HUAWEI TECHNOLOGIES CO.,LTD.

CQI Adjustment Based on Dynamic BLER Target Trial Report

 CQI adjustment based on dynamic BLER target

Contents
1 Overview ...............................................................................................................................................

2 Basic Concepts .......................................................................................................................................

2.1 CQI Adjustment Process ........................................................................................................

2.2 Note Issue .............................................................................................................................

3 Feature Prerequisites ............................................................................................................................

3.1 Dependencies on Hardware ..................................................................................................

3.2 Dependencies on Other Features ..........................................................................................

3.3 License ..................................................................................................................................

4 Network Impact ....................................................................................................................................

5 Feature Deployment .............................................................................................................................

5.1 Data Preparation ...................................................................................................................

5.2 Activation ..............................................................................................................................

5.3 Activation Observation .........................................................................................................

5.4 Deactivation(Using MML Commands) ..................................................................................

5.5 Performance Monitoring ......................................................................................................

6 Trial Area ...............................................................................................................................................

7 Trial Result .............................................................................................................................................

7.1 DT Result ................................................................................................................................

7.2 OSS KPIs: ...............................................................................................................................

8 Conclusion ...........................................................................................................................................

1 OVERVIEW

Page 2
 CQI adjustment based on dynamic BLER target

The CQI measures the channel conditions of a UE and is reported from the UE to the NodeB.
Without this feature, the NodeB determines an appropriate TBS based on the reported CQI, system
resources, and the TFRC policy. If the reported CQI and related conditions remain the same, the NodeB
does not change the TBS because it does not consider the ever-changing radio environments.

The constant changes in radio environments, caused by multipath effects and UE mobility, lead
to fluctuating channel quality. Under these circumstances, choosing a TBS based on the reported CQI
makes it difficult to always achieve the optimum downlink throughput.

With the feature CQI adjustment based on dynamic BLER target, the NodeB monitors the
channel quality fluctuations for HSDPA users in a cell in real time and dynamically selects a proper BLER
target based on the monitoring result. The NodeB then uses the BLER target to adjust the CQI reported
by the UE. Based on the adjusted CQI, the NodeB determines an appropriate TBS to achieve higher
downlink throughput for HSDPA users and higher cell throughput.

2 BASIC CONCEPTS
2.1 CQI ADJUSTMENT PROCESS
CQI adjustment based on dynamic BLER target is performed in each TTI. The following describes
the adjustment process:

1. Based on the CQI reported by the UE, the NodeB checks the actual radio environment, which is
affected by multipath effects and UE mobility.
2. Based on the actual radio environment and channel quality of the UE, the NodeB obtains an optimum
BLER target, which helps to achieve the highest possible throughput for the UE.
3. Based on the ACK, NACK, or DTX indication from the UE in the current TTI and on the optimum BLER
target, the NodeB calculates the CQI offset, which can be a positive or negative number. The NodeB
then uses the CQI offset to adjust the CQI.
4. Based on the adjusted CQI, the NodeB selects an appropriate TBS by using the TFRC algorithm.

2.2 NOTE ISSUE

The required BLER target may be high in some environments; therefore this feature is not suitable for
networks that limit the BLER target.

This feature requires that both the network and UE support HSDPA. This feature is applicable to all
HSDPA terminals except for the terminals that are configured with MIMO. Different terminals may have
different performance for the same TB size. Some terminals may have greater BLERs. This feature
adjusts the TB size for terminals based on data transmission performance to achieve optimized
performance.

3 FEATURE PREREQUISITES

Page 3
 CQI adjustment based on dynamic BLER target

3.1 DEPENDENCIES ON HARDWARE

The BTS3812E and BTS3812AE must be configured with the EBBI, EBOI, or EDLP board.
The BBU3806 must be configured with the EBBC or EBBCd board.
The BBU3900 must be configured with the WBBPb, WBBPd or WBBPf board.

3.2 DEPENDENCIES ON OTHER FEATURES

WRFD-010610 HSDPA Introduction Package

3.3 LICENSE

4 NETWORK IMPACT
System Capacity

This feature increases the downlink throughput for HSDPA users and cells by up to 10%.

Network Performance

Calculation for adjusting the CQI increases the downlink load of the NodeB DSP slightly.

5 FEATURE DEPLOYMENT
5.1 DATA PREPARATION
Table5-1 lists the data to prepare before deploying CQI Adjustment Based on Dynamic BLER Target.

Table 5-1 Data to prepare before deploying CQI Adjustment Based on Dynamic BLER Target

Page 4
 CQI adjustment based on dynamic BLER target

5.2 ACTIVATION
Run the NodeB MML command SET ULOCELLMACHSPARA with CQI Adjust Algorithm Switch of non-
Conversational Service set to CQI_ADJ_BY_DYN_BLER(CQI Adjusted by Dynamic BLER).

//Activating CQI Adjustment Based on Dynamic BLER Target

SET ULOCELLMACHSPARA: ULOCELLID=0, CQIADJALGOFNONCON=CQI_ADJ_BY_DYN_BLER;

5.3 ACTIVATION OBSERVATION

Run the NodeB MML command LST ULOCELLMACHSPARA. In this step, ensure that CQI Adjust
Algorithm Switch of non-Conversational Service is set to CQI_ADJ_BY_DYN_BLER.

5.4 DEACTIVATION(USING MML COMMANDS)

Run the NodeB MML command SET ULOCELLMACHSPARA. In this step, set CQI Adjust Algorithm
Switch of non-Conversational Service to NO_CQI_ADJ (Not CQI Adjust Algorithm).

//Deactivating CQI Adjustment Based on Dynamic BLER Target

SET ULOCELLMACHSPARA: CQIADJALGOFNONCON=NO_CQI_ADJ;

5.5 PERFORMANCE MONITORING

This feature helps increase cell throughput and UE throughput. You can query the values of the
following counters to track changes in cell throughput:

VS.HSDPA.MeanChThroughput: an RNC counter that measures the average downlink throughput of

individual MAC-d flows for HSDPA in a cell
VS.DataOutput.Mean: a NodeB counter that measures the average cell throughput at the MAC-
hs/MAC-ehs layer within a measurement period

The values of the preceding counters increase after this feature is activated. Note that gains brought by
this feature vary in different scenarios.

Page 5
 CQI adjustment based on dynamic BLER target

It is recommended that you observe the gains brought by this feature in scenarios that meet the
following conditions:
The fluctuation in traffic volume (VS.IuPS.BytesPayldIntact.Rx + VS.IuPS.BytesPayldBgrd.Rx) is less
than 10%.
The fluctuation in the average CQI value (1 x VS.CQI1 + 2 x VS.CQI2 + … + 35 x VS.CQI35)/(VS.CQI0 +
VS.CQI1 + … + VS.CQI35) is less than 0.4 dBm.

In most cases, the gains can be observed in the scenarios that meet the preceding conditions. If the
gains cannot be observed, observe the gains based on drive test results.

6 TRIAL AREA
Cluster AIN TAYA will be used as a pilot cluster for this trail;

RNC Cluster Node B

Rouiba RNC AIN TAYA 3A16X104
Rouiba RNC AIN TAYA 3A16X132
Rouiba RNC AIN TAYA 3A16X398
Rouiba RNC AIN TAYA 3A16X524
Rouiba RNC AIN TAYA 3A16X671
Rouiba RNC AIN TAYA 3A16X705
Rouiba RNC AIN TAYA 3A16X710

Page 6
 CQI adjustment based on dynamic BLER target

7 TRIAL RESULT
7.1 DT RESULT

Date Average Throughput Avg CQI Carrier 1 Avg CQI Carrier 2

Before 3.45 25.7 25.3
After 4.02 26.4 26.2

HSDPA Throughput

Page 7
 CQI adjustment based on dynamic BLER target

HSDPA Throughput Distribution

Page 8
 CQI adjustment based on dynamic BLER target

HSDPA CQI Distribution (Carrier 1 & 2)

Page 9
 CQI adjustment based on dynamic BLER target

HSDPA ACK Distribution (Carrier 1 & 2)

Page 10
 CQI adjustment based on dynamic BLER target

HSDPA NACK Distribution (Carrier 1 & 2)

Page 11
 CQI adjustment based on dynamic BLER target

HSDPA Retransmission Distribution (Carrier 1 & 2)

Page 12
 CQI adjustment based on dynamic BLER target

7.2 OSS KPIS:

Page 13
 CQI adjustment based on dynamic BLER target

8 CONCLUSION
CQI Adjustment Based on BLER was implemented for cluster Ain Taya on 30 th Novenmber 2014, refer to
DT result we conclude increase in Average HSDPA Throughput by 16%, this increase is correlated with
clear change in below counters:

HSDPA Throughput distribution show increase of throughput sample between 4 and 5 Mbps from
29.1% to 48.9%

CQI distribution show increase of good CQI sample (25 to 30) by ~14%

The improvement in CQI conducted to improve in ACK and NACK Rate and mainly in Retransmission
Rate which decreased by around 12%

Clear improvement also in BLER as shown in PS BLER trend.

Page 14