on the MS side.


Speech processing on the MS side



2 BTS

On the BTS side, the TMU performs speech exchange with the BSC, and
the DSP performs speech coding/decoding. Figure 6 shows the speech
processing on the BTS side.



Speech processing on the BTS side


3 BSC

The BSC modules other than the GTCS perform transparent transmission
on the speech signals. Instead of participating in the speech
coding/decoding, these modules are only responsible for the
establishment of the speech channel, wiring, and speech connection.
For the transparent transmission process, see the BSC6000 speech
process figure.


1 FTC Processing on Speech

Coding/decoding is performed on the speech signals and rate
adaptation is performed on the data signals so that the communication
between a GSM subscriber and a PSTN subscriber is realized and the
transparent transmission on the SS7 signaling over the A interface is
implemented.



Handling process in the GTCS


2 FTC Loopback

In a loopback, a message is transmitted by a transmission device or
transmission channel and then is received by the same to check the
health of the hardware and the settings of the software parameters.
The FTC loopback is one of the most commonly used method for locating
the transmission problems and for checking whether the settings of
the trunk parameters are accurate.




4 UMG

The UMG performs the coding/decoding conversion. Different
coding/decoding algorithms have different impacts on the speech
quality. If the communication is performed between different
networks, if the MSs use different coding/decoding algorithms, or if
the same coding/decoding uses different rates to perform
communications, the coding/decoding conversion is required.
Generally, the UMG8900 coding/decoding algorithm uses the codec
cascading to perform speech conversions. As shown in Figure 8, codec
A is cascaded with codec B. First, the compressed code stream is
restored to the PCM linear code through the corresponding decoder.
Then, the PCM linear code is encoded through another coding/decoding
algorithm. The codecs involve lots of redundancy operations, so the
speech quality is degraded to some extent.


Codec cascading


Factors That Affect the MOS in GSM

The MOS is affected by many factors, such as the background noise,
mute suppression, low-rate coder, frame error rate, echo, mobile
terminal (MS). Here, the frame error rate pertains to the frame
handling strategy (handling of frame loss during signaling
transmission), frame stealing, bit error, handover, and number of
online subscribers (congestion degree). During the speech
propagation, several NEs participate in the speech handling: MS, BTS,
TC, and MGW. The following paragraphs describe the impact of each NE
on the speech quality.


1 Introduction to GSM Speech Acoustic Principles

In a radio network, the basic processing of speech data involves
source sampling, source coding, framing, Um-interface radio
transmission, internal NE processing, handover, terrestrial
transmission, and source decoding at the receive end.

A fault in any segment of the speech transmission will result in bit
errors, thus leading to poor speech quality.

For the wireless communication system, the speech quality is
significantly affected by the Um interface, that is, the radio
transmission part. An intrinsic characteristic of radio transmission
is time-variant fading and interference. Even for a normally
functioning network, the radio transmission characteristics are
changing from time to time. For a radio network, the radio
transmission has a great impact on the speech quality. A speech
signal is transmitted to the BSS system over the Um interface. Then,
the signal is transmitted within the BSS system through the standard
and non-standard interfaces. The process requires the transmission
lines to be stable and the port BER to be lower than the predefined
threshold. If a transmission alarm is generated, the related speech
transmission lines should be checked. If the speech quality is poor,
a port BER test should be conducted.


2 Impact of Field Intensity and C/I on the Speech Quality

For the wireless communication system, the speech quality is
significantly affected by the Um interface, that is, the radio
transmission part. An intrinsic characteristic of the radio
transmission is time-variant fading and interference. Even for a
normally functioning network, the radio transmission characteristics
are changing from time to time. For a radio network, the radio
transmission has a great impact on the speech quality.

If the changes in the signal field intensity do not cause the BER/FER
to be greater than zero, the RXQUAL remains zero. In this case, the
speech quality is not affected theoretically. If the changes in the
signal filed intensity cause the BER/FER to be greater than zero
(equivalently some interference exists), the C/I and the field
intensity have a great impact on the MOS.

Both the in-network interference and the out-network interference may
affect the C/I and the receive quality and degrade the demodulation
capability of the BTS. This will lead to continuous bit errors and
faulty parsing of speech frames. Thus, frame loss may occur, causing
adverse effect on the speech quality.


3 Impact of Handover on the Speech Quality

The GSM network uses hard handovers, so a handover from a source
channel to a target channel definitely causes loss of downlink speech
frames on the Abis interface. Therefore, audio discontinuity caused
by handovers is inevitable during a call. Hence, the handover
parameters should be properly set to avoid frequent handovers. In
addition, the audio discontinuity caused by handovers should be
minimized to improve the speech quality.


4 Impact of DTX on the Speech Quality


1 If the DTX is enabled for a radio network, comfort noise and voice
activity detection (VAD) are introduced. Affected by the background noise
and system noise, the VAD cannot be totally exact. This definitely leads to
the clipping of speech signals. Thus, the loss of speech frames and the
distortion of speech may occur, and the speech quality and MOS test may be
greatly affected. When the Comarco device marks a speech score, the
statistics on the clipping are collected. Generally, the value of the
clipping has a positive correlation with the clipped portion of speech.
Therefore, if the intrusive algorithm is used, the MOS is definitely low.

Table 2 describes the result of the lab test.


Impact of DTX on the speech quality
"Impact of DTX on the Speech Quality "
" "
"FR "1. If the uplink DTX of the FR is enabled, the PESQ decreases "
" "by about 0.053 on average. Varying from sample to sample, the "
" "decrease of PESQ ranges from 0.03 to 0.08. "
" "2. If the downlink DTX of the FR is enabled, the PESQ "
" "decreases by about 0.054 on average. Varying from sample to "
" "sample, the decrease of PESQ ranges from 0.02 to 0.12. "
" " "
" " "
"FAMR12.2"1. If the uplink DTX of the FAMR12.2 is enabled, the PESQ "
" "decreases by about 0.05 on average. Varying from sample to "
" "sample, the decrease of PESQ ranges from 0.01 to 0.33. "
" "2. If the downlink DTX of the FAMR12.2 is enabled, the PESQ "
" "decreases by about 0.08 on average. Varying from sample to "
" "sample, the decrease of PESQ ranges from 0.02 to 0.20. "
" " "
" " "
" " "
"HAMR5.9 "1. If the uplink DTX of the HAMR5.9 is enabled, the PESQ "
" "decreases by about 0.018 on average. Varying from sample to "
" "sample, the decrease of PESQ ranges from 0.01 to 0.07. "
" "2. If the downlink DTX of the HAMR5.9 is enabled, the PESQ "
" "decreases by about 0.079 on average. Varying from sample to "
" "sample, the decrease of PESQ ranges from 0.05 to 0.11. "
" " "
" " "




5 Impact of Speed (Frequency Deviation) on the Speech Quality

Generally, at a speed of 200 km/h, the BER increases and the speech
quality deteriorates because of multi-path interference. If the speed
is increased to 400 to 500 km/h, a certain frequency deviation occurs
in the signals received by the BTS from the MS because of the Doppler
effect. The uplink and downlink frequency deviations may accumulate
to 1,320 Hz to 1,650 Hz. Thus, the BTS cannot correctly decode the
signals from the MS.

With the development of high-speed railways and maglev trains, mobile
operators pay increasing attention to the speech quality in high-
speed scenarios. In 2007, Dongguan Branch of China Mobile requested
Huawei to optimize the speech quality for the railways in Dongguan
under the coverage of Huawei equipment. After optimizing the speech
quality, Huawei enabled the HQI (HQI indicates the percentage of
quality levels 0-3 to quality levels 0-7 in the measurement report)
to be 97.2%, which is the competitor's level. In addition, the
highest HQI reached 98.5%. The percentage of SQIs distributed between
20 and 30, however, is only 40% and that distributed between 16 and
20 is also only 40%. The distribution of the highest SQIs is sparser
than that (about 90%) with the same speech quality at a low speed.
Therefore, high speed greatly affects the speech quality. Ensure that
the speed is stable during acceptance tests or comparative tests.





6 Impact of Speech Coding Rate on the Speech Quality

The speech coding schemes are HR, FR, EFR, and AMR.

Each speech coding scheme maps to an MOS. Table 3 lists the mapping
between the speech coding scheme and the MOS value.


Mapping between the speech coding scheme and the MOS value






7 Impact of Transmission Quality on the Speech Quality

Generally, if the transmission quality is poor, the BER and the slip
rate are high and the transmission is intermittent. The statistics on
OBJTYPE LAPD involve the retransmission of LAPD signaling, LAPD bad
frame, and overload. These counters are used to monitor the
transmission quality on the Abis interface. If too many bad frames
are generated or if the signaling retransmission occurs frequently,
the transmission quality is probably poor. From the perspective of
principle, poor transmission quality is equivalent to the loss of
some speech frames. If the speech frames are lost, the speech quality
deteriorates greatly.




Method of Analyzing the Problem of Low MOS


1 Process of Analyzing the Problem of Low MOS

The MOS aims at an end-to-end communication. The communication
involves many NEs and interfaces. The fault in any NE or interface
will cause high BER, thus leading to low MOS. If the MOS is low, the
involved NEs and interfaces should be checked in succession.

Figure 9 shows the fault location flow.



Fault location flow


2 Method of Solving the Problem of Low MOS


1 Consistency Check and Sample Check

The consistency check involves the test devices, the MSs that serve
the test devices, and the grading standards adopted by the test
devices. Different test devices adopt different grading standards and
are served by different MSs. These differences lead to various
combinations, which will definitely cause differences in the opinion
scores. Even if the same device uses different grading standards, the
difference in the opinion scores is large. For example, if you use
the Comarco and DSLA to test the speech quality of the same speech
code, the MOS with the Comarco is lower than the MOS with the DSLA.

The Comarco and the DSLA adopt different grading standards, test
samples, and test MSs.



If the test samples are different, the test results differ
irrespective of whether the environment (for example, shielded
cabinet in non-interference environment), MS, wireless equipment,
core network equipment, and parameter setting are the same.
Therefore, the speech samples for the speech tests before and after
the network replacement must be the same. The following table lists
the mapping between the speech sample and the MOS. According to Table
4, the MOS varies according to the speech sample. The tests of a
large number of speech samples show that American English has the
highest MOS, German has the second highest MOS, and Spanish has the
third highest MOS.


Mapping between speech sample and MOS
"Network "Speech "MOS "
"Type "Sample " "
"900M "French "3.4 "
"900M "Italian "3.46 "
"900M "Arabic "3.5 "
"900M "Russian "3.54 "
"900M "Japanese"3.54 "
"900M "Greek "3.57 "
"900M "Spanish "3.59 "
"900M "German "3.61 "
"900M "American"3.64 "
" "English " "




2 Um Interface Check

The GSM speech codes use the Un-equal Error Protection (UEP)
mechanism. Figure 10 shows the data transmission and clipping.
The differences between the speech data transmission on the air
interface of GSM and that of WCDMA/CDMA2000 are as follows:
Cyclic redundancy check (CRC): For the GSM, the CRC of the full-rate
TCH checks only three bits. The error check capability of the GSM is
far weaker than that of the CDMA2000 and WCDMA. For the GSM, the CRC
of the enhanced full-rate TCH checks ten bits. The error check
capability of the GSM is close to that of the 3G.
Error correction coding: For the GSM, sub-stream C does not have
error correction coding, so the error probability is large.
Power control: The GSM does not have fast power control. Therefore,
the burst fading or interference cannot be resisted and the errors in
the radio transmission cannot be reduced quickly. Power control
improves the speech quality by reducing the BER and FER.

9. Speech data transmission on the Um interface
(schematic drawing)

Like the CDMA2000, the GSM also uses the frame stealing method to
transmit some signaling. The frame stealing method has an impact on
the speech quality. If continuous frame stealing occurs, the speech
quality is greatly affected.
In the GSM system, if the full-rate speech coding is used, the CRC of
sub-stream A checks only three bits and the error check capability is
limited. The errors that cannot be detected through the CRC also
affect the speech quality. Hence, the speech quality can be reflected
only when the measurement of the remaining bit error rate (RBER) is
performed.
The RBER cannot be measured, but the GSM system provides an
alternative method, that is, to measure the demodulation BER. In
other words, first, perform error correction on the demodulation
result; second, encode the obtained result; third, compare the
demodulation result with the encoded result. Thus, the BER in the
radio transmission can be reflected indirectly. The standard
measuring value that corresponds to BER is RXQUAL. Therefore, for
high speech quality, the BER must be reduced and the receive quality
on the Um interface must be improved.
For the enhanced full rate (EFR), the statistics of FER can basically
reflect the speech quality because the 10-bit CRC is used.
From the perspective of the Um interface, the factors that affect the
speech quality are sub-stream A, BER (or RXQual), and frame stealing.
Only RxQual, however, can solve the problem of poor speech quality
through network optimization.


1 Coverage- and Interference-Related Problem Check

If the network coverage is poor, it is definite that many areas in
the network have poor receive quality. Therefore, the speech quality
is affected.

The interference leads to an increase of BER on the radio link. The
increase may exceed the demodulation capacity of the BTS so that
speech frames cannot be identified. Thus, the speech frames may be
lost and thus the speech is discontinuous.

To solve the two types of problems, refer to the corresponding guide:



G-Guide to Eliminating Interference - 20050311-A-1.0

G-Guide to Analyzing Network Coverage - 20020430-A-1.0


2 Low MOS due to Handovers

Low MOS is caused by not only frequent handovers but also the
following factors.

1. The GSM network uses hard handovers, so a handover from a source
channel to a target channel definitely causes loss of downlink speech
frames on the Abis interface. As a consequence, audio discontinuity
caused by handovers is inevitable during a call. Therefore, the
handover-related parameters must be checked to avoid frequent
handovers.

2. The handover is not reasonable. For example, a call is handed over
to a cell with poor quality because of configurations, and thus the
MOS is low.
3. The parameter settings are improper, so the handover is slow. If
the QoS of the serving cell is poor for a long time, the speech call
cannot be handed over to a better neighboring cell in time. Thus, the
speech quality is always poor, leading to low MOS, handover failure,
and call drops.
4. Some networks disable the bad quality handover, so the MOS is low.


5. The intra-cell handover is configured as asynchronous handover, so
the connection on the Um interface is long, leading to low MOS.




3 Occupation Ratios of Half Rate and Low AMR Rate

All the MOS tests using the PESQ algorithm adopt intrusive speech
scores, which are process values. If the existing network has several
types of speech coding, the conduct of speech quality DT test or CQT
test leads to channel handovers and AMR speech coding rate handovers.
Several types of speech coding may be involved in the speech grading
process. Therefore, the network speech quality test is performed on
different types of speech coding. The speech quality test value of
the high coding rate is low, and the speech quality test value of the
low coding rate is high. When the transmission quality on the Um
interface is stable, the MOS is low if the occupation ratio of the
half rate is high. Therefore, the full rate and the high AMR rate
coding are recommended.




3 BTS Check


1 Software Version Check

Check for the version-related problems that have been detected.
The old BTS uses a too early version and is incompatible with the new
BTS, so the speech problems occur.

2 Whether the Uplink and Downlink DTX Function Is Enabled

DTX means VAD and silent frames. Replacing the speech with silent
frames is a kind of distortion, which brings about difficulties for
all the perceptual models to predict the MOS. Generally, the 50ms
clipping (VAD) at the front end and rear end does not have a great
impact on the subjective impression. In the case of clipping during
the speech, however, replacing the speech with silent frames after
the packet loss significantly affects the subjective impression. If
50 ms is lost, the MOS is decreased by one. For the PESQ, each 50ms
clipping generally leads to the decrease in the MOS of about 0.5,
irrespective of the location. The VAD cannot be 100% correct, so the
speech quality definitely deteriorates if the uplink and downlink DTX
function is enabled during the MOS test.


3 Hardware Factors

The audio discontinuity caused by BTS hardware fault affects the MOS. Bugs
in the speech processing part of the hardware also affect the speech
quality. You are advised to confirm with the R&D personnel that no
identified problems exist in the version.

4 Abis Transmission Check

The networks built by Huawei cover many parts of the world. The
development levels of the basic communication and data communication
vary from region to region. In addition, the cost of investing and
leasing the transmission lines is high. Therefore, different regions
use different transmission types: microwave transmission, circuit
transmission, optical transmission, and satellite transmission. Here,
the quality of microwave transmission is very prone to weather
conditions. Different BERs of different transmission types definitely
lead to different transmission quality. Therefore, different networks
of different mobile operators should be compared on the basis of the
same transmission type.

The alarms to be checked include Broken LAPD Link and Excessive Loss
of E1/T1 Signals in an Hour.

In addition, the Monitoring the Port BER function of the BSC and BER
tester (E7580A) can be used to check whether the Abis interface has
bit errors.


5 BSC Check


1 Whether the TFO and EC Functions Are Enabled

During a call from an MS to another, if the calling MS and called MS
use the same speech service type, the times of speech coding/decoding
can be reduced by one through in-band signaling negotiation. Thus,
the speech quality can be improved. When the EC function is enabled,
the speech quality can be improved if the echo occurs during the
call. If there is no bit error, enabling the TFO function can improve
the speech quality by more than 0.25 score.


Impact of TFO on the improvement of speech quality (GSM Rec. 06.85)
"DMOS "EP0 "EP1 "EP2 "
"HR ".85 ".68 ".39 "
"FR ".53 ".53 ".35 "
"EFR ".32 ".46 ".19 "




2 Whether Local Switch Is Enabled

The local switch consists of BSC local switch and BTS local switch.
For the BSC local switch, the calling MS and called MS should be
served by the same BSC. Thus, the Ater interface and local
transmission resources are saved. For the BTS local switch, the
calling MS and called MS should be served by the same BTS or BTS
group. Thus, the Ater interface and Abis interface transmission
resources are saved. When the BSC local switching is used, the TC
coding/decoding is not required if the transcoding function is
implemented in the core network, thus improving the speech quality.
When the BTS local switching is used, the TC coding/decoding is not
required because the speech signals do not pass the BSC. This also
improves the speech quality.


6 A Interface Transmission Check

The rules for checking the A interface transmission is similar to
those for checking the Abis interface transmission. You can refer to
the section Abis Transmission Check.

To check the A interface transmission, you have two methods: first,
query the BSC alarms (for example, the Loss of E1/T1 Signals alarm)
to determine whether intermittence occurs on the A interface; second,
use a BER tester to check whether bit errors occur on the A interface
transmission.


7 MGW Check

If this problem does not occur when you use an MS to call another MS
during the MOS test, you can skip this section.

As is mentioned in section UMG, if the communication is performed
between different networks, if the MSs use different coding/decoding
algorithms, or if the same coding/decoding uses different rates to
perform communications, the coding/decoding conversion is required.
The inter-code conversion, however, may adversely affect the speech
quality.

Therefore, if you use an MS to call a fixed-line phone during the MOS
test, you should check whether the deterioration of the speech
quality is caused by the following: whether the route between the MS
and the fixed-line phone passes through two UMGs and whether the two
UMGs use the speech compression algorithm.


8 Miscellaneous (Comparison of MOS Before and After Network Replacement)

In a network replacement project, if the MOS deviation occurs before
and after the network replacement, the following factors should be
considered:


1 Test Speed

Generally, the drive speed should be stable (at about 30 km/h) during
the test. If the drive speed is low, the test is equivalent to the
fixed-point CQT test and thus the test result is high.

In addition, if the drive speed is high (at more than 200 km/h), the
generated frequency deviation affects the speech quality. In this
case, the BTS frequency deviation algorithm should be enabled to
improve the speech quality.


2 Test Route and Test Time

The DT test of speech quality objectively reflects the coverage and
receive quality of a network. In a network, it is definite that some
areas have good speech quality and other areas have poor speech
quality. During the DT test of speech quality, the trunk coverage
lines of the target network should be tested completely and the
important branch lines should also be tested. A test route should not
be tested repeatedly. If you test the areas with good speech quality
repeatedly, the speech quality in the DT test becomes high. If you
test the areas with poor speech quality repeatedly, the speech
quality in the DT test becomes low.

You should also check whether the test time is consistent. In
different periods, the traffic models of the existing network are
different. The busy traffic hours in each day occur regularly.
Therefore, the congestion during traffic peaks is heavy, thus causing
more in-network interference. According to the statistics about the
receive quality on the Um interface, the receive quality deteriorates
during busy hours and the corresponding SQI decreases. Therefore, to
ensure the test consistency, you are advised to choose the same test
period.

For example, Huawei has conducted comparison tests at 4:00 a.m. and
9:00 p.m (busy hour) in Tieling. The results show that the QoS on the
Um interface in the early morning is very good and that during busy
hours is very poor. Accordingly, the speech quality in the early
morning is good and that during busy hours is poor. Therefore, the
same test periods should be selected for the comparison test.


3 Frequency Reuse Degree

For mobile communications, frequency is the most important resource.
With the rapid development of mobile communications, the number of
mobile subscribers increases sharply. To meet the increasing capacity
requirements, all the mobile operators try to raise the frequency
reuse degree within their own frequency bands. The increase of the
frequency reuse degree, however, definitely brings about large
network interference. If the frequency reuse degree is high, the
interference is strong. Thus, the network quality is poor and the
speech quality is poor. This may adversely affect the user
experience. Therefore, the speech quality of the mobile operators
with different frequency reuse degrees cannot be compared directly.
For example, China Unicom adopts a plan with high frequency reuse
degree to reach the same cell configuration of BTSs for China Mobile,
so the speech quality of China Unicom is definitely lower than that
of China Mobile. In a word, if the frequency reuse degree is high,
the test MOS is low.


4 Engineering Installation Quality Issues

According to the experience, check that the connector (on the DDF) on
each transmission segment is properly connected and that there are no
exposed stubs. For optical transmission, check that optical connector
is clean and that the transmission BER is not high.
The poor engineering quality in the antenna system also causes the
MOS to decrease. The speech quality may deteriorate because of errors
in engineering installation, for example, loose connector,
misconnection, or poor coverage.



Test Methods and Suggestions


1 Test Tool Selection and Test Suggestions

1. Normally, the test tools are selected according to the
requirements of the mobile operators. At present, China Mobile
accepts the PESQ as the evaluation standard of the existing network
and Ding Li or Hua Xing as the test tool. The overseas mobile
operators use different evaluation standards and use such test tools
as DSLA, Cormarco, and QVOICE.
2. During the bidding, the acceptance standard, test tool, speech
sample, acceptance area (recommended to exclude the suburb areas with
poor coverage), calling method, test duration, test time, and test
route are determined for the convenience of future acceptance.

2 Suggestions on the Test of the Existing Network

1. It is recommended that you use short call samples as the test
samples to avoid some blind areas or poor-coverage areas. For the
network that has good coverage and that does not require frequent
handovers, long call samples are recommended.
2. Both Nokia6680 and Samsung zx10 can be used as the test MSs. Note
that Nokia6680 does not support half rate and has outdoor antenna
(no vehicle body loss) and that Samsung zx10 supports half rate
and does not have outdoor antenna. In the case of outdoor antenna
(vehicle body loss should be considered), it is recommended that
Nokia6680 be used as the test MS.
3. The areas with good coverage and only a few handovers should be
selected as the test routes.
4. During the test, it is recommended that you use an MS to call a
fixed-line phone. Thus, the MOS is high.
5. The DTX function should be disabled.
6. The drive speed during the drive test should not be too high.
7. It is recommended that the idle hours be selected as the test
time. Thus, the network C/I is high.
8. During the test, it is recommended that the channels with good
speech coding quality be occupied, for example, EFR and AMR full-
rate channels.
9. The TFO function should be enabled if the version is correct. Note
that the TFO function is valid only for the call from an MS to
another.



MOS Cases


1 Differences Between Speech Signal Process and Signaling Process


1 GSM Speech Signal Process

MS-BTS - GEIUB-GTNU-GEIUT-GEIUT- GTNU-GDSUC-GTNU-GEIUA-MSC…MS



BSC6000 speech signal process


2 Signaling Process

MS-BTS - GEIUB-GGNU-GXPUM -GGNU-GEIUT-GEIUT-GTNU-GEIUA –MSC…MS

Here, the internal BSC signaling process contains the signaling
handling process on the Ater interface, which is omitted in this
document.

The previous process indicates that the speech signal process and the
signaling process are different in terms of the path. The measurement
of KPIs is mainly performed at the signaling measurement points in
the calling process. The speech MOS indicates the audio experience of
the end user. The signaling process and the speech signal process are
different. Therefore, if the KPIs are good, the MOS is not definitely
high. Good KPI is only a necessary condition of high MOS. The speech
MOS is closely related to the transmission quality on the Um
interface, interference, C/I, frame erase ratio (FER), SQI, and SNR.





2 Identified MOS Problems

After the handling of MOS problems on the existing network and the
crisis handling of the speech MOS, some devices of Huawei that affect
the MOS are detected. If the MOS of the existing network is low and
if the problem of low MOS cannot be solved after optimization, you
can refer to the Problem Description column in the following table to
check whether the version is incorrect.

Table 6 lists only the problem-solved versions. To check whether the
onsite version is correct, consult the product maintenance
department.


Identified MOS problems



"Problem"Problem "Problem Description "Relate"Affected "Problem-Solve"
"Number " " "d "Channel "d Version "
" " " "Produc" " "
" " " "t " " "
"1 "In the case of"The frame loss on the "DPU(TC"FAMR/HAMR/"V9R8C01B048SP"
" "FAMR/HAMR and "uplink during the ") "FR "01 "
" "FR, one frame "FAMR/HAMR and FR speech" " " "
" "is lost and "leads to a sharp " " " "
" "then the frame"decrease in the MOS. " " " "
" "is " " " " "
" "retransmitted." " " " "
"2 "In case of "The frame loss on the "DPU(TC"EFR/HR "V9R8C01B048SP"
" "frame loss "uplink during the ") " "01 "
" "during a "EFR/HR speech leads to " " " "
" "handover, the "a sharp decrease in the" " " "
" "smoothness "MOS. " " " "
" "handling " " " " "
" "performed on " " " " "
" "the signals " " " " "
" "over the " " " " "
" "EFR/HR " " " " "
" "channels does " " " " "
" "not take " " " " "
" "effect. " " " " "
"3 "Random bit "When the TFO is "DPU(TC"EFR/FR/HR "V9R8C01B048SP"
" "errors when "established, the MOS is") " "01 "
" "TFO "lower than the expected" " " "
" "established "value and there are " " " "
" " "random bit errors. " " " "
"4 "Permanent loss"The uplink DTX is "DPU(TC"HAMR7.4 "V9R8C01B048SP"
" "of one frame "enabled in the case of ") " "01 "
" "during "HAMR7.4. During the " " " "
" "handover to "transition from " " " "
" "half rate and "non-speech to speech, " " " "
" "permanent loss"the MOS is decreased by" " " "
" "of one frame "one frame. " " " "
" "during " " " " "
" "activation " " " " "
" "under HAMR " " " " "
" "7.4k " " " " "
"5 "The uplink DTX"The uplink DTX is "DPU(TC"EFR/HARM6."V9R8C01B048SP"
" "is enabled and"enabled in the case of ") "7/HARM7.4 "01 "
" "the speech "EFR and HAMR. During " " " "
" "quality under "the transition from " " " "
" "EFR and HAMR "non-speech to speech, " " " "
" "obviously "the MOS is decreased by" " " "
" "deteriorates. "one frame. " " " "
" "Damage is " " " " "
" "introduced on " " " " "
" "the TC side. " " " " "
"6 "The internal "If a call is made "DPU(TC"All the "V9R8C01B048SP"
" "clock is slow."repeatedly on the same ") "speech "01 "
" "External "channel, audio " "channels " "
" "interruption "discontinuity occurs. " " " "
" "should be used" " " " "
" "to locate the " " " " "
" "period of 20 " " " " "
" "ms. " " " " "
"7 "SID_FIRST "In the test speech "DSP "FAMR " "
" "frame for FAMR"sample, two SP frames "(BTS) " "V100R008C02B2"
" " "contain the SID_FIRST " " "01 or "
" " "frame. In this case, " " "V100R001C07B4"
" " "the BTS misinterprets " " "15 "
" " "and discards the first " " " "
" " "speech frame after the " " " "
" " "SID frame. Thus, the " " " "
" " "MOS decreases. " " " "
"8 "SID_FIRST_INH "In the test speech "DSP "HARM " "
" "frame for HAMR"sample, two SP frames "(BTS) " "V100R008C02B2"
" " "contain the " " "01 or "
" " "SID_FIRST_INH frame. In" " "V100R001C07B4"
" " "this case, the BTS " " "15 "
" " "reports the " " " "
" " "SID_FIRST_INH frame as " " " "
" " "the NO_SP frame. Thus, " " " "
" " "the TC misinterprets " " " "
" " "and discards the first " " " "
" " "speech frame after the " " " "
" " "NO_SP frame. As a " " " "
" " "result, the MOS " " " "
" " "decreases. " " " "
"11 "Frequent "After the uplink DTX is"DSP "HARM " "
" "adjustment to "enabled, the adjustment"(BTS) " "V100R008C02B2"
" "downlink rate "(adjustment is made " " "01 or "
" "when uplink "when silent frames are " " "V100R001C07B4"
" "DTX enabled "transmitted and " " "15 "
" " "adjustment is not made " " " "
" " "when speech frames are " " " "
" " "transmitted) is made on" " " "
" " "the downlink coding in " " " "
" " "the case of half-rate " " " "
" " "AMR multirate set. If " " " "
" " "the DTX is disabled, " " " "
" " "however, a fixed rate " " " "
" " "is always occupied. " " " "
" " "Therefore, the " " " "
" " "adjustment is not " " " "
" " "caused by the C/I. " " " "
"12 "Reporting of "During the synchronous "DSP "All the " "
" "HO_DET ahead "handover, the HO_DET is"(BTS) "speech "V100R008C02B2"
" "of time during"reported ahead of time." "channels "01 or "
" "synchronous "Thus, the uplink speech" " "V100R001C07B4"
" "handover "frames on the old " " "15 "
" " "channel are lost and " " " "
" " "the handover disruption" " " "
" " "is long. The occurrence" " " "
" " "possibility of this " " " "
" " "problem during the lab " " " "
" " "test is about 5%-10%. " " " "
"13 "One speech "During the intra-BSC "DSP "All the " "
" "frame lost on "asynchronous handover, "(BTS) "speech "V100R008C02B2"
" "old channel "one frame out of the " "channels "01 or "
" "during "uplink speech frames is" " "V100R001C07B4"
" "asynchronous "lost. This problem " " "15 "
" "handover "occurs on the three " " " "
" " "types of MSs. The " " " "
" " "occurrence possibility " " " "
" " "of this problem during " " " "
" " "the lab test is about " " " "
" " "30%-50%. " " " "










Feedback on MOS or Speech Problems

To better compare the network quality before and after the network
replacement, a comprehensive test should be conducted before the
network replacement and the trunk roads, important branch roads, and
important public places in the original network must be tested. A
test report on the original network should be provided. The test
report should include the following contents: RxQual (including the
mean values, peak values, and mean square errors), SQI (including the
mean values, peak values, and mean square errors), C/I (including the
mean values, peak values, and mean square errors), test route and
speed, and dotted output figure (the dotted contents should be
provided on the basis of the previous three counters).


1 Test Requirements

1. Test time and periods: The test must be conducted at 9:00-12:00 and
17:00-20:00 on workdays (Monday through Friday).

2. The test routes must evenly cover the trunk roads in the urban areas
without repeated coverage. The round-the-city express ways, viaducts,
and roads between the urban areas and the air port must be tested.

3. In the urban areas, the test speed should equal the normal drive speed.
No limitation is set on the test speed.

4. Irrespective of the traffic, the city with a population of more than 500
thousand should be tested for three days and the city with a population
of more than 200 thousand should be tested for two days. The test should
last six hours for each test day.

5. Dialing requirements:

The test MSs should be located inside the vehicle and both the
calling MS and called MS should be connected to the test
instruments. The GPS receiver should be connected to conduct
the test.
Both the GSM calling MS and called MS for the test should be
of auto dualband.
The MSs should be dialed mutually. The dialing, answering, and
onhook of the MSs should be automatic. Each call should last
180 seconds with a call interval of 20 seconds. If call failure
or call drop occurs, another call attempt should be made after
20 seconds. The call interval is set according to the
requirements of the mobile operator.
6. Daemon data analysis: All the tests must use the same test instruments
and Daemon data processing software.

7. Normally, the test tools are selected according to the requirements of
the mobile operators. At present, China Mobile accepts the PESQ as the
evaluation standard of the existing network and Hua Xing as the test
tool. The overseas mobile operators use different evaluation standards
and use such test tools as SwissQual, QVoice, and Cormarco.


8. The evaluation of the Um interface on the existing network should be
complete and the statistics on RxQual, C/I, and SQI should be provided.
The three counters should have the mean values, peak values, mean square
errors in different periods, and distribution interval list of different
values. During the test, the GPS should be dotted and the log files of
the TEMS test should be archived.

9. When the network of several cities is replaced, the speech
problems should be reported. For different cities, the test should
be conducted according to the different requirements mentioned in
this chapter. The test reports should be archived. The dot
information about the local e-map should be provided for the
future network optimization of the areas with poor quality.
During each test, the mean speed per hour should be recorded and
archived. Dot statistics can be performed on the GPS.

2 Requirements for Configuration Data in Existing Network

The QoS of the existing Huawei network varies according to the
economic development degree, network coverage, network user
density, network density, network planning, frequency reuse
degree, and external interference in the local area. Networks with
different QoSs have different configurations and different
configurations have different impacts on the network. For the R&D
personnel to learn the existing network, the configurations of the
existing network should be provided.

Table 7 lists the network configuration parameters that should be
provided.


Network configuration parameters to be provided
"Network Configuration "Test Result "
"Uplink/downlink DTX " "
"UL PC Allowed " "
"DL PC Allowed " "
"Radio frequency hopping " "
"Baseband frequency hopping " "
"Transmit diversity " "
"TFO " "
"EC " "
"Whether the core network uses IP " "
"bearing " "
"Transmission mode of each " "
"interface " "
"Frequency resources " "
"Configuration of main BTS models " "
"Setting of the handover threshold" "
"Setting of the power control " "
"threshold " "
"Setting of the coding rate and " "
"the use proportion " "
"RxQual in the drive test of the " "
"entire network " "
"SQI in the drive test of the " "
"entire network " "
"C/I in the drive test of the " "
"entire network " "


-----------------------
Session processing

A/D and D/A conversions

Speech coding/decoding, DTX

20ms speech frame

Sub-stream A

Sub-stream B

Sub-stream C

Sub-stream A


Sub-stream B

Sub-stream C

CRC

1/2 coding

Sub-stream C

TDMA frame