Huawei 434789548 Massive MIMO TDD ERAN15 1 Draft A PDF
Huawei 434789548 Massive MIMO TDD ERAN15 1 Draft A PDF
Huawei 434789548 Massive MIMO TDD ERAN15 1 Draft A PDF
Issue Draft A
Date 2018-12-30
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective
holders.
Notice
The purchased products, services and features are stipulated by the contract made between Huawei and the
customer. All or part of the products, services and features described in this document may not be within the
purchase scope or the usage scope. Unless otherwise specified in the contract, all statements, information,
and recommendations in this document are provided "AS IS" without warranties, guarantees or
representations of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in the
preparation of this document to ensure accuracy of the contents, but all statements, information, and
recommendations in this document do not constitute a warranty of any kind, express or implied.
Website: http://www.huawei.com
Email: [email protected]
Contents
1 Change History.............................................................................................................................. 1
1.1 eRAN15.1 Draft A (2018-12-30)................................................................................................................................... 1
3 Overview....................................................................................................................................... 11
3.1 Introduction...................................................................................................................................................................11
3.2 Application Scenarios...................................................................................................................................................13
5.1.2.2 Impacts....................................................................................................................................................................38
5.1.3 Requirements............................................................................................................................................................. 39
5.1.3.1 Licenses.................................................................................................................................................................. 39
5.1.3.2 Software..................................................................................................................................................................39
5.1.3.3 Hardware................................................................................................................................................................ 39
5.1.3.4 Others......................................................................................................................................................................39
5.1.4 Operation and Maintenance.......................................................................................................................................39
5.2 MU-MIMO................................................................................................................................................................... 39
5.2.1 Principles................................................................................................................................................................... 40
5.2.2 Network Analysis...................................................................................................................................................... 43
5.2.2.1 Benefits................................................................................................................................................................... 43
5.2.2.2 Impacts....................................................................................................................................................................43
5.2.3 Requirements............................................................................................................................................................. 43
5.2.3.1 Licenses.................................................................................................................................................................. 43
5.2.3.2 Software..................................................................................................................................................................45
5.2.3.3 Hardware................................................................................................................................................................ 46
5.2.4 Operation and Maintenance.......................................................................................................................................46
5.2.4.1 Data Configuration................................................................................................................................................. 47
5.2.4.1.1 Data Preparation.................................................................................................................................................. 47
5.2.4.1.2 Using MML Commands...................................................................................................................................... 49
5.2.4.1.3 Using the CME.................................................................................................................................................... 49
5.2.4.2 Activation Verification........................................................................................................................................... 49
5.2.4.3 Network Monitoring............................................................................................................................................... 50
5.2.4.4 Possible Issues........................................................................................................................................................ 52
6.2 MU Beamforming.........................................................................................................................................................61
6.2.1 Principles................................................................................................................................................................... 62
6.2.2 Network Analysis...................................................................................................................................................... 66
6.2.2.1 Benefits................................................................................................................................................................... 66
6.2.2.2 Impacts....................................................................................................................................................................66
6.2.3 Requirements............................................................................................................................................................. 68
6.2.3.1 Licenses.................................................................................................................................................................. 68
6.2.3.2 Software..................................................................................................................................................................70
6.2.3.3 Hardware................................................................................................................................................................ 71
6.2.4 Operation and Maintenance.......................................................................................................................................71
6.2.4.1 Data Configuration................................................................................................................................................. 71
6.2.4.1.1 Data Preparation.................................................................................................................................................. 71
6.2.4.1.2 Using MML Commands...................................................................................................................................... 76
6.2.4.1.3 Using the CME.................................................................................................................................................... 76
6.2.4.2 Activation Verification........................................................................................................................................... 77
6.2.4.3 Network Monitoring............................................................................................................................................... 77
6.3 TM9.............................................................................................................................................................................. 81
6.3.1 Principles................................................................................................................................................................... 81
6.3.2 Operation and Maintenance.......................................................................................................................................81
6.3.2.1 Data Configuration................................................................................................................................................. 81
6.3.2.1.1 Data Preparation.................................................................................................................................................. 81
6.3.2.1.2 Using MML Commands...................................................................................................................................... 83
6.3.2.1.3 Using the CME.................................................................................................................................................... 84
6.3.2.2 Activation Verification........................................................................................................................................... 84
6.3.2.3 Network Monitoring............................................................................................................................................... 84
6.4 TM9 Hybrid Precoding.................................................................................................................................................84
6.4.1 Principles................................................................................................................................................................... 84
6.4.2 Network Analysis...................................................................................................................................................... 85
6.4.2.1 Benefits................................................................................................................................................................... 85
6.4.2.2 Impacts....................................................................................................................................................................85
6.4.3 Requirements............................................................................................................................................................. 86
6.4.3.1 Licenses.................................................................................................................................................................. 86
6.4.3.2 Software..................................................................................................................................................................87
6.4.3.3 Hardware................................................................................................................................................................ 88
6.4.3.4 Others......................................................................................................................................................................88
6.4.4 Operation and Maintenance.......................................................................................................................................88
6.4.4.1 Data Configuration................................................................................................................................................. 88
6.4.4.1.1 Data Preparation.................................................................................................................................................. 88
6.4.4.1.2 Using MML Commands...................................................................................................................................... 89
6.4.4.1.3 Using the CME.................................................................................................................................................... 89
6.4.4.2 Activation Verification........................................................................................................................................... 90
6.4.4.3 Network Monitoring............................................................................................................................................... 90
8 Parameters................................................................................................................................... 199
9 Counters...................................................................................................................................... 200
10 Glossary..................................................................................................................................... 201
11 Reference Documents............................................................................................................. 202
1 Change History
This chapter describes changes not included in the "Parameters", "Counters", "Glossary", and
"Reference Documents" chapters. These changes include:
l Technical changes
Changes in functions and their corresponding parameters
l Editorial changes
Improvements or revisions to the documentation
Technical Changes
Change Description Parameter Change Base Station Model
Added advanced massive Added the 3900 and 5900 series base
MIMO IRC. For details, see PUSCH_RECEIVE_ENH stations
7.7 IRC. _SWITCH option to the
CellMMAlgo.MMAlgoOpt
Switch parameter.
Added optimized switching Added the 3900 and 5900 series base
to beamforming in heavy- CellBf.HeavyLoadSwitch- stations
load scenarios. For details, ToBfOpt parameter.
see 6.7.1.3 Optimized
TMA.
Added optimized TMA for Added the 3900 and 5900 series base
4R and 8R UEs. For details, CellBfMimoParaCfg.Ue4R stations
see 6.7.1.3 Optimized Or8RBfMimoAdaptOptSw
TMA. parameter.
Added intelligent selection Added the 3900 and 5900 series base
of massive MIMO CellMMAlgo.MmParamInt stations
parameters (trial). For elSelectSw parameter and its
details, see 7.13 Intelligent options
Parameter Selection. BIG_EVENT_THLD_LE
ARN_SW and
BIG_EVENT_PDCCH_C
APC_ADAPT_SW.
Added the SRS IC function. Added the 3900 and 5900 series base
For details, see 7.4 SRS NI CellUlIcAlgo.SrsIcSwitch stations
Joint Suppression. parameter.
Added the description that Added the 3900 and 5900 series base
the maximum number of NON_AS_UE_MAX_BF_ stations
beamforming layers for both RANK_CTRL_SW option
fake antenna selection UEs to the
and non-antenna selection CellBfMimoParaCfg.BfMi
UEs is fixed to 1 to prevent moAlgoOptSwitch
their entry into dual-stream parameter.
beamforming (TM8) when
MU beamforming pairing of
antenna selection UEs in
dual-stream beamforming is
enabled in massive MIMO
cells and the
NON_AS_UE_MAX_BF_
RANK_CTRL_SW option
is selected under the
CellBfMimoParaCfg.BfMi
moAlgoOptSwitch
parameter. For details, see:
l 7.9.1 Principles
l 7.9.4.1.1 Data
Preparation
l 7.9.4.1.2 Using MML
Commands
Added tilt fine-tuning Added the 3900 and 5900 series base
corresponding to the BfAnt.TiltFractionalPart stations
beamforming weight. For parameter.
details, see 4.4.1.1 Data
Preparation.
Added remote AAU Added the 3900 and 5900 series base
azimuth adjustment. For BfAnt.BfantBearing stations
details, see 4.4.1.1 Data parameter.
Preparation.
Added beamforming weight Added the 3900 and 5900 series base
granularity optimization. BF_WEIGHT_GRAN_OP stations
For details, see 6.1.4.1.1 T_SWITCH option to the
Data Preparation and CellBfMimoParaCfg.BfMi
6.1.4.1.2 Using MML moAlgoOptSwitch
Commands. parameter.
Added access control Added the 3900 and 5900 series base
optimization based on MM_ACCESS_CONTRO stations
uplink and downlink UE- L_OPT_SWITCH option
perceived rates. For details, to the
see 7.10 Access Control CellMMAlgo.MMAlgoOpt
Optimization Based on Switch parameter.
Uplink and Downlink UE-
Perceived Rates.
Added PDCCH resource Added the 3900 and 5900 series base
reservation for big-packet HEAVY_LOAD_BIG_PK stations
services in heavy-load T_SCH_OPT_SW option
scenarios. For details, see to the
section 7.11 PDCCH CellPdcchAlgo.PdcchOpti
Resource Reservation for mizationSwitch parameter.
Big-Packet Services in
Heavy-Load Scenarios.
Added optimized PDCCH Added the 3900 and 5900 series base
aggregation level selection HEAVY_LOAD_BIG_PK stations
near the cell center. For T_SCH_OPT_SW option
details, see 7.12 Optimized to the
PDCCH Aggregation CellPdcchAlgo.PdcchOpti
Level Selection Near the mizationSwitch parameter.
Cell Center.
Added the TM8 pairing Added the 3900 and 5900 series base
optimization function. For CellBf.Tm8PairingOptSwit stations
details, see 7.9 MU ch parameter.
Beamforming Pairing of
Antenna Selection UEs in
Dual-Stream
Beamforming.
Added support of UEs in Added the 3900 and 5900 series base
dual-stream beamforming CellBf.DualStreamBfAlgoS stations
(TM8) for multiplexing w parameter and its option
feedback mode. For details, ACK_FEEDBACK_IN_M
see 7.9.1 Principles. UX_MODE_SW.
Added massive MIMO Added the 3900 and 5900 series base
uplink overload MmUlHeavyLoadOptS- stations
optimization. For details, witch option to the
see 7.3 Uplink Overload CellUlschAlgo.VmimoOptA
Optimization. lgoSwitch parameter.
Added MCS index selection Added the 3900 and 5900 series base
optimization for downlink MM_DL_SCH_AND_MC stations
paired UEs with small- S_OPT_SWITCH option to
amount data transmission the
and cell-level outer-loop CellMMAlgo.MMAlgoOpt
adjustment optimization in Switch parameter.
beamforming. For details,
see 7.8 Scheduling
Optimization.
Editorial Changes
Revised descriptions in this document.
This document only provides guidance for feature activation. Feature deployment and feature
gains depend on the specifics of the network scenario where the feature is deployed. To achieve
the desired gains, contact Huawei professional service engineers.
Software Interfaces
Any parameters, alarms, counters, or managed objects (MOs) described in this document
apply only to the corresponding software release. For future software releases, refer to the
corresponding updated product documentation.
Trial Features
Trial features are features that are not yet ready for full commercial release for certain
reasons. For example, the industry chain (terminals/CN) may not be sufficiently compatible.
However, these features can still be used for testing purposes or commercial network trials.
Anyone who desires to use the trial features shall contact Huawei and enter into a
memorandum of understanding (MoU) with Huawei prior to an official application of such
trial features. Trial features are not for sale in the current version but customers may try them
for free.
Customers acknowledge and undertake that trial features may have a certain degree of risk
due to absence of commercial testing. Before using them, customers shall fully understand not
only the expected benefits of such trial features but also the possible impact they may exert on
the network. In addition, customers acknowledge and undertake that since trial features are
free, Huawei is not liable for any trial feature malfunctions or any losses incurred by using the
trial features. Huawei does not promise that problems with trial features will be resolved in
the current version. Huawei reserves the rights to convert trial features into commercial
features in later R/C versions. If trial features are converted into commercial features in a later
version, customers shall pay a licensing fee to obtain the relevant licenses prior to using the
said commercial features. If a customer fails to purchase such a license, the trial feature(s)
will be invalidated automatically when the product is upgraded.
Unless otherwise stated, descriptions in this document apply to all RATs. If a description does
not apply to all RATs, the specific RAT that it does apply to will be stated.
For example, in the statement "TDD cells are compatible with enhanced MU-MIMO", "TDD
cells" indicates that this function cannot be used in non-TDD cells.
NOTE
The features described in 4 Basic Massive MIMO Functions and 5.1 Receive Diversity are
prerequisites for setting up massive MIMO cells. The operation and maintenance (O&M) of the features
are described in 4.4 Operation and Maintenance.
3 Overview
3.1 Introduction
Massive MIMO is widely regarded as a key update of multiple-antenna technology in the
4.5G era. It uses a large number of antenna arrays to perform 3D beamforming and multi-
layer multi-user multiplexing, significantly improving system capacity and 3D coverage.
The following figure shows hardware evolution from traditional MIMO sites to massive
MIMO sites.
Uplink multi-user MIMO Up to four layers can share Up to eight layers can share
(MU-MIMO for short) the same time-frequency the same time-frequency
resource. resource.
NOTE
For details about traditional MIMO, see MIMO and Beamforming (TDD).
l Hotspot coverage
In typical hotspot areas, user density is high and many online UEs create extremely
strong demand for capacity in both the uplink and the downlink. Typical hotspots include
dense urban areas, central business districts (CBDs), public squares, and stadiums.
Massive MIMO can improve system capacity through spatial multiplexing. Up to 8
layers in the uplink for MU-MIMO pairing and up to 16 layers in the downlink for MU
beamforming pairing significantly improve the cell throughput and meet capacity
demands in hotspot areas.
l Tall-building coverage
In tall buildings, users are distributed across many floors, and a normal site cannot
extend coverage vertically very far. The vertical beamwidth (approximately 6.5°) makes
it impossible for a traditional site to cover multiple floors.
Massive MIMO enables the eNodeB to adjust the vertical broadcast beamwidth to up to
35°, enhancing vertical coverage of broadcast beams. In this way, more floors can be
covered by a single site. Massive MIMO strengthens power, array, and diversity gains,
and enables energy to be more concentrated by implementing 3D beamforming on traffic
beams. The gains offset the propagation and penetration losses and improve the link
quality for indoor UEs in tall buildings.
4.1 Principles
Basic massive MIMO functions include broadcast beamforming and TDLEOFD-121601
Massive MIMO Introduction. TDLEOFD-121601 Massive MIMO Introduction is a
prerequisite for activating a massive MIMO cell.
Broadcast Beamforming
Broadcast beamforming enables the eNodeB to apply weighting on broadcast beams to adjust
their coverage scope.
The weighting designed for typical coverage scenarios has been written into the beam weight
file, which is included in the eNodeB software package. After the beam weight file is
activated, the eNodeB configures broadcast beamwidth on both the horizontal and vertical
planes based on the setting of the BfAnt.CoverageScenario parameter, satisfying broadcast
coverage demands in various application scenarios. A vertical beamwidth of up to 35°
significantly improves the vertical coverage scope in tall buildings.
l If fewer than 32 antennas remain operational, the massive MIMO cell is deactivated.
l If more than eight antennas are faulty, neither PDCCH SDMA in massive MIMO nor
multi-user split SDMA takes effect.
This feature is free from parameter control. It is enabled by default for massive MIMO cells.
– The proportion of downlink large IP packets is greater than the threshold specified
by the ServiceIdentifyPara.MassFlowBigPacketRateThld parameter.
– The cumulative number of downlink IP packets is greater than the threshold
specified by the ServiceIdentifyPara.MassFlowPacketNumberThld parameter.
l Normal
All other services are considered normal services.
4.2.1 Benefits
This feature increases power and array gains. Massive MIMO enables the eNodeB to adjust
broadcast beams and downlink traffic beams both horizontally and vertically. It achieves
better uplink and downlink coverage performance than 8T8R multiple-antenna technologies.
The gains are more significant on the vertical plane.
NOTE
l For details about the power and array gains, see Beamforming (TDD).
l For 3.5 GHz massive MIMO cells, it is recommended that inter-frequency split be enabled so that a
single site can provide six sectors using two frequencies, improving cell coverage. For details about
inter-frequency split, see Soft Split Resource Duplex (TDD).
4.2.2 Impacts
Network Impacts
l System capacity
This feature improves the average cell capacity, peak cell throughput, average single-UE
throughput, and CEU throughput in both the uplink and downlink.
l Network performance
– This feature improves network coverage in both the uplink and downlink.
– Massive MIMO cells have a larger target SRS power for SRS power control than
normal cells. This increases the downlink spectral efficiency and cell throughput as
well as SRS interference.
Function Impacts
Function Function Switch Reference Description
Name
4.3 Requirements
4.3.1 Licenses
Feature ID Feature Name Model Sales Unit
NOTE
For details about the capacity license usage of a massive MIMO cell, see License Management.
4.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
4.3.3 Hardware
Boards
l BBU: BBU3910 or BBU5900
l BBPs
– UBBPem
– UBBPf1
– UBBPf3
– UBBPfw1 (only available in the BBU5900)
l Main control board: UMPTe
NOTE
For details about the maximum number of BBPs supported by each BBU and the maximum number of
massive MIMO cells supported by each BBP, see the hardware description of the corresponding BBU.
RF Modules
l AAU5271
l AAU5281
l AAU5613 (compatible only with the UBBPfw1 and UBBPf3)
Cells
The following table describes the frequency, cell bandwidth, and subframe configurations
supported by each RF module.
NOTE
l The UBBPf1, UBBPf3, or UBBPfw1 does not support uplink-downlink subframe configuration 1.
l Special subframe configuration 5 can be used only with uplink-downlink subframe configuration 2,
not with uplink-downlink subframe configuration 1.
4.3.4 Others
Due to new functions or features, new beam weight files may be introduced in the new
software version after an upgrade. When upgrading or rolling back the software version,
perform operations on the beam weight file according to the following table.
The new beam weight file is The beam weight file does The original beam weight
not used in the target not need to be activated. file does not need to be
software version. activated.
The new beam weight file is Run the ACT BFANTDB Run the ACT BFANTDB
used in the target software command to reactivate the command to activate the
version. beam weight file. original beam weight file.
NOTE
If a new beam weight file is installed after a software upgrade and then a software rollback is performed:
l Before the original beam weight file is activated, the new beam weight file is still stored in the
eNodeB.
l After the old beam weight file is activated, the new beam weight file is deleted.
Port 1 Port No. RRUCHAIN.CONNPN Set this parameter to the CPRI port
1 number 1 of the trunk chain.
Port 2 Port No. RRUCHAIN.CONNPN Set this parameter to the CPRI port
2 number 2 of the trunk chain.
Before configuring a beamforming antenna, run the ACT BFANTDB command with the
OPMODE parameter set to SWFILE to activate the beam weight file.
The following operations are not required when a massive MIMO cell is to be set up at an
existing site:
l Activating the minimum configuration
l Configuring operator information
l Configuring clock information
l Configuring the local time zone and DST information
//Configuring the WeightNO parameter to the same value as the DEVICENO parameter
MOD EUCELLSECTOREQM: LocalCellId=0, SectorEqmId=0, WeightNO=0;
//(Optional) Enabling service-based dynamic SRS allocation when there are 200 to
500 UEs in the cell
MOD SRSCFG: LocalCellId=0, SrsCfgPolicySwitch=SrvBasedSRSAdjAlgo-1;
MOD SERVICEIDENTIFYPARA: HeartbeatPacketLengthThld=180,
HeartbeatPacketNumberThld=4, MassFlowBigPacketRateThld=80, MassFlowDuration=2,
MassFlowPacketNumberThld=200;
If a function requires the execution of the ACT BFANTDB command for activation after cells are
activated, this function can be activated by running this command, but not by using the CME.
l The query results include historical calibration results. Identify the calibration record based on the
activation time of the corresponding cell and the calibration time to determine the calibration result.
l Beamforming and PDCCH SDMA can be activated only when channel calibration is successful.
----End
NOTE
To define a channel calibration failure alarm, perform the following operations on the U2020:
Step 1 Choose Performance > Threshold Management > Threshold Settings. The Threshold
Settings tab page is displayed.
Step 2 Choose eNodeB > Measurement of CellSectorEQUIP Performance from the navigation
tree, and then click Add at the bottom right part of the tab page. The Add Threshold dialog
box is displayed.
Step 3 Select one or more NEs in the Object tab of the Add Threshold dialog box.
Step 4 On the Basic tab page, set Name, Period, and Activation time.
Step 5 On the Advanced tab page, perform the following operations and then click OK.
An alarm is reported if the duration of the channel calibration failure is greater than or equal
to the threshold plus the offset. It is cleared if the duration falls below the threshold minus the
offset.
l Set Direction to Ascending.
l Set Function subset to Network/Measurements Related to Algorithm(LTE)/Cell
Sector Algorithm Measurement.
l Set Counter name to L.CellSectorEqpt.UNA.Dur.Cali.
l Set Threshold based on the calculation result of Period set on the Basic tab page
multiplied by 80% minus the offset. In the example, set Threshold to 9, namely 15 x
80% – 3.
l Set Offset based on the calculation result of Period set on the Basic tab page multiplied
by 20%. In the example, set Offset to 3, namely 15 x 20%.
----End
Antenna Fault
This feature enables the eNodeB to periodically check for antenna faults. The eNodeB reports
ALM-29243 Cell Capability Degraded upon detecting an antenna fault.
In this situation, the value of the alarm parameter Specific Problem is Antenna channel
exceptions. The number of functional antennas can be identified based on the values of the
TX Channel Numbers In The Cell and RX Channel Numbers In The Cell parameters.
l The cell can still work properly as long as 32 or more antennas are operational. However,
faulty antennas may cause network performance, such as the cell throughput, to
deteriorate. The impact is directly proportionate to the number of faulty antennas.
l If fewer than 32 antennas remain operational, the massive MIMO cell is deactivated.
l If more than eight antennas are faulty, neither PDCCH SDMA in massive MIMO nor
multi-user split SDMA takes effect.
5.1.1 Principles
This section describes TDLEOFD-12160301 UL 64-Antenna Receive Diversity.
Receive diversity enables an eNodeB to receive signals through multiple antennas and then
combine the received signals using a specified signal combining algorithm. In this way, the
SINR is maximized and diversity and array gains are achieved. The gains are proportionate to
the number of receive antennas. For details about uplink receive diversity, see MIMO.
Receive diversity is free from parameter control. It is deployed when a massive MIMO cell is
configured and the license has been purchased and activated.
Maximum ratio combining (MRC) and interference rejection combining (IRC) are typically
used for signal combining during receive diversity. For details, see MRC and IRC Receivers.
5.1.2.1 Benefits
This feature increases array gains. Massive MIMO enables the eNodeB to adjust broadcast
beams and downlink traffic beams both horizontally and vertically. It achieves better uplink
and downlink coverage performance than 8T8R multiple-antenna technologies. The gains are
more significant on the vertical plane.
NOTE
5.1.2.2 Impacts
Network Impacts
l System capacity
This feature offers greater diversity gains and array gains than traditional uplink 2, 4, or
8-antenna receive diversity, improving uplink cell coverage, uplink average cell
throughput, and CEU throughput.
l Network performance
This feature provides a larger vertical coverage scope and stronger vertical coverage than
traditional 2, 4, or 8-antenna receive diversity. Accordingly, interference may increase in
contrast with that of a normal cell, leading to increasing measured interference counter
values.
Function Impacts
Function Function Switch Reference Description
Name
Control channel PrachIrcSwitch and MRC and IRC Physical Random Access
receivers PucchIrcSwitch options Receivers CHannel (PRACH) IRC,
under the if being enabled, may
CellAlgoSwitch.IrcSwitc affect network access
h parameter performance in massive
MIMO scenarios. It is
recommended that the
PRACH IRC function be
disabled. You are advised
to enable the Physical
Uplink Control CHannel
(PUCCH) IRC function
in interference scenarios.
5.1.3 Requirements
5.1.3.1 Licenses
Feature ID Feature Name Model Sales Unit
5.1.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
5.1.3.3 Hardware
See 4.3.3 Hardware.
5.1.3.4 Others
For details, see 4.3.4 Others.
5.2 MU-MIMO
The following table lists MU-MIMO features and the maximum number of pairing layers that
can be paired for each feature.
NOTE
5.2.1 Principles
Overview
MU-MIMO is enabled in massive MIMO scenarios when the UlVmimoSwitch option is
selected under the CellAlgoSwitch.UlSchSwitch parameter. When this function is enabled,
the eNodeB attempts to pair UEs in each transmission time interval (TTI). Paired UEs use the
same time-frequency resource for uplink transmission. This technology improves the average
uplink cell throughput. If pairing fails for a given UE, the UE uses independent uplink time-
frequency resources for uplink transmission.
UE Pairing
The maximum number of MU-MIMO pairing layers in a massive MIMO cell is specified by
the CellUlschAlgo.MaxLayerMMVMIMO parameter.
The principles for selecting UEs to be paired are as follows:
l High-speed UEs are not selected for MU-MIMO pairing due to their constantly changing
channel environment and the poor stability of their radio quality. Pairing such UEs may
cause system performance to deteriorate.
l If the paired UEs have high SINRs and their channels are approximately orthogonal,
interference between the UEs can be effectively mitigated. Such UEs are suitable for
MU-MIMO. If UE channels are strongly correlated or the SINRs are low, interference
between the UEs cannot be effectively mitigated and MU-MIMO may decrease system
performance. Pairing of such UEs is not recommended.
NOTE
The correlation between UE channels is calculated based on uplink feedback. This similarity
between uplink channels is highly correlated to the distance between UEs. The correlation is high
if two UEs are close to each other, and is low if they are not.
l The following parameters are used to control the types of UEs to be paired or UE pairing
policy:
Parameter ID Description
MU-MIMO Enhancement
l Resource randomization
After MU-MIMO is enabled, interference on the PUSCH may increase. You are advised
to enable resource randomization to ensure that the UE-perceived data rate in
neighboring cells does not decrease after MU-MIMO takes effect. This function is
controlled by the VMIMOResRandomSwitch option under the
CellUlschAlgo.VmimoOptAlgoSwitch parameter.
l MCS index selection optimization
The uplink MU-MIMO pairing algorithm has been optimized so that optimal MCS
indexes are selected after pairing. This function is recommended in big-event heavy-load
massive MIMO scenarios. This function takes effect only for cells set up on the UBBPf1,
UBBPf3, or UBBPfw1. It can be used with uplink 2x64 MU-MIMO, uplink 4-layer MU-
MIMO, or uplink 8-layer MU-MIMO. This function is controlled by the
VmimoOptResSwitch option under the CellUlschAlgo.VmimoOptAlgoSwitch
parameter. Selecting this option reduces the initial block error rate (IBLER) after MU-
MIMO pairing, and in big-event heavy-load scenarios, improves the uplink spectral
efficiency.
l Pairing optimization
The mechanism for determining the number of uplink MU-MIMO pairing layers has
been optimized so that five to eight layers can be paired when the number of uplink RBs
is small. This function is recommended in massive MIMO scenarios. This function takes
effect only on cells set up on the UBBPf1 and requires uplink 8-layer MU-MIMO to be
enabled. This function is controlled by the CellUlschAlgo.EnhancedVmimoSwitch
5.2.2.1 Benefits
This feature increases multiplexing gains. Massive MIMO supports up to eight layers for
uplink MU-MIMO, with theoretical uplink peak throughput approximately eight times that
provided by uplink receive diversity in 8T8R networking scenarios.
NOTE
5.2.2.2 Impacts
Network Impacts
l A maximum of two, four, or eight UEs can use the same time and frequency resources to
increase the uplink spectral efficiency and uplink cell throughput. After this feature is
enabled, the probability of uplink control channel element (CCE) allocation failures may
increase because the number of uplink scheduled UEs increases.
l If uplink 4-layer MU-MIMO or uplink 8-layer MU-MIMO is enabled:
– The number of UEs paired within each TTI may change, leading to fluctuating
interference between paired UEs. As a result, the BLER of the uplink PUSCH may
increase.
– After a 4T/8T cell is reconstructed into a massive MIMO cell, the uplink and
downlink IBLERs may increase.
l If pairing optimization for uplink 8-layer MU-MIMO is enabled:
– If interference from neighboring cells changes significantly with time, this function
may decrease the average UE throughput, spectral efficiency, and average cell
throughput in the uplink.
– When the uplink physical resource block (PRB) load is not heavy, the single-UE-
perceived rate may decrease. When the uplink PRB load is heavy, the pairing
performance of uplink fifth to eighth layers improves significantly and the uplink
and downlink BLERs may increase.
Function Impacts
None
5.2.3 Requirements
5.2.3.1 Licenses
l The operator has purchased and activated the following license for TDLEOFD-12160302
UL 2x64 MU-MIMO. Only one unit of the license control item Massive MIMO UL 2-
Layers Extended Processing Unit License(TDD) is required.
l The operator has purchased and activated the license for TDLEOFD-121609 UL 4-Layer
MU-MIMO. Two units of the license control item Massive MIMO UL 2-Layers
Extended Processing Unit License(TDD) are required.
Feature ID Feature Model License Sales Unit
Name Control Item
l The operator has purchased and activated the license for TDLEOFD-121610 UL 8-Layer
MU-MIMO. Four units of the license control item Massive MIMO UL 2-Layers
Extended Processing Unit License(TDD) are required.
Feature ID Feature Model License Sales Unit
Name Control Item
5.2.3.2 Software
Prerequisite Functions
l TDLEOFD-12160302 UL 2x64 MU-MIMO requires the following functions.
Function Function Switch Reference Description
Name
5.2.3.3 Hardware
See 4.3.3 Hardware. The UBBPf1 must be deployed for pairing optimization, and the
UBBPf1, UBBPf3, or UBBPfw1 must be deployed for MCS index selection optimization.
MU-MIMO Pairing
Step 1 On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace Management.
Step 2 From the navigation tree of the Signaling Trace Management tab page, choose LTE > Cell
Performance Monitoring, and then double-click Multi User-MIMO Monitoring.
Step 3 In the displayed dialog box, select an eNodeB and set Local Cell ID for the cell to be traced.
Then, click Finish to start a trace task.
Step 4 Use multiple UEs to access the network, adjust the RSRP of the UEs to be greater than –75
dBm, and then perform uplink FTP services on the UEs.
Step 5 Double-click the trace task to verify the value of UL MIMO UE Pair Num.
l If the value of UL MIMO UE Pair Num is not N/A, uplink MU-MIMO has taken
effect. UL MIMO UE Pair Num indicates the total number of UEs paired for uplink
MIMO.
l If the value of UL MIMO UE Pair Num 2 Layers is not N/A, uplink 2x64 MU-MIMO
has taken effect.
l If the value of UL MIMO UE Pair Num 3 Layers or UL MIMO UE Pair Num 4
Layers is not N/A, uplink 4-layer MU-MIMO has taken effect.
l If any of the values of UL MIMO UE Pair Num 5 Layers, UL MIMO UE Pair Num
6 Layers, UL MIMO UE Pair Num 7 Layers, or UL MIMO UE Pair Num 8 Layers
is not N/A, uplink 8-layer MU-MIMO has taken effect.
Step 6 Use multiple UEs to access the network, adjust the RSRP of the UEs to be greater than –75
dBm, and then perform uplink VoLTE services on the UEs.
Step 7 Double-click the trace task to verify the value of VoLTE UE VMIMO Pair Num.
If the value of VoLTE UE Mimo Pair Num is not N/A, VoLTE UEs have been paired for
MU-MIMO.
----End
MU-MIMO Enhancement
Step 1 Run the LST CELLULSCHALGO command. If VMIMOResRandomSwitch:On is
displayed under the Vmimo Optimization Algorithm Switch parameter, resource
randomization has been enabled.
Step 2 Run the LST CELLULSCHALGO command. If VmimoOptResSwitch:On is displayed
under the Vmimo Optimization Algorithm Switch parameter, MCS index selection
optimization has been enabled.
Step 3 Run the LST CELLULSCHALGO command to query the value of the Enhanced VMIMO
Switch parameter. If the value is ON, pairing optimization has been enabled.
----End
Signaling Tracing
Step 1 On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace Management.
Step 2 In the navigation tree of the Signaling Trace Management tab page, choose LTE > Cell
Performance Monitoring > Throughput Monitoring. Check the throughput of the eNodeB
and the cells under it.
Step 3 In the navigation tree of the Signaling Trace Management tab page, choose Trace Type >
LTE > Cell Performance Monitoring > Multi User-MIMO Monitoring. Check the values
of the following monitoring items.
l UL MIMO UE Pair Num
l UL MIMO UE Pair Num 2 Layers through UL MIMO UE Pair Num 8 Layers
l UL MIMO 2Layers PairRB through UL MIMO 8Layers PairRB
UL MIMO UE Pair Num 3 Layers through UL MIMO UE Pair Num 8 Layers and UL MIMO
3Layers PairRB through UL MIMO 8Layers PairRB are not suitable for uplink 2x64 MU-
MIMO. UL MIMO UE Pair Num 5 Layers through UL MIMO UE Pair Num 8 Layers and
UL MIMO 5Layers PairRB through UL MIMO 8Layers PairRB are not suitable for uplink
4x64 MU-MIMO.
If the monitoring item corresponding to an MU-MIMO feature has a non-zero value, the
feature has been activated. If the value of Mimo UE Pair Num or the number of paired RBs
is always 0, see 5.2.4.4 Possible Issues.
NOTE
MU-MIMO pairing performance is subjected to channel quality, user distributions, service status, and
other factors. Therefore, it is acceptable if the number of paired layers does not reach the maximum
within a given measurement period.
Step 4 In the navigation tree of the Signaling Trace Management tab page, choose LTE > Cell
Performance Monitoring > Multi User-MIMO Monitoring. Check the value of VoLTE
UE Mimo Pair Num.
l If the value of VoLTE UE Mimo Pair Num is not 0, VoLTE UEs have been paired for
MU-MIMO.
l If the value of VoLTE UE Mimo Pair Num is always 0, see 5.2.4.4 Possible Issues.
----End
Counter-based Monitoring
Step 1 Monitor counters in the following function subsets:
l Total Number of TBs Transmitted in the Uplink in Massive MIMO
l Total Number of TBs That Fail to Be Transmitted in the Uplink in Massive MIMO
l Number of Bits Successfully Transmitted in the Uplink in Massive MIMO
l Total Number of RBs Transmitted in the Uplink in Massive MIMO Scenarios
l Number of RBs That Are Successfully Paired for VMIMO UEs in a Cell
l Number of RBs That Can Be Paired for VMIMO UEs in a Cell
l Number of RBs That Are Successfully Paired for VMIMO layer2 UEs in a Cell
Step 2 Monitor the counters related to the number of PRBs actually scheduled for UEs in a cell. The
number of PRBs is measured by UE, and therefore the total number of PRBs scheduled in a
20 MHz cell may exceed 100, except for the counters L.ChMeas.PRB.UL.Used.Avg and
L.ChMeas.PRB.UL.PUSCH.Avail.
Counter ID Counter Name Description
----End
Step 1 Run the DSP LICINFO command to check whether the feature license is available in the
command output.
l If no, apply for and load a new license.
l If yes, go to the next step.
Step 2 On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace Management.
Step 3 In the navigation tree of the Signaling Trace Management tab page, choose Trace Type >
LTE > Cell Performance Monitoring > Users Statistic Monitoring. Create and start a
tracing task, and check the value of General Users Number.
l If there are five or fewer UEs, the possibility of UE pairing is low and the MU-MIMO
feature does not take effect. It is recommended that there be more than five UEs.
l If there are enough UEs, go to the next step.
Step 4 In the navigation tree of the Signaling Trace Management tab page, choose Trace Type >
LTE > User Performance Monitoring > Quality of Channel Monitoring. Create and start a
tracing task, and check SINRs.
l If the SINRs of the monitored UEs are less than 1 dB, there is a high probability that the
UEs cannot be paired and MU-MIMO does not take effect.
l If the SINRs of the monitored UEs are greater than or equal to 1 dB but MU-MIMO does
not take effect, contact Huawei technical support.
----End
NOTE
Unless otherwise specified, UEs incapable of transmit antenna selection include fake antenna selection
UEs. For details about fake antenna selection UEs, see 7.9.1 Principles.
6.1 3D Beamforming
6.1.1 Principles
The overall procedure of beamforming in massive MIMO scenarios is the same as that of
traditional beamforming. For details, see Beamforming (TDD). Figure 6-1 shows the overall
beamforming process.
NOTE
l In massive MIMO scenarios, the eNodeB automatically starts the initial channel calibration 2
minutes after a cell is set up. The subsequent channel calibrations are performed on each carrier at a
fixed period of 10 seconds.
l The channel calibration result can be queried by running the DSP CELLCALIBRATION
command. Beamforming and PDCCH SDMA can be activated only when channel calibration is
successful.
1. Channel calibration: The eNodeB adjusts the changes in phase, amplitude, and delay
(Delta 1 through Delta 64) of calibration signals to be consistent with each other after
they pass through different AAU channels. This ensures the reciprocity between uplink
and downlink channels.
2. Weight calculation: The eNodeB estimates the PDSCH characteristics based on UE-
reported SRSs and calculates the beamforming weight.
3. Weighting and beamforming: The eNodeB weights the transmitted signals to form a
narrow beam targeted towards the UE. In this way, the eNodeB differentiates UEs
horizontally and vertically.
6.1.2.1 Benefits
This feature increases beamforming gains. Massive MIMO enables the eNodeB to adjust
broadcast beams and downlink traffic beams both horizontally and vertically. It achieves
better uplink and downlink coverage performance than 8T8R multiple-antenna technologies.
The gains are more significant on the vertical plane.
NOTE
6.1.2.2 Impacts
Network Impacts
l Impact on system capacity: This feature improves the downlink cell throughput and
spectral efficiency.
l Impact on network performance: This feature improves the downlink coverage
capability.
Function Impacts
Function Function Switch Reference Description
Name
6.1.3 Requirements
6.1.3.1 Licenses
Feature ID Feature Name Model Sales Unit
6.1.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
6.1.3.3 Hardware
See 4.3.3 Hardware.
NOTE
The cell will reset after SRS configurations are modified by running the MOD SRSCFG command.
NOTE
The 3GPP release version of the UE can be queried in the RRC_UE_CAP_INFO message.
Step 2 On the U2020, perform a Uu interface trace task and check the UE transmission mode based
on the value of the transmissionMode field in the explicitValue IE.
Step 3 Check whether the changes of transmission modes during the UE movement match what is
listed in 6.7.1.2 TMA Principles. If yes, beamforming and the related TMA have taken
effect.
----End
Step 1 Verify that the feature, hardware, and license meet the deployment requirements.
Step 3 Run the DSP CELLCALIBRATION command to check the channel calibration results. If
channel calibration has failed, check for uplink interference.
Step 4 If a UE cannot switch to TM7, check the RSRP of the UE. If the RSRP is low, move the UE
to another location and then retry the network access. If the UE cannot switch to TM9, run the
DSP CELLCSIRSPARACFG command to check the value of the CsiRs State parameter. If
the value is not Normal, rectify the fault based on the cause value.
----End
6.2 MU Beamforming
The following table lists 64-antenna MU beamforming features and the maximum number of
layers that can be paired.
NOTE
"Layer" in the feature names refers to the maximum number of multiplexing layers.
6.2.1 Principles
Overview
MU beamforming is enabled in massive MIMO scenarios when the MuBfSwitch option is
selected under the CellAlgoSwitch.MuBfAlgoSwitch parameter. When this function is
enabled, the eNodeB attempts to pair UEs in each TTI.
The eNodeB performs MU beamforming as follows:
l Determines the UEs to be paired and the number of downlink data streams to be
transmitted to each UE based on UE pairing conditions.
l Generates orthogonal weight values for downlink data streams based on the zero forcing
principle.
l Imposes weight on data symbols and DMRS symbols for each data stream using the
generated weight values.
l Combines the weighted symbols and transmits them over the antenna ports.
In this way, multiple data streams are transmitted in the downlink using the same time-
frequency resource, enhancing the downlink spectral efficiency.
UE Pairing
correlations. The pairing rule increases the number of paired layers, but may
reduce the system capacity.
3. The eNodeB uses the same time-frequency resources to transmit data to both of the
paired UEs.
Figure 6-8 uses 4-layer MU beamforming pairing as an example to illustrate the RBG
allocation for paired UEs. UEs with low spatial channel correlations, such as UEs 1, 2, 3,
and 4, can be paired for MU beamforming and are scheduled at different layers. MU
beamforming allows UEs to be paired to occupy different amounts of RB resources.
Therefore, unused layer 3 resources can be allocated to UEs 5 and 6 that are highly
correlated with UE 3 if UEs 5 and 6 are low correlated with other UEs that use the same
resource for data transmission.
NOTE
UE pairing uses spatial resources and the scheduling priority of UEs changes after MU
beamforming is enabled. Therefore, MU beamforming has an impact on the features or functions
related to scheduling priorities, such as QoS management.
MCS index selection is subject to CQIs and the CQI adjustment reliability depends on
the number of downlink data packets. If there are a small number of downlink data
packets, the CQI adjustment reliability is poor. When there are many downlink data
packets, the CQI adjustment reliability is high.
In massive MIMO scenarios, the eNodeB selects MCS indexes for dual-stream
beamforming based on the MCS indexes of single-stream beamforming. When there are
a small number of downlink packets in single-stream beamforming, the MCS indexes of
single-stream beamforming are unreliable, which makes it impossible for the eNodeB to
select optimal MCS indexes for MU beamforming.
MU beamforming-specific MCS index selection optimization is introduced to resolve
this issue. With this function, if the number of MU beamforming data packets is greater
than the threshold specified by the CellCqiAdjAlgo.MuBfPacketThld parameter, the
eNodeB determines MCS indexes for MU beamforming based on 4-bit CQI information
reported by UEs, measured BFGain value of the eNodeB, and HARQ feedback on MU
beamforming data packets. This function enhances the accuracy of MCS index selection
in MU beamforming scenarios.
Beamforming weights calculated based on SRS measurements may not match ever-
changing UE channel conditions, affecting the MCS index selection accuracy for pairing.
If many UEs move at a speed of 5 km/h or less, selecting the
MU_PAIRING_MCS_EXACT_SWITCH option under the
CellMMAlgo.MMAlgoOptSwitch parameter improves the MCS index selection
accuracy for pairing. This optimization has no gain in scenarios where UEs are
stationary. If many UEs move at a speed of greater than 5 km/h, this optimization has
limited gains or even no gains due to exceedingly violent channel condition changes. The
MCS index fluctuates with signal quality when there are many low-speed UEs,
improving the MCS index selection accuracy and the downlink spectral efficiency. The
UEs that are not suitable for pairing are not paired, which may reduce the number of
paired layers.
l Moving UE optimization
This function is controlled by the HIGH_SPEED_OPTI_SCHEME option under the
CellBf.MovingUeMuBfScheme parameter. When this option is selected, the eNodeB
adjusts the MU beamforming grouping mechanism for moving UEs. The adjustment
ensures that the MU beamforming weights of moving UEs are orthogonal. This function
is suitable when downlink 8- or 16-layer MU beamforming is enabled.
This function increases the downlink cell throughput and spectral efficiency in mobility
scenarios, but the average number of pairing layers may decrease.
l Big-event moving UE protection
This function is controlled by the BIG_EVENT_PROTECT_SCHEME option under
the CellBf.MovingUeMuBfScheme parameter. When this option is selected, the eNodeB
adjusts the UE grouping mechanism for MU beamforming pairing of moving UEs. The
adjustment ensures that the MU beamforming weights of moving UEs are orthogonal.
This function is suitable when downlink 8- or 16-layer MU beamforming is enabled.
This function improves MU beamforming pairing performance of moving UEs in big-
event scenarios. In heavy-load scenarios, however, this function may increase the
average PDCP service data unit (SDU) transmission delay.
NOTE
NOTE
l Before this function was introduced, the initial CQI outer-loop adjustment value is 0 dB and cannot
be changed in MU beamforming scenarios.
l It is recommended that the CellMMAlgo.MassiveMimoInitDeltaCqi parameter be set to 0. Confirm
the value of the parameter with Huawei engineers during network optimization.
6.2.2.1 Benefits
This feature increases multiplexing gains. Massive MIMO supports up to 16 layers (or 24
layers as a trial version only) for downlink MU beamforming, which delivers 7 times or 9.9
times as much as theoretical downlink peak throughput as downlink 2x2 MIMO in 8T8R
networking scenarios does.
NOTE
6.2.2.2 Impacts
Network Impacts
l This function enables an eNodeB to simultaneously transmit up to 8, 16, or even 24 (for
trial) data streams, increasing the downlink cell throughput and spectral efficiency.
l The number of UEs paired within each TTI may change, leading to fluctuating
interference between paired UEs. As a result, the block error rate (BLER) of the
downlink PDSCH may increase.
l The downlink time-domain adaptive extended scheduling function improves the user
perceived rate of downlink heavy-traffic services.
l The MU beamforming-specific MCS index selection optimization function reduces the
IBLER.
l After a 4T/8T cell is reconstructed into a massive MIMO cell, the uplink and downlink
IBLERs may increase.
l When downlink 24-layer MU beamforming take effect, the
CellBf.MassiveMIMOMubfPairRule parameter must be set to
MU_WITHOUT_CORR_PAIR_RULE. As a result, the number of pairing layers and
BLER increase and the MCS index decreases.
Function Impacts
Function Function Reference Description
Name Switch
6.2.3 Requirements
6.2.3.1 Licenses
l The operator has purchased and activated the license for TDLEOFD-121604 DL 8-Layer
MU-Beamforming. Four units of the license control item Massive MIMO DL 2-Layers
Extended Processing Unit License(TDD) are required.
Feature ID Feature Model License Sales Unit
Name Control Item
l The operator has purchased and activated the license for TDLEOFD-121605 DL 16-
Layer MU-Beamforming. Eight units of the license control item Massive MIMO DL 2-
Layers Extended Processing Unit License(TDD) are required.
l The operator has purchased and activated the license for TDLEOFD-131603 DL 24-
Layer MU-Beamforming (Trial). 12 units of the license control item Massive MIMO
DL 2-Layers Extended Processing Unit License(TDD) are required.
Feature ID Feature Model License Sales Unit
Name Control Item
6.2.3.2 Software
Prerequisite Functions
l TDLEOFD-121604 DL 8-Layer MU-Beamforming requires the following functions.
Function Name Function Switch Reference
6.2.3.3 Hardware
See 4.3.3 Hardware. Downlink 24-layer MU beamforming (trial) requires a UBBPf1,
UBBPf3, or UBBPfw1 to be configured.
Table 6-3 Optimization parameters for downlink time-domain adaptive extended scheduling
Parameter Parameter ID Setting Notes
Name
Mass Flow Big ServiceIdentify- Set this parameter to a large value when
Packet Rate Para.MassFlowBigPacke- most heavy-traffic services are video or
Threshold tRateThld download services and to a small value
when most heavy-traffic services are
web browsing services.
Step 2 Use the U2020 to perform signaling tracing. For details, see 6.2.4.3 Network Monitoring. If
the measured value of any of the monitoring objects listed in Table 6-4 does not equal 0, MU
beamforming has been activated
Number of Available Number of RBs that can be paired for massive MIMO MU
Massive MIMO MUBF beamforming. It is the average of the cumulative total
Pairing RB (Num) number of RBs paired with the massive MIMO MU
beamforming at each TTI at the end of the monitoring period.
----End
Counter-based Monitoring
Step 1 Monitor counters in the following function subsets:
l Number of PRBs for MUBF UEs at Each Layer
l Number of TTIs During Which UEs Are Successfully Paired in the Downlink
l Number of Layers Paired for Massive MIMO MU Beamforming
l Total Number of RBs Transmitted in the Downlink in Massive MIMO Scenarios
l Total Number of TBs Transmitted in the Downlink in Massive MIMO
l Number of Bits Successfully Transmitted in the Downlink in Massive MIMO
l Total Number of TBs Incorrectly Transmitted in the Downlink in Massive MIMO
----End
When massive MIMO MU beamforming is enabled, the number of pairing layers and PRBs
increases and the measured values of the following counter may reach 64 bits.
Signaling Tracing
l Cell-level MU beamforming monitoring
Step 1 On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace Management.
Step 2 From the navigation tree on the left of the Signaling Trace Management tab page, choose
LTE > Cell Performance Monitoring > Multi-User BF (Cell) Monitoring. The Multi-User
BF (Cell) Monitoring dialog box is displayed.
Step 3 In the displayed dialog box, enter the tracing task name in the Trace Name text box, select
the eNodeB to be monitored, and click Next.
Step 4 Enter the local cell ID, and select Massive MIMO MUBF Pair in the Test Items area.
Step 5 On the Signaling Trace Management tab page, double-click the trace task that you have
created to view the monitoring result. For the detailed description of the monitoring result,
press F1 to obtain the online help.
----End
l UE-level MU beamforming monitoring
Step 1 On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace Management.
Step 2 From the navigation tree on the left of the Signaling Trace Management tab page, choose
LTE > User Performance Monitoring > Multi-User BF (User) Monitoring. The Multi-
User BF (User) Monitoring dialog box is displayed.
Step 3 In the displayed dialog box, enter the tracing task name in the Trace Name text box, select
the eNodeB to be monitored, and click Next.
Step 4 In the displayed dialog box, enter the values in the MMEc and mTMSI text boxes, and click
Finish.
Step 5 On the Signaling Trace Management page, double-click the tracing task that you have
created to view the values of the measurement items Indep-Indep Number of successful
MUBF pairing RB and Indep-Joint Number of successful MUBF pairing RB in the
monitoring result. For the detailed description of the monitoring result, press F1 to obtain the
online help.
----End
6.3 TM9
For details about the network analysis and running environment of the TM9 feature, see
Beamforming (TDD).
6.3.1 Principles
The following TM9 features are available in massive MIMO scenarios:
For details about TM9, see Beamforming (TDD). The availability of the TM9 features
increases the UE transmission modes in massive MIMO scenarios. TM9 hybrid precoding
requires downlink 2-layer MIMO based on TM9 or downlink 4-layer MIMO based on TM9 to
be enabled.
CSI-RS Period CellCsiRsParaCfg.CsiRs Set this parameter to ms80. The cell will
Period reset if the value is changed.
uplink CA is enabled
MOD CELLBFMIMOPARACFG: LocalCellId=0, SccBfMimoAdaptiveSwitch=ON;
//Disabling OTDOA
MOD ENODEBALGOSWITCH: LcsSwitch=OtdoaSwitch-0;
6.4.1 Principles
TM9 hybrid precoding exploits the array antenna form of massive MIMO networking to
further improve the performance of the following TM9 features:
For details about the preceding TM9 features, see Beamforming (TDD).
6.4.2.1 Benefits
l Improved downlink cell throughput and spectral efficiency
l Improved throughput of TM9 UEs
l Total increased proportion of 4R UEs working in TM9 rank 3 or 4
It is recommended that basic massive MIMO functions be used with TM9 hybrid precoding,
especially when there are many TM9-capable 4R UEs. TM9 hybrid precoding must be
enabled if split SDMA is required for closed-loop TM9 UEs.
6.4.2.2 Impacts
Network Impacts
l The gain may be negative when the TM9 UE penetration rate is low.
l After this feature is enabled, the gains for 8R UEs not located at the cell center or edge
are greater than the gains for 4R UEs at the same position.
Function Impacts
Function Function Switch Reference Description
Name
6.4.3 Requirements
6.4.3.1 Licenses
Feature ID Feature Name Model Sales Unit
6.4.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
6.4.3.3 Hardware
See 4.3.3 Hardware.
6.4.3.4 Others
For details, see 4.3.4 Others.
l Beam split weight values are available only when the beam weight file inside the eNodeB
software package is activated. If no beam split weight value is available, enabling TM9
hybrid precoding will lead to cell faults.
l Run the ACT BFANTDB command with the OPMODE parameter set to SWFILE to
activate the beam weight file.
6.5.1 Principles
Multi-user split SDMA enhances the capacity of massive MIMO cells in which MU
beamforming is enabled. It improves downlink service performance of massive MIMO cells
through SDMA on PDSCH split beams in the following scenarios:
l Moving UEs need to be served.
l SRS interference needs to be suppressed.
It is recommended that this function be used with TM9 hybrid precoding so that multi-user
split SDMA takes effect on closed-loop TM9 rank 1 and rank 2 UEs served by the SCC. In
this way, downlink service performance improves in densely populated urban areas with
strong coverage. For details about TM9 hybrid precoding, see 6.4 TM9 Hybrid Precoding.
This feature is controlled by the SplitSDMASwitch option under the
CellAlgoSwitch.MuBfAlgoSwitch parameter. Multi-user split SDMA is disabled if this
option is deselected. The following table describes the principles applicable when this option
is selected.
If... Then...
Table 6-7 Mechanism for entering and exiting multi-user split SDMA
Whether TM9 Entering or Condition
Hybrid Precoding Exiting Multi-
Is Enabled User Split SDMA
NOTE
l UEs with limited SRS resources include moving UEs and UEs subjected to SRS interference.
l A UE is identified as the UE subjected to SRS interference when the SRS SINR of the UE less than
the value of the CellMMAlgo.SplitSdmaUeIdenSrsSinrThld parameter for a specified number of
consecutive times.
If... Then...
– UEs subjected to limited SRS resources support only split SDMA and can
participate in split SDMA pairing.
– For normal UEs, MU beamforming applies. Normal UEs can be paired with UEs
subjected to limited SRS resources or closed-loop TM9 rank 1 and rank 2 UEs
served by the SCC for SDMA pairing.
6.5.2.1 Benefits
This function improves the average downlink spectral efficiency and average downlink user-
perceived rate in target application scenarios.
The following table describes the application scenarios of massive MIMO split SDMA.
Strong coverage scenarios, Characteristics: strong capacity demand, high user density,
such as central business scattered user distribution, and high proportion of UEs
districts (CBDs) and In such scenarios, multi-user split SDMA improves the
campuses number of paired layers for closed-loop TM9 rank 1 and rank
2 UEs served by the SCC.
Mobility scenarios, such Characteristics: heavy traffic, strong capacity demand in the
as traffic hubs and rally uplink and downlink, high user density, scattered user
scenarios distribution, and most UEs having mobility characteristics
suitable for beam-domain pairing
In such scenarios, multi-user split SDMA increases the
beamforming beamwidth and reduces interference between
beams and the mobility sensitiveness of SDMA.
Scenarios with limited Characteristics: many users, huge capacity demand, and
SRS resources, such as clearly insufficient SRS resources or severe uplink
stadiums, concerts, and interference, leading to low SRS SINR
contiguous networking In such scenarios, multi-user split SDMA improves pairing
areas performance regardless of SRS weight estimation.
The following counters can be used to evaluate the SRS
interference in the local cell:
l L.UL.SINR.SRS.Index0
l L.UL.SINR.SRS.Index1
l L.UL.SINR.SRS.Index2
l L.UL.SINR.SRS.Index3
l L.UL.SINR.SRS.Index4
Determine whether to enable multi-user split SDMA based on
the SRS interference. For example, if the proportion of
samples whose SRS SINR is within range 0 is greater than
40%, there is SRS interference in the local cell and multi-user
split SDMA is recommended. The proportion of samples
whose SRS SINR is within range 0 is calculated using the
following formula:
Proportion of samples whose SRS SINR is within range 0 =
L.UL.SINR.SRS.Index0/(L.UL.SINR.SRS.Index0 +
L.UL.SINR.SRS.Index1 + L.UL.SINR.SRS.Index2 +
L.UL.SINR.SRS.Index3 + L.UL.SINR.SRS.Index4)
6.5.2.2 Impacts
Network Impacts
l This function improves the average downlink spectral efficiency of the cell.
l This function improves the average downlink UE-perceived rate.
l If the number of downlink scheduled UEs increases in the subframes where both uplink
and downlink CCE resources are allocated, more CCEs are used in the downlink and
fewer CCEs are available in the uplink. As a result, the number of UEs scheduled in the
uplink decreases, affecting uplink capacity.
Function Impacts
Function Function Switch Reference Description
Name
6.5.3 Requirements
6.5.3.1 Licenses
Feature ID Feature Name Model Sales Unit
6.5.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
6.5.3.3 Hardware
See 4.3.3 Hardware.
6.5.3.4 Others
For details, see 4.3.4 Others.
l Beam split weighting values are available only when the beam weight file inside the
eNodeB software package is activated. If no beam split weighting value is available,
enabling multi-user split SDMA will lead to cell faults.
l Run the ACT BFANTDB command with the OPMODE parameter set to SWFILE to
activate the beam weight file.
NOTE
The parameter settings in the following examples are used for reference only. Set the parameters based
on network requirements.
//Activating the beam weight file in the software package
ACT BFANTDB: OPMODE=SWFILE;
//Turning on the split SDMA switch
MOD CELLALGOSWITCH: LocalCellId=0, MuBfAlgoSwitch=SplitSDMASwitch-1;
//(Optional) Configuring the PDSCH split beam configuration index to the default
beam configuration template and the split SDMA adaptation period to 15 minutes,
and enabling split SDMA for moving UEs
MOD CELLBF: LocalCellId=0, PdschSplitBeamConfigIndex=CONFIG_DEFAULT,
SplitSdmaAdaptionPeriod=PERIOD_15MIN, MovingUeMuBfScheme=SPLIT_SDMA_SCHEME;
//(Optional) Configuring the threshold of the proportion of UEs with insufficient
SRSs, the threshold of the proportion of TM9 for triggering split SDMA, the SRS
SINR threshold for identifying UEs available for split SDMA, and split SDMA beam
RSRP isolation threshold based on the network plan
MOD CELLMMALGO: LocalCellId=0, ASplitSdmaSrsLmtUePropThld=10,
ASplitSdmaTm9UePropThld=10, SplitSdmaUeIdenSrsSinrThld=-12,
SplitSdmaBeamRsrpIsltnThld=5;
//(Optional) Enabling multi-user split SDMA for closed-loop TM9 UEs
MOD CELLALGOSWITCH: LocalCellId=0, EnhMIMOSwitch=TM9HybridPrecodingSwitch-1,
MuBfAlgoSwitch=TM9SplitSdmaSwitch-1;
You can perform signaling tracing on the U2020 to check whether multi-user split SDMA
takes effect. For details, see "user-level beam-domain monitoring items" in Monitoring
Through Signaling Tracing.
Counter-based Monitoring
The following counters can be used to monitor resource usages in different transmission
modes.
e. On the Signaling Trace Management tab page, double-click the trace task that you
have created to view the monitoring result. For the detailed description of the
monitoring result, press F1 to obtain the online help.
l User-level beam-domain monitoring items: to monitor the PDSCH target beam and
working beam information of a UE working in multi-user split SDMA
a. On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace
Management.
b. In the navigation tree on the left of the Signaling Trace Management tab page,
choose Trace Type > LTE > User Performance Monitoring > Beam Space
(User) Monitoring. In the displayed dialog box, select PDSCH Beam Defect.
c. In the displayed dialog box, enter the tracing task name in the Trace Name text
box, select the eNodeB to be monitored, and click Next.
d. Enter the mobility management entity code (MMEC) and MME temporary mobile
subscriber identity (mTMSI) of the UE to be traced, and click Finish. The trace task
is started on the U2020.
e. In the Signaling Trace Management window, double-click the created trace task to
view the monitoring results of PDSCH Beam Flag, PDSCH Beam Identifier,
PDSCH Beam SRS RSRP. For the detailed description of the monitoring result,
press F1 to obtain the online help.
Step 1 Check whether the feature, hardware and license control meet the requirements described in
6.5.3 Requirements.
Step 2 Check whether parameters are properly set by referring to 6.5.4.1.1 Data Preparation.
Step 3 Run the DSP BFANTDB command and check whether the beam weight file is activated by
referring to 6.5.4.1.2 Using MML Commands.
----End
6.6.1 Principles
This feature increases the PDCCH capacity. After the beam weight file is loaded, the PDCCH
wide beam is split into multiple PDCCH narrow beams, as shown in Figure 6-10. The half-
power beamwidth (HPBW) of the narrow beams is similar to that of the wide beam. The
eNodeB determines whether the UE is an independent-scheduling UE or a joint-scheduling
UE based on beam associations and isolations of the UE. Each PDCCH narrow beam is
independently transmitted to an independent-scheduling UE, enabling PDCCH resources to
be shared by multiple UEs served by different beams and enhancing PDCCH capacity. The
beams in the working beam list of joint-scheduling UEs perform joint transmission in a way
of scheduling same as that used before the split.
NOTE
l The eNodeB automatically identifies the CCE aggregation level of a UE, free from any
parameter control.
l The SINR is low for UEs with high CCE aggregation levels. For such UEs, adjust the
setting of the CellPdcchAlgo.SplitBeamPdcchSdmaThldOfs parameter to slightly
decrease the isolation threshold so that the UEs are more probably scheduled as
independent-scheduling UEs and more probably enter PDCCH SDMA.
4. Determines UE attributes based on the number of narrow PDCCH beams in the working
beam list of the UE.
– If the working beam list of the UE contains only one narrow PDCCH beam, the UE
is an independent-scheduling UE.
– If the working beam list of the UE contains multiple narrow PDCCH beams, the UE
is a joint-scheduling UE.
6.6.2.1 Benefits
This feature increases the PDCCH capacity and the average UE throughput in the uplink and
downlink.
This feature is recommended when all of the following conditions are met in massive MIMO
scenarios:
6.6.2.2 Impacts
Network Impacts
l This feature increases the PDCCH capacity, allows more UEs to be scheduled, and
improves the average UE throughput in the uplink and downlink when all of the
following conditions are met. This feature may decrease the spectral efficiency, cell
throughput, and UE-perceived rate if any of the following conditions is not met:
– There are big events.
– Uplink PDCCH resources are severely limited.
– Uplink resources are more limited than downlink resources.
– UEs are evenly distributed on the horizontal plane.
l PDCCH beam split and coverage changes induced by this feature may lead to
fluctuations in the number of UEs, reduce CQIs, and affect TM3 UE performance.
Consequently, this feature has an impact on the PRB usage, pairing performance,
spectral efficiency, cell throughput, and UE-perceived rate.
Function Impacts
None
6.6.3 Requirements
6.6.3.1 Licenses
Feature ID Feature Name Model Sales Unit
6.6.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
6.6.3.3 Hardware
See 4.3.3 Hardware.
6.6.3.4 Others
For details, see 4.3.4 Others.
l Beam split weighting values are available only when the beam weight file is activated. If
no beam split weighting value is available, enabling PDCCH SDMA will lead to cell
faults.
l Run the ACT BFANTDB command with the OPMODE parameter set to SWFILE to
activate the beam weight file.
NOTE
The parameter settings in the following examples are used for reference only. Set the parameters based
on network requirements.
Step 1 On the U2020 client, choose Monitor > Signaling Trace > Signaling Trace Management.
Step 2 From the navigation tree on the left of the Signaling Trace Management tab page, choose
LTE > Cell Performance Monitoring > DCI Statistic Monitoring. On the displayed DCI
Statistic Monitoring dialog box, select PDCCH SDMA in Split Beam DCI STATISTIC.
Step 3 Set Local cell ID, Monitor Period (ms), and UE type, and click Finish. The monitoring task
is initiated.
Step 4 On the Signaling Trace Management tab page, double-click the trace task that you have
created to view the monitoring result. PDCCH SDMA in massive MIMO has taken effect if
the number of CCEs successfully paired within the monitoring period is not 0.
----End
Fault handling:
Step 1 Check whether feature, hardware and license control meet the requirements described in 6.6.3
Requirements.
Step 2 Check whether parameters are properly set by referring to 6.6.4.1.1 Data Preparation.
Step 3 Check UE distributions. If the UEs are located near each other, the UEs may receive the same
beam, increasing the probability that no CCE resources are paired. In this situation, adjust the
UE locations or add UEs in dispersed locations and then monitor the pairing performance
again.
----End
Fault description: The feature does not take effect after the eNodeB software is upgraded,
though the feature switch has been turned on.
Fault handling:
Step 1 Check whether feature, hardware and license control meet the requirements described in 6.6.3
Requirements.
Step 2 Check whether parameters are properly set by referring to 6.6.4.1.1 Data Preparation.
Step 3 Run the DSP BFANTDB command and check whether the beam weight file is activated by
referring to 6.6.4.1.2 Using MML Commands.
----End
6.7 TMA
6.7.1 Principles
transmission mode for a UE based on the radio channel quality of the UE after it accesses the
network. In this way, the spectral efficiency is improved.
According to Table 6-11, TMA can be performed among TM2, TM3, TM7, TM8, and PMI
feedback-dependent TM9 in massive MIMO scenarios. Table 6-12 illustrates the TMA
principles.
Figure 6-12 TMA for low-speed UEs compliant with 3GPP Release 8 or 9 and incapable of
transmit antenna selection
Figure 6-13 TMA for low-speed UEs compliant with 3GPP Release 9 and capable of transmit
antenna selection
Figure 6-14 TMA for low-speed UEs compliant with 3GPP Release 10 and incapable of
transmit antenna selection
Figure 6-15 TMA for low-speed UEs compliant with 3GPP Release 10 and capable of
transmit antenna selection
TM7 TM7
This setting improves the proportion of beamforming,
downlink spectral efficiency, and average downlink cell
throughput. When the network is lightly loaded and the
proportion of small packets is high, the UE-perceived rate may
decrease.
After this function is enabled, the eNodeB periodically checks the service load of a cell. In
heavy-load scenarios, the eNodeB calculates the overall spectral efficiency of a cell separately
for cell-level TMA and the target beamforming mode. Then, the eNodeB automatically
selects the transmission mode that maximizes the spectral efficiency of the cell. This function
may cause the cell spectral efficiency to fluctuate due to cell-level TMA attempts.
Step Description
Step Description
Determine whether the cell load A cell is heavily loaded if both of the following
is heavy. conditions are met:
l Average downlink PRB usage of the cell > Value of
the CellBfMimoParaCfg.HeavyLoadDlPrbUsa-
geThld parameter
l Average number of UEs in a cell > Value of the
CellBfMimoParaCfg.HeavyLoadAvgUeCountThl
d parameter
6.7.2.1 Benefits
The eNodeB adaptively selects a proper transmission mode for a UE based on the radio
channel quality of the UE, maximizing the spectral efficiency.
6.7.2.2 Impacts
Network Impacts
l TMA enabled but transmission mode control for UE access disabled: The proportion of
TM2 UEs may be high due to the time required for switching from TM2 to TM7. As a
result, the number of downlink paired layers is less than expected.
l TMA and transmission mode control for UE access both enabled: When the network is
lightly loaded and the proportion of small packets is high, the UE-perceived rate may
decrease.
l Heavy-load TMA boost: The cell spectral efficiency may fluctuate due to cell-level
TMA tries.
l Optimized TMA for 4R and 8R UEs increases the proportion of beamforming, downlink
UE-level spectral efficiency, and downlink UE-perceived data rate. The proportion of
PMI feedback-dependent TM9 may decrease.
l If WBB UEs have been online for a long time, switching of a WBB UE to TM7 when
the cell is in the heavy-load beamforming state under optimized switching to
beamforming in heavy-load scenarios increases the number of pairing layers, spectral
efficiency, and UE-perceived data rate in the downlink.
l In heavy-load scenarios, optimized heavy-load TMA boost under optimized switching to
beamforming in heavy-load scenarios enables the cell-level historical spectral efficiency
to more accurately adapt to changes in the number of UEs and load status. It reduces the
overhead of heavy-load TMA detections when the cell is in heavy-load beamforming
state.
Function Impacts
None
6.7.3 Requirements
6.7.3.1 Licenses
None
6.7.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
6.7.3.3 Hardware
See 4.3.3 Hardware.
Step 2 Run the LST CELLBFMIMOPARACFG command to check whether the value of UE
Access Transmission Mode is TM7. If yes, transmission mode control for UE access has
been enabled.
Step 3 Run the LST CELLBF command to check whether the value of Cell Highload Thd for
Switch to BF is 0. If no, switching to TM7 in multi-layer pairing scenarios has been enabled.
Step 5 Run the LST CELLBFMIMOPARACFG command to check whether the value of
Handover BF Threshold Adjust Switch is On. If yes, handover-specific beamforming
threshold adjustment has been enabled.
Step 6 Run the LST CELLBF command. If the value of Heavy-Load Switch to BF Optimization
is not 255, optimized switching to beamforming in heavy-load scenarios has been enabled.
----End
The following table lists the counters related to the number of WBB UEs after optimized
switching to beamforming in heavy-load scenarios is enabled.
After heavy-load TMA boost and optimized heavy-load TMA boost are enabled, the downlink
UE pairing rate and spectral efficiency improve in heavy-load scenarios. The following
counters used to observe the performance of the functions.
1526747714 L.HeavyLoad.TmaToTmTarget.Num.DL.Cell
1526747715 L.HeavyLoad.TmTargetToTma.Num.DL.Cell
6.8.1 Principles
This feature enables power coordination between the serving massive MIMO cell and the
neighboring non-massive MIMO cells that cause strong interference to the serving cell. In this
way, this feature improves the downlink spectral efficiency of massive MIMO cells. Figure
6-17 shows the principles of enhanced coordinated scheduling-based power control.
NOTE
In inter-eNodeB networking scenarios, the serving cell interacts with the neighboring cells causing
strong interference through the eX2 interface when the USU3910 is configured or through the ODI
interface when the USU3900 is configured.
l If two values are selected, the eNodeB adjusts the downlink transmit power of the cell
between the value of the CellDlpcPdschPa.PaPcOff parameter and the two values of
this parameter.
For example:
If the UE whose transmit power is to be updated uses QPSK or beamforming, the eNodeB
changes the transmit power of all UEs of this type. In this scenario, the transmit power change
is transmitted only to the physical layer and is not notified to the UE through RRC signaling.
NOTE
This step does not take effect for voice UEs, push to talk (PTT) UEs, enhanced Machine Type
Communication (eMTC) UEs, and Coordinated MultiPoint transmission/reception (CoMP) UEs.
6.8.2.1 Benefits
This function is recommended only when all of the following conditions are met:
l Networking: The serving cell is a massive MIMO cell and multiple non-massive MIMO
macro neighboring cells cause strong interference to the serving cell.
l Load requirements: The PRB usage of any cell is greater than 80%.
l Overlapping coverage rate: More than 40% of UEs in the massive MIMO cell are subject
to interference from two neighboring cells. More than 40% of UEs in non-massive
MIMO cells are subject to interference from another non-massive MIMO cell.
NOTE
If the UE-reported RSRP difference between a neighboring cell and the serving cell is less than 6
dB, the UE is an interfering UE.
l CQI value: The average CQI value of the massive MIMO cell is low due to interference
from neighboring cells.
When all of the preceding conditions are met, enabling this function improves the downlink
spectral efficiency of the massive MIMO cell and that of the entire network. The downlink
spectral efficiency of non-massive MIMO neighboring cells coordinating with the serving
cell, however, may decrease.
6.8.2.2 Impacts
Network Impacts
l Inter-cell power coordination improves the downlink spectral efficiency of heavy-load
massive MIMO cells in densely populated urban areas.
l If the inter-cell inter-RRU distance is small and the overlapping coverage area between
cells is large, the inter-cell power coordination gain increases.
l If the distribution of downlink PRB usages between cells is imbalanced, the gain of
inter-cell power coordination increases.
l Coordinated scheduling based power control requires downlink RSRP measurements.
This feature is recommended for networks where most UEs move at speeds less than 10
km/h. If most UEs move at higher speeds, eNodeBs cannot track fast interference
changes.
l When the cell attribute changes from a macro cell to a micro cell due to the configuration
modification, the PA of the cell may fail to be updated. As a result, the cell performance
deteriorates.
Function Impacts
None
6.8.3 Requirements
6.8.3.1 Licenses
Feature ID Feature Name Model Sales Unit
6.8.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
6.8.3.3 Hardware
The serving cell is a massive MIMO cell. For details about hardware requirements, see 4.3.3
Hardware.
The neighboring cells are non-massive MIMO cells. The hardware requirements are as
follows:
l Site type: 3900 and 5900 series base stations
l Boards: BBPs other than the LBBPc or LBBPd
l RF modules: no requirements
6.8.3.4 Others
PLMN information must be configured for the eNodeBs that serve the cells in the coverage
area of this feature. If the PLMN information is not configured, routes cannot be set up.
l The parameter settings in the following examples are used for reference only. Set the parameters
based on network requirements.
l After this feature is enabled, it takes about 5 minutes to set up a coordinating neighboring cell for the
first time.
MOD CELLCSPCPARA: LocalCellId=0, eCspcPCAdjUeNumTh=1, eCspcA3Offset=-12,
CelleCspcSwitch=ON, eCspcPCAdjRange=DB_6_P_A-1&DB_4DOT77_P_A-1;
----End
Inter-frequency load steering in massive MIMO scenarios enables the eNodeB to transfer the
uplink synchronized UEs suitable for massive MIMO between inter-frequency non-massive
MIMO cells and massive MIMO cells, improving traffic volume and spectral efficiency of
massive MIMO cells. For details, see Intra-RAT Mobility Load Balancing.
7.2.1 Principles
Figure 7-2 Background for introducing preferential scheduling of UEs with low aggregation
levels
Figure 7-3 Principle of preferential scheduling of UEs with low aggregation levels
This function takes effect only when all of the following conditions are met:
l Uplink-downlink subframe configuration 2 is used.
l A massive MIMO cell is set up.
l The MuBfSwitch option is selected under the CellAlgoSwitch.MuBfAlgoSwitch
parameter.
7.2.2.1 Benefits
PDCCH resource allocation has been optimized in massive MIMO scenarios, preventing cell
capacity from being suppressed when PDCCH resources are insufficient. This function
improves the average cell throughput and average UE throughput in the downlink. This
function can achieve the largest gain if the ratio of aggregation level 8 is greater than or equal
to 10% and the ratio of aggregation level 1 is greater than or equal to 30%.
NOTE
7.2.2.2 Impacts
Network Impacts
l If both PDCCH and PDSCH resources are limited, PDCCH capacity enhancement may
decrease the average downlink cell throughput and downlink spectral efficiency.
l Preferential scheduling of UEs with low aggregation levels may slightly decrease the
CEU perceived rate.
l Enhanced PDCCH symbol adaptation may decrease the average cell and UE throughput
in the downlink in moderate- or light-load scenarios.
NOTE
Function Impacts
None
7.2.3 Requirements
7.2.3.1 Licenses
None
7.2.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.2.3.3 Hardware
See 4.3.3 Hardware.
Step 2 Run the LST CELLPDCCHALGO command to query the value of the PDCCH Symbol
Number Adjust Switch parameter. Enhanced symbol adaptation has been enabled if the
value is Enhanced CFI Adaption On.
----End
NOTE
7.3.1 Principles
Uplink overload optimization is categorized into uplink retransmission scheduling
optimization and uplink power control optimization.
7.3.2.1 Benefits
l Uplink retransmission scheduling optimization
This feature improves the uplink spectral efficiency in uplink heavy-load scenarios. This
function achieves the maximum gain when the uplink CCE allocation failure rate is less
than 10% and the uplink PRB usage is greater than 80%.
l Uplink power control optimization
This function improves the uplink spectral efficiency in uplink heavy-load scenarios.
This function achieves the maximum gain when the uplink PRB usage is greater than
60%.
7.3.2.2 Impacts
Network Impacts
l Uplink retransmission scheduling optimization
– This function reduces the impact of RB truncations caused by retransmission
scheduling and improves the uplink PRB usage and uplink spectral efficiency.
– The CCE usage increases, the probability of CCE allocation failures increases, the
number of PDCCH symbols increases, and the proportion of PDCCH DTXs slightly
increases.
NOTE
Function Impacts
None
7.3.3 Requirements
7.3.3.1 Licenses
Uplink overload optimization is controlled by the license of massive MIMO introduction.
7.3.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.3.3.3 Hardware
For details, see 4.3.3 Hardware.
7.4.1 Principles
In a single massive MIMO cell, the target value of SRS power control can be increased to
increase the SRS SINR, improving MU beamforming performance. When multiple massive
MIMO cells are configured for contiguous coverage and cell overlapping is severe, increasing
the target value of SRS power control may increase interference between neighboring cells.
As a result, the SRS SINR may decrease and MU beamforming performance may be affected.
To solve the preceding problem, TDLEOFD-131614 SRS NI Joint Suppression is introduced.
This feature covers three functions: SRS resource allocation optimization for interference
avoidance, SRS power control optimization, and SRS interference cancellation (IC).
cells are deployed on the same eNodeB and coverage overlap is severe in big-event scenarios.
This function is controlled by the SRSCfg.SrsInterfAvoidOptSwitch parameter and requires
massive MIMO introduction to be enabled.
NOTE
The ANTICLOCKWISE_COMB1 value is used for performance guarantee only in certain special
networking scenarios. For example, the coverage directions of two inter-eNodeB cells are the opposite
of each other. It is not recommended that the SRSCfg.SrsInterfAvoidOptSwitch parameter be set to
ANTICLOCKWISE_COMB1. Consider the networking conditions if you want to use the value.
Otherwise, the setting may lead to the identical comb preferentially allocated for SRS transmission of
the UEs that interfere with each other in the overlapping coverage areas, increasing interference.
This function is recommended only when massive MIMO cells are involved to provide
contiguous coverage.
This function is enabled when the SrsPcSwitch option is selected under the
CellAlgoSwitch.UlPcAlgoSwitch parameter and the CellPcAlgo.SrsPcStrategy parameter is
set to SRSPC_SINR_ADAPTIVE. Each massive MIMO cell periodically calculates the
average SRS NP of the cell and determines whether to send SRS NP indicator to cells in the
interference neighboring cell list based on the following principles.
If... Then...
The average SRS noise power (NP) value is Sends the SRS-NP-HIGH IE to neighboring
greater than the upper threshold of –112 cells in the interference neighboring cell list.
dBm, and there are more than 100 UEs in
the cells
The average SRS NP value is less than the Sends SRS-NP-LOW signals to neighboring
lower threshold of –115 dBm cells in the interference neighboring cell list.
The average SRS NP value ranges from the Sends SRS-NP-HOLD signals to
upper threshold to the lower threshold neighboring cells in the interference
neighboring cell list.
After receiving the SRS-NP-Indicator message, the massive MIMO or non-massive MIMO
cell determines whether to adjust the UE-level target SRS SINR value for power control
based on the following principles.
If... Then...
The SRS-NP-HIGH signal sent by the Decreases the UE-level target SRS SINR
neighboring cell is received value for power control.
The SRS-NP-LOW signal sent by the Increases the UE-level target SRS SINR
neighboring cell is received value for power control.
If... Then...
The SRS-NP-HOLD signal sent by the Does not adjust the UE-level target SRS
neighboring cell is received SINR value for power control.
NOTE
l The UE-level target SRS SINR value for power control is an internal system parameter and does not
need to be manually configured. Its initial value is controlled by the CellPcAlgo.SrsPcSinrTarget
parameter.
l The SRS NP value refers to the interference noise intensity measured on the SRS symbol, which is
obtained through SRS measurement.
SRS IC
As shown in Figure 7-6, SRS IC reduces or eliminates the SRS interference of neighboring
cells by means of interference cancellation, improving the SRS SINR.
This function is recommended only when massive MIMO cells are involved to provide
contiguous coverage. It is recommended that SRS SDNR be also enabled to further improve
performance.
7.4.2.1 Benefits
l SRS resource allocation optimization for interference avoidance: Improves MU
beamforming performance when multiple massive MIMO cells are deployed on a single
eNodeB.
l SRS power control optimization: Improves MU beamforming performance by adaptively
adjusting the target SRS SINR value for power control.
l SRS IC: Reduces or eliminates SRS interference of neighboring cells and improves MU
beamforming performance.
7.4.2.2 Impacts
Network Impacts
l MU beamforming performance improves.
l SRS resource allocation optimization for interference avoidance may prolong the SRS
period of certain UEs when UEs are unevenly distributed in the cell. A prolonged SRS
period affects the beamforming performance of moving UEs and may decrease the
downlink cell throughput.
l SRS power control optimization is designed to increase the average SRS SINR in
contiguous coverage areas. If the uplink SRS SINR of a single light-load cell is high, this
function may decrease its SRS SINR in a contiguous coverage area. However, the
average SRS SINR improves for the coverage area in which SRS power control is
enabled.
l SRS power control optimization increases the number of RRC connection
reconfiguration messages, leading to an increased service drop rate.
l After SRS IC is enabled:
– Uplink performance improves, which may increase the number of uplink Transfer
Control Protocol (TCP) packets. As a result, the uplink spectral efficiency may
decrease.
– Uplink pilot timeslot (UpPTS) interference decreases, improving the SRS SINR
distributions and affecting the number of downlink pairing layers.
Function Impacts
None
7.4.3 Requirements
7.4.3.1 Licenses
Feature ID Feature Name Model Sales Unit
7.4.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.4.3.3 Hardware
See 4.3.3 Hardware. There are no hardware requirements, except that the UBBPem does not
support SRS IC.
//Enabling SRS IC
//Configuration of the local cell
MOD PUSCHCFG: LocalCellId=0, GroupHoppingEnabled=BOOLEAN_FALSE,
SeqHoppingEnabled=BOOLEAN_FALSE;
MOD CELLALGOSWITCH: LocalCellId=0, SrsAlgoSwitch=SrsTimeMeasSwitch-1;
MOD NCELLSRSMEASPARA: LocalCellId=0, SrsAutoNCellMeasSwitch=ON,
NCellMeasSwitch=NCellDmrsMeasSwitch-0&NCellSrsTimeMeasSwitch-1;
MOD CELLULICALGO: LocalCellId=0, SrsIcSwitch=MASSIVE_MIMO_SRS_IC_SW-1;
MOD CELLMMALGO: LocalCellId=0, MMAlgoOptSwitch=SRS_SDNR_SWITCH-1;
//Configurations of the neighboring cell
MOD PUSCHCFG: LocalCellId=1, GroupHoppingEnabled=BOOLEAN_FALSE,
SeqHoppingEnabled=BOOLEAN_FALSE;
MOD CELLALGOSWITCH: LocalCellId=1, SrsAlgoSwitch=SrsTimeMeasSwitch-1;
MOD NCELLSRSMEASPARA: LocalCellId=1, SrsAutoNCellMeasSwitch=ON,
NCellMeasSwitch=NCellDmrsMeasSwitch-0&NCellSrsTimeMeasSwitch-1;
----End
7.5.1 Principles
SRS Time-Domain Measurement
This function enables SRS measurement in the time domain. Beam weight values are
calculated based on SRS measurement results. In massive MIMO scenarios, SRS time-
domain measurement is recommended to improve SRS measurement accuracy and SRS
SINR, enhancing the performance of SU beamforming and MU beamforming.
This function is controlled by the SrsTimeMeasSwitch option under the
CellAlgoSwitch.SrsAlgoSwitch parameter. When this option is selected, SRS time-domain
measurement is enabled. When this option is deselected, SRS time-domain measurement is
disabled. In this case, frequency-domain measurement is used.
SRS SDNR
This function uses massive antennas to identify SRS spatial-domain characteristics and takes
SRS interference of UEs in neighboring cells into account during channel estimation. In this
way, this function improves the channel estimation accuracy, the average downlink UE
throughput, and average number of MU beamforming pairing layers. It is recommended that
this function be enabled in massive MIMO scenarios. In addition, it is recommended that SRS
time-domain measurement be used with SRS SDNR to achieve optimal performance.
7.5.2.1 Benefits
l SRS time-domain measurement
This function improves the SRS measurement accuracy and SRS SINR, enhancing the
performance of SU beamforming and MU beamforming.
l SRS SDNR
This function improves the channel estimation accuracy, the average downlink UE
throughput, and average number of MU beamforming pairing layers. The gain of this
function decreases if the target cell is opposite the interfering cell. However, the overall
gain is not negative.
NOTE
7.5.2.2 Impacts
Network Impacts
l SRS time-domain measurement
This function improves the SRS measurement accuracy and SRS SINR, enhancing the
performance of SU beamforming and MU beamforming.
l SRS SDNR
– This function improves the channel estimation accuracy, the average downlink UE
throughput, and average number of MU beamforming pairing layers. The gain of
this function decreases if the target cell is opposite the interfering cell. However, the
overall gain is not negative.
– SRS SDNR has impact on the following counters:
n L.UL.SINR.SRS.Index0
n L.UL.SINR.SRS.Index1
n L.UL.SINR.SRS.Index2
n L.UL.SINR.SRS.Index3
n L.UL.SINR.SRS.Index4
Function Impacts
None
7.5.3 Requirements
7.5.3.1 Licenses
None
7.5.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.5.3.3 Hardware
See 4.3.3 Hardware.
SRS SDNR
Run the LST CELLMMALGO command to query the status of the SRS_SDNR_SWITCH
option under the MMAlgoOptSwitch parameter. If SRS_SDNR_SWITCH:On is displayed,
SRS SDNR is enabled.
SRS SDNR
Counter ID Counter Name Description
NOTE
7.6.1 Principles
The weight filtering function has been introduced. WTTx UEs generally remain stationary
and channel quality changes are typically caused by external noises. Weight filtering can resist
noise and improve channel quality. This function takes effect only in massive MIMO cells and
is recommended in WTTx or WTTx+MBB hybrid scenarios.
With this function, the eNodeB identifies WTTx UEs based on SPIDs or QCIs, or using
device-pipe identification and performs weight filtering on such UEs when the
CellBf.BfWeightOptDegreeFactor parameter is set to a non-zero value and the radio channel
quality is less than the value of the CellBf.BfWeightOptChnQltyThldOfs parameter plus the
channel quality threshold. The channel quality threshold is fixed to 0 dB in the current
version.
7.6.2.1 Benefits
This function improves the downlink spectral efficiency of CEUs with low SINRs in WTTx
or WTTx+MBB scenarios where massive MIMO cells are configured.
7.6.2.2 Impacts
Network Impacts
In WTTx scenarios or WTTx+MBB hybrid scenarios, this feature:
l Increases the downlink spectral efficiency and MCS indexes and may reduce the number
of RBs paired for MU beamforming at each layer for UEs with low SINRs. The gain is
inversely proportionate to the SINR. The gain is not significant for UEs with high SINRs
and even is negative in mobile scenarios.
l Slightly improves the downlink spectral efficiency of the entire network if most UEs are
allocated with short SRS periods and are not located near the cell center.
Function Impacts
Function Function Switch Reference Description
Name
7.6.3 Requirements
7.6.3.1 Licenses
Beamforming weight optimization uses the license of Massive MIMO introduction.
7.6.3.2 Software
Prerequisite Functions
Function Function Switch Reference Description
Name
7.6.3.3 Hardware
See 4.3.3 Hardware.
7.7 IRC
This section describes TDLEOFD-150613 Advanced Massive MIMO IRC.
7.7.1 Principles
On the basis of uplink 2x64 MU-MIMO, IRC improves physical-layer reception performance
as follows:
l Uses high-level IRC to better mitigate interference from neighboring cells.
l Optimizes the interference estimation mechanism in multi-user pairing scenarios,
enhances the antenna equalization robustness, and improves demodulation performance
and post-pairing throughput.
7.7.2.1 Benefits
Advanced massive MIMO IRC achieves higher average uplink cell throughput and uplink
spectral efficiency than 64T64R multiple-antenna reception or IRC in heavy-load scenarios.
The gains are significant when the uplink PRB load is heavy and the number of uplink paired
layers, interference from neighboring cells, and the number of interference sources increase.
The gains can be maximized when the uplink PUSCH PRB usage of a cell is greater than
60%, the average number of uplink pairing layers is greater than 3, and the measured value of
the L.UL.Interference.Avg counter is greater than -108 dBm.
If the equivalent uplink CCE usage relative to Control Format Indicator (CFI) adaptation is
greater than 90%, uplink CCE resources are limited. As a result, the number of uplink paired
UEs is limited, and the gains of advanced massive MIMO IRC are affected. Therefore, it is
recommended that the CellPdcchAlgo.CceMaxInitialRatio parameter be set to 10_1 when
advanced massive MIMO IRC is enabled in uplink heavy-load scenarios.
NOTE
7.7.2.2 Impacts
Network Impacts
l Advanced massive MIMO IRC may increase the interference noise strength due to the
changes in the interference noise measurement mechanism, but demodulation
performance is not affected.
l After advanced massive MIMO IRC takes effect, uplink performance improves. When
uplink scheduling is complete and no service is available in the downlink for a period of
time, UEs may enter the RRC idle mode more quickly. As a result, the number of paging
messages increases, resulting in negative gains in the downlink.
Function Impacts
None
7.7.3 Requirements
7.7.3.1 Licenses
Feature ID Feature Name Model Sales Unit
7.7.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference
7.7.3.3 Hardware
This feature requires the UBBPf1. For details about other hardware requirements, see 4.3.3
Hardware.
7.8.1 Principles
Scheduling optimization involves the following functions: selective HARQ scheduling in
subframe 3 or 8, downlink RB allocation optimization, MCS index selection optimization for
downlink paired UEs with small-amount data transmission, and cell-level outer-loop
adjustment optimization in beamforming.
PRBs of the subframe are used or the number of paired layers is low. To solve this issue,
selective HARQ scheduling in subframe 3 or 8 is introduced. This function is controlled by
the SF_3_8_HARQ_SELECTIVE_SCH_SW option under the
CellPdcchAlgo.PdcchOptimizationSwitch parameter. When this option is selected, the
eNodeB selects the UEs requiring a small number of CCEs and many RBs, for downlink
HARQ retransmission scheduling in subframe 3 or 8 if all of the following conditions are
met:
l A massive MIMO cell is set up.
l Uplink-downlink subframe configuration 2 is used.
l The CellPdcchAlgo.PDCCHAggLvlAdaptStrage parameter is set to
STRATEGYBASEDONCAPACITY.
l The downlink PRB usage is greater than 60%.
l The downlink equivalent CCE usage is less than 60%, the control format indicator (CFI)
value of subframe 3 or 8 is 3, and 12 or fewer downlink CCEs are available.
It is recommended that this function be used with preferential scheduling for UEs with low
aggregation levels (for initial transmission optimization) to increase the number of PRBs
scheduled in subframe 3 or 8, increasing the average cell throughput and average UE
throughput in the downlink.
MCS Index Selection Optimization for Downlink Paired UEs with Small-
Amount Data Transmission
If the amount of downlink data to be initially transmitted is small for paired UEs, the transport
block size (TBS) selected for the initial transmission may be greater than the amount of data
to be transmitted. To resolve this issue, MCS index selection optimization for downlink paired
UEs with small-amount data transmission is introduced. This function is controlled by the
MM_DL_SCH_AND_MCS_OPT_SWITCH option under the
CellMMAlgo.MMAlgoOptSwitch parameter. When this option is selected, the eNodeB
selects a small MCS index while maintaining the number of allocated RBs. This reduces
transmissions of invalid data and the number of retransmissions and increases the average
downlink throughput of paired UEs.
beamforming is introduced. This function optimizes the mechanism for adjusting the initial
CQI outer-loop value. It is controlled by the MM_DL_SCH_AND_MCS_OPT_SWITCH
option under the CellMMAlgo.MMAlgoOptSwitch parameter. After the option is selected:
1. The eNodeB determines UE locations based on the 5-bit full-band CQI reported by UEs.
If the reported 5-bit full-band CQI value is less than or equal to 9, the UE is located at
the cell edge. If the CQI value is greater than 9 but is less than or equal to 15, the UE is
located between the cell center and cell edge. In other situations, the UE is near the cell
center. The eNodeB measures the cell-level outer-loop CQI adjustment value for various
locations in the cell based on each codeword.
NOTE
The reported 5-bit full-band CQI value is calculated based on the 4-bit CQI value reported by the
UE. The conversion formula is as follows: 5-bit full-band CQI value = 2 x 4-bit CQI value – 4
2. When a new UE accesses the network, the eNodeB determines the cell-level outer-loop
CQI adjustment value based on the UE type and uses it as the initial outer-loop CQI
value of the UE.
7.8.2.1 Benefits
l Selective HARQ scheduling in subframe 3 or 8 prevents limited cell capacity caused by
insufficient PDCCH resources in massive MIMO cells, increasing the average downlink
UE and cell throughput.
l Downlink RB allocation optimization increases the average downlink throughput and
downlink spectral efficiency in common scenarios. In light- or medium-load scenarios,
the average downlink UE throughput and downlink spectral efficiency may decrease, and
the PDCCH-related counter values, such as the proportion of DTXs, may slightly
deteriorate, and the service drop rate may slightly increase.
l MCS index selection optimization for downlink paired UEs with small-amount data
transmission reduces the number of retransmissions, shortens the scheduling delay of
small-traffic services, and improves the average downlink throughput of paired UEs and
downlink spectral efficiency.
l Cell-level outer-loop adjustment optimization in beamforming facilitates CQI
convergence in burst service scenarios, improving service channel reliability and cell
spectral efficiency.
NOTE
l The gains of downlink RB allocation optimization depend on the proportion of UEs allocated with a
small number of RBs and proportion of the first and last RBGs scheduled.
l Average downlink UE throughput = (L.Thrp.bits.DL – L.Thrp.bits.DL.LastTTI)/
L.Thrp.Time.DL.RmvLastTTI
l Downlink cell throughput = L.Thrp.bits.DL/L.Traffic.Sch.DL.Num
l Downlink spectral efficiency = L.Thrp.bits.DL/(L.Cell.DL.PDSCH.Tti.Num x
L.ChMeas.PRB.DL.DrbUsed.Avg)
l Proportion of PDCCH DTXs = (L.ChMeas.PDCCH.DL.DTXNum.AggLvl1 +
L.ChMeas.PDCCH.DL.DTXNum.AggLvl2 + L.ChMeas.PDCCH.DL.DTXNum.AggLvl4 +
L.ChMeas.PDCCH.DL.DTXNum.AggLvl8)/(L.ChMeas.PDCCH.AggLvl1Num +
L.ChMeas.PDCCH.AggLvl2Num + L.ChMeas.PDCCH.AggLvl4Num +
L.ChMeas.PDCCH.AggLvl8Num)
7.8.2.2 Impacts
Network Impacts
l Selective HARQ scheduling in subframe 3 or 8
– This function prevents the limited cell capacity caused by insufficient PDCCH
resources and improves the cell throughput and average UE throughput in the
downlink.
– The average downlink cell throughput may decrease.
NOTE
Function Impacts
None
7.8.3 Requirements
7.8.3.1 Licenses
None
7.8.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.8.3.3 Hardware
See 4.3.3 Hardware.
Step 2 Run the LST CELLMMALGO command to query the status of the
DL_RB_ASSIGN_OPT_SWITCH option under the Massive MIMO Algorithm
Optimization Switch parameter. Downlink RB allocation optimization has been enabled if
DL_RB_ASSIGN_OPT_SWITCH:On is displayed.
Step 3 Run the LST CELLMMALGO command to query the setting of the
MM_DL_SCH_AND_MCS_OPT_SWITCH option under the Massive MIMO Algorithm
Optimization Switch parameter. If MM_DL_SCH_AND_MCS_OPT_SWITCH:On is
displayed, MCS index selection optimization for downlink paired UEs with small-amount
data transmission or cell-level outer-loop adjustment optimization in beamforming has been
enabled.
----End
7.9.1 Principles
In massive MIMO cells, antenna selection UEs can deliver better network performance when
working in dual-stream beamforming and being paired for MU beamforming. MU
beamforming pairing of antenna selection UEs in dual-stream beamforming is enabled when
the CellBf.AntSelUEMubfPairMode parameter is set to DUAL_STREAM_PAIR, the
FakeAntSelIdentificationSwitch option is selected under the CellAlgoSwitch.BfAlgoSwitch
parameter, and the CellBf.Tm8PairingOptSwitch parameter is set to ON.
l Offset to the threshold for switching the transmission mode of UEs capable of transmit
antenna selection from TM3 rank 2 to dual-stream beamforming:
CellBfMimoParaCfg.AsUeTM3Rank2ToDualBfOffset
l Offset to the threshold for switching the transmission mode of UEs capable of transmit
antenna selection from dual-stream beamforming to TM3 rank 2:
CellBfMimoParaCfg.AsUeDualBfToTM3Rank2Offset
l Offset to the threshold for switching the transmission mode of UEs capable of transmit
antenna selection from single-stream beamforming to dual-stream beamforming:
CellBfMimoParaCfg.AsUeBfSingleToDualOffset
identification period, unidentified fake antenna selection UEs may mistakenly enter TM8. As
a result, the performance is affected.
A fake antenna selection UE refers to the UE of which the chip supports transmit antenna
selection but the RF module does not support transmit antenna selection. As a whole, fake
antenna selection UEs do not support transmit antenna selection. However, these UEs claim
their support for transmit antenna selection in the reported RRC_UE_CAP_INFO message.
Figure 7-8 shows the differences between antenna selection UEs and fake antenna selection
UEs.
Figure 7-8 Differences between antenna selection UEs and fake antenna selection UEs
NOTE
Unless otherwise specified, UEs incapable of transmit antenna selection include fake antenna selection
UEs.
7.9.2.1 Benefits
This function improves the downlink UE-perceived data rate and downlink cell spectral
efficiency.
7.9.2.2 Impacts
Network Impacts
This function improves the downlink UE-perceived data rate and downlink cell spectral
efficiency.
Function Impacts
Function Name Function Switch Reference Description
7.9.3 Requirements
7.9.3.1 Licenses
None
7.9.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.9.3.3 Hardware
See 4.3.3 Hardware.
7.9.3.4 Others
l UEs support TM8.
l UEs support transmit antenna selection.
l UEs comply with 3GPP Release 9 or later.
Algorithm Switch parameter, scheduling optimization for antenna selection UEs has been
enabled.
Step 4 Run the LST CELLBFMIMOPARACFG command. If
NON_AS_UE_MAX_BF_RANK_CTRL_SW:Off is displayed under the BfMimo
Algorithm Optimization Switch parameter, non-antenna selection UEs cannot enter dual-
stream beamforming (TM8).
----End
Step 2 After identifying the UE, the eNodeB sends an RRC_CONN_RECFG message to instruct the
UE to disable transmit antenna selection, as shown in Figure 7-10. This indicates that fake
antenna selection UE identification has taken effect.
----End
Table 7-8 Monitoring object for MU beamforming pairing of antenna selection UEs in dual-
stream beamforming
Monitoring Object Meaning
7.10.1 Principles
In a massive MIMO cell, if the uplink or downlink UE-perceived rate is less than a specified
target value and congestions occur in the uplink or downlink, the eNodeB uses the access
control (AC barring) mechanism to reduce the UE access opportunity, improving user
experience.
7.10.2.1 Benefits
After this function is enabled in heavy-load scenarios, you can set a proper target uplink or
downlink UE-perceived rate to improve user experience based on the optimal cell capacity
and performance gains.
7.10.2.2 Impacts
Network Impacts
This function prevents the uplink and downlink UE-perceived rates from dropping to 0 Mbit/s
and ensures that the target rate is reached.
Function Impacts
None
7.10.3 Requirements
7.10.3.1 Licenses
None
7.10.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference
7.10.3.3 Hardware
See 4.3.3 Hardware.
7.11.1 Principles
In massive MIMO cells, if PDCCH resources are insufficient in heavy-load scenarios, UEs
performing large-packet services may not be scheduled. As a result, not all the PRBs in the
cell are used or there are still free RB resources in the spatial domain. To solve this issue,
PDCCH resource reservation for big-packet services in heavy-load scenarios is introduced
NOTE
PDCCH resource reservation for big-packet services in heavy-load scenarios and optimized PDCCH
aggregation level selection near the cell center are controlled by the same switch.
7.11.2.1 Benefits
This feature improves the downlink cell throughput and spectral efficiency in heavy-load
scenarios.
7.11.2.2 Impacts
Network Impacts
This feature improves the downlink cell throughput and spectral efficiency in heavy-load
scenarios.
Function Impacts
None
7.11.3 Requirements
7.11.3.1 Licenses
None
7.11.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference
7.11.3.3 Hardware
See 4.3.3 Hardware.
7.12.1 Principles
PDCCH aggregation levels are selected based on the CQI values adjusted based on the
PDSCH BLER. However, in massive MIMO cells, the PDSCH BLER is high due to mobility
and abnormal UEs. This affects the PDCCH aggregation level selection accuracy and severely
limits the PDCCH capacity. To solve this issue, PDCCH resource reservation for big-packet
services in heavy-load scenarios is introduced.
NOTE
PDCCH resource reservation for big-packet services in heavy-load scenarios and optimized PDCCH
aggregation level selection near the cell center are controlled by the same switch.
7.12.2.1 Benefits
This function reduces the CCE usage, increases the cell-level PDCCH capacity, and improves
the cell spectral efficiency if all of the following conditions are met:
l PDCCH resources are insufficient (ratio of uplink or downlink CCE allocation failures >
40% and equivalent CCE efficiency > 60%).
l There are abnormal UEs.
l MU beamforming pairing performance is poor or the downlink MCS index is low.
7.12.2.2 Impacts
Network Impacts
If there are UEs with reported CQI values greater than expected, such as IRC-powered UEs,
this function may affect the PDCCH reliability and decrease the UE-perceived data rate due to
lack of outer-loop CQI control.
Function Impacts
None
7.12.3 Requirements
7.12.3.1 Licenses
None
7.12.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference
7.12.3.3 Hardware
See 4.3.3 Hardware.
7.13.1 Principles
This feature enables the eNodeB to automatically optimize parameters related to massive
MIMO cells in big-event scenarios.
CellPdcchAlgo.PdcchCapacityImproveSwitch ON
CellDlschAlgo.RbgAllocStrategy ADAPTIVE
CellPdcchAlgo.CceMaxInitialRatio 10:1
CellPdcchAlgo.PdcchOptimizationSwitch ON
CellPdcchAlgo.PdcchBlerTarget 15
CellPdcchAlgo.PdcchOutLoopAdjBaseStep 1
CellPdcchAlgo.PdcchPowerEnhancedSwitch ON
CellPdcchAlgo.PdcchPwrCtrlUserNumThd 280
CellUlschAlgo.UlSchAbnUeThd 11
CellPdcchAlgo.NackDtxRatioThd 50
CellPdcchAlgo.PdcchAggLvlCLAdjustSwitch OFF
7.13.2.1 Benefits
In heavy-load scenarios, this feature improves the average downlink UE throughput, average
downlink cell throughput, and downlink cell spectral efficiency.
7.13.2.2 Impacts
NOTE
Intelligent parameter selection has no impact on the network or functions. However, some function
control switches are turned on after parameter adjustment. For details about the network and function
impacts of these functions, see the corresponding feature parameter descriptions.
Network Impacts
None
Function Impacts
None
7.13.3 Requirements
7.13.3.1 Licenses
Feature ID Feature Name Model Sales Unit
7.13.3.2 Software
Prerequisite Functions
Function Name Function Switch Reference Description
7.13.3.3 Hardware
See 4.3.3 Hardware.
8 Parameters
The following hyperlinked EXCEL files of parameter reference match the software version
with which this document is released.
l Node Parameter Reference: contains device and transport parameters.
l eNodeBFunction Parameter Reference: contains all parameters related to radio access
functions, including air interface management, access control, mobility control, and radio
resource management.
NOTE
You can find the EXCEL files of parameter reference for the software version on the live network from
the product documentation delivered with that version.
FAQ: How do I find the parameters related to a certain feature from parameter
reference?
Step 2 On the Parameter List sheet, filter the Feature ID column. Click Text Filters and choose
Contains. Enter the feature ID, for example, LOFD-001016 or TDLOFD-001016.
Step 3 Click OK. All parameters related to the feature are displayed.
----End
9 Counters
The following hyperlinked EXCEL files of performance counter reference match the software
version with which this document is released.
l Node Performance Counter Summary: contains device and transport counters.
l eNodeBFunction Performance Counter Summary: contains all counters related to radio
access functions, including air interface management, access control, mobility control,
and radio resource management.
NOTE
You can find the EXCEL files of performance counter reference for the software version used on the live
network from the product documentation delivered with that version.
FAQ: How do I find the counters related to a certain feature from performance counter
reference?
Step 2 On the Counter Summary(En) sheet, filter the Feature ID column. Click Text Filters and
choose Contains. Enter the feature ID, for example, LOFD-001016 or TDLOFD-001016.
Step 3 Click OK. All counters related to the feature are displayed.
----End
10 Glossary
11 Reference Documents
1. Beamforming (TDD)
2. Cloud BB Overview
3. CSPC
4. D-MIMO (TDD)
5. DL CoMP (TDD)
6. eMBMS
7. eMTC
8. eX2 Self-Management
9. LCS
10. Dynamic Power Sharing Between LTE Carriers
11. MIMO
12. MRC and IRC Receivers
13. License Management
14. Relay
15. SFN
16. Turbo Receiver
17. UL CoMP
18. VoLTE
19. WBB
20. WTTx Turbo Beamforming (TDD)
21. Specified User Coordinated Scheduling
22. Extended Cell Range
23. Scheduling
24. Interference Detection and Suppression
25. Interference Randomizing (TDD)
26. High Speed Mobility
27. Power Control
28. USU3900-based Multi-BBU Interconnection