Uplink Power Control Rule for Uplink Multi-Streams TransmissionIEEE 802.16 Presentation Submission Template (Rev. 9) Document Number: IEEE S802.16m-09/1596
Date Submitted:2009-07-08
Source: Rongzhen Yang, Apostolos Papathanassiou, E-mail: [email protected]; Wei Guan, Hujun Yin, Yang-seok Choi | [email protected] Corporation Dong-cheol Kim, Wookbong Lee, HanGuy Cho, Jin Sam Kwak [email protected] LG Electronics
Venue:.IEEE 802.16m Session#62, San Francisco, USA Category: AWD comments / Area: Chapter 15.3.9.4 (Uplink Power Control) “Comments on AWD 15.3.9.4 Uplink Power Control” 802.16m amendment working document
Base Contribution:IEEE C802.16m-09/ 1524
Purpose: The analysis to support uplink CSM power control rule in IEEE C802.16m-09/ 1524.
Notice:This document does not represent the agreed views of the IEEE 802.16 Working Group or any of its subgroups. It represents only the views of the participants listed in the “Source(s)” field above. It is offered as a basis for discussion. It is not binding on the contributor(s), who reserve(s) the right to add, amend or withdraw material contained herein.
Release:The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that
this contribution may be made public by IEEE 802.16.
Patent Policy:The contributor is familiar with the IEEE-SA Patent Policy and Procedures:
<http://standards.ieee.org/guides/bylaws/sect6-7.html#6> and <http://standards.ieee.org/guides/opman/sect6.html#6.3>.Further information is located at <http://standards.ieee.org/board/pat/pat-material.html> and <http://standards.ieee.org/board/pat >.
Background & Status
• Uplink Tx Power behavior for CSM is rarely discussed in the standard and academic study, options for Power Control PSD decision:– Tx PSD per RB?– Tx PSD per stream?– Tx PSD per MS (or Tx PSD per antenna)?
• It is very straightforward thinking: decrease the Tx power for CSM to keep nearly the same PSD and interference for each RB/sector (Tx PSD per RB).
Power Degrade for CSM (Tx PSD per RB)Pros and Cons
• Pros:– It is very straightforward to keep the IoT level of neighbor BS
• Cons:– The CSM paired AMS is penalized for power degrade.– For low interference and cover limited scenarios, or for inner users in
interference limited scenarios: the MS have much lower interference to neighbor sectors.
For example: 1. Cover Limited & Low Interference Scenarios: two MS transmit with
maximum power for SIMO, and then paired for CSM due to matched channel status, -3dB power degrade their performance and link budget a lot.
2. Inner MS in Interference Limited Scenario: two inner MS have high MCS with very low Tx power for closing to BS, - 3 dB power only degrade their performance a lot but only reduce a little bit of overall interference at neighbor sectors.
Interference Status Discussion for current OLPC algorithm
• SINRMIN is used to guarantee the celledge performance, if MS is limited by this threshold, it means that its SE is kept by using the price of high interference to other sector. Those MS is not suitable to be paired for CSM: more power on the allocated RB will get less SE gain in home sector than SE loss in neighbor sectors (by SMST theory), even not considering the co-channel interference of CSM.
• For the MS with higher SINROPT, the interference to other sectors is lower .
I.E:
Only MS have higher MCS level than SINRMIN are selected for CSM pairing.
rDL
MIN
NSIR
SINR 1),
10(^10max10log10 OPTSINR
Evaluation Setting
• Evaluation based on the high interference scenario defined by PC/LA DG (500m ISD, 3km/h eITU PedB, 1x2 SIMO/CSM, other details in backup)
• CSM Pairing Rules:– Brute Force Algorithm with perfect channel estimation
– CSM Pairing only apply the AMS with MCS level higher than 4 (decided by SINRMIN = 0 dB)
• Performance Evaluation for:– Case 1: power degrade -3 dB from SIMO to CSM
– Case 2: same power level for SIMO/CSM selection
Performance Curve Comparison
0.8 0.9 1 1.1 1.2 1.3 1.4 1.50.02
0.03
0.04
0.05
0.06
0.07
0.08
Sector SE
Cel
l edg
e S
EPerformance comparison
SIMO
CMIMO/SIMO,MCSLimit4,CMIMOBoost0dB
CMIMO/SIMO,MCSLimit4,CMIMOBoost-3dB
For Detail Comparison
CMIMO Selection Probability Comparison
0.2 0.4 0.6 0.8 1.0 0
0.05
0.1
0.15
0.2
0.25
0.3
values
CM
IMO
pro
babi
lity
CMIMO probability for different values of
CMIMO/SIMO,MCSLimit4,CMIMOBoost0dB
CMIMO/SIMO,MCSLimit4,CMIMOBoost-3dB
Detail Comparison of selected performance points
In the high interference scenario, we can see that:• “CSM -3 dB” can keep the same IoT level for the same gamma value (0.8)• “CSM -0 dB” can achieve the same IoT level and similar performance by gamma
value 0.6
Conclusion: • In high interference scenario, for the rule of “CSM -0 dB”, the IoT must to be
controlled by gamma value and CMIMO pairing rules.• “CSM -3 dB” provides the simple way to control IoT but degrades the
performance.
Scheme Gamma Value
Sector SE Celledge SE IoT Mean IoT Std
SIMO Only 0.8 1.2023 0.0404 9.9735 1.0268
CSM -3 dB 0.8 1.2583 0.0417 10.0662 1.0268
CSM -0 dB 0.6 1.2681 0.0446 10.049 1.1388
0 0.2 0.4 0.6 0.8 1.0 1.2 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
values
CM
IMO
pro
babi
lity
CMIMO probability for different values of
CMIMO/SIMO,CMIMOBoost0dB
CMIMO/SIMO,CMIMOBoost-3dB
0 500 10000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Throughput(kbps)
CD
F o
f T
hrou
ghpu
t
CDF of User Throughput
SIMO,=1.0
CMIMOBoost=0dB,=0.6
CMIMOBoost=-3dB,=1.0
Evaluation of Lower Interference Case
• One specific scenario (1x4, IMT-Adv InH) is selected for further evaluation:– Low overall interference with much higher MCS level.– Major interference is contributed by celledge MS
Scheme Gamma Value Sector SE Celledge SE IoT Mean IoT Std
SIMO Only 1.0 3.0618 0.1429 4.4131 1.5239
CSM -3 dB 1.0 4.6500 0.1847 4.8008 1.8761
CSM -0 dB 0.6 4.9391 0.2246 4.9461 1.4873
Recommendation• Because:
– Keep PSD per RB for CSM (-3 dB) is a direct and simple way to control interference, but may degrade the inner MS performance;
– Keep PSD per MS for CSM (- 0 dB) will maintain the inner MS performance but require the additional design of CSM pairing algorithm and gamma value changes that increase the complexity;
• One harmonized solution is recommended in Contribution IEEE C802.16m-09_0xxx:
– One signaling parameter in MAC power control message (PMC_RSP) from ABS indicate the power change behavior of CSM for each AMS (beta value).
Here,is set to be zero or one by one bit of MAC power control mode
signaling TNS is the Total Number of Streams in the LRU indicated by UL A-MAP IE. In
case of SU-MIMO, this value shall be set to Mt where Mt is the number of streams for one user. In case of CSM, TNS is the aggregated number of streams. In case of control channel transmission, this value shall be set to one.
)(10log10),10
)((^10max10log10 TNSSINR
dBSINRSINR DLIoT
MINopt
Backup 1:
1x2 Simulation Setting and Details of Performance Result
Uplink SLS Simulation Key Parameters(decided by PCLA DG as PC EMD)
Parameter Value Parameter Value
Carrier frequency (GHz) 2.5 GHz Site to site distance (m) 500m
System bandwidth (MHz) 10 MHz Channel eITU-Ped B, 3km/h
Reuse factor 1 Max power in MS (dBm) 23dBm
Frame duration (Preamble+DL+UL)
5ms Antenna Config 1x2 SIMO
Number of OFDM
symbols in UL Frame18 HARQ On (Max retrans: 4/Sync)
FFT size (tone) 1024 Target PER 0.2
Useful tone 864 Link to system mapping RBIR
Number of LRU 48 Scheduler type PF
LRU type DRUResource Assignment
Block8 LRU
Number of users
per sector10 Penetration loss (dB) 20dB
CMIMO support Yes Control Overhead0 for SE calculation (not
defined yet)
CMIMO/SIMO Switch - CMIMO boost -3dB
- Harq is CMIMO - MCS Limitation 4
CMIMO Probability
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0
0.05
0.1
0.15
0.2
0.25
values
CM
IMO
pro
babi
lity
CMIMO probability for different values of
Throughput performance
Gamma
Values
Sector throughput
(in Mbps)
Cell-edge throughput
(in Kbps)Sector SE Cell-Edge SE
0 2.7081 253.44 0.7222 0.0676
0.2 3.2115 253.5304 0.8564 0.0676
0.4 3.8607 253.7111 1.0295 0.0677
0.6 4.3471 229.0447 1.1592 0.0611
0.8 4.7186 156.2744 1.2583 0.0417
1.0 5.0281 125.2292 1.3408 0.0334
1.2 5.2014 97.6715 1.387 0.026
1.4 5.2758 70.9813 1.4069 0.0189
1.6
User throughput CDF
0 500 1000 1500 2000 25000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
User throughput (in kbps)
CD
FUser throughput distribution for different values of
=0=0.2=0.4=0.6=0.8=1.0=1.2=1.4
IoT CDF
Gamma
Values
Mean IoT
(in dB)
IoT Std
(in dB)
0 6.3425 1.0864
0.2 6.4624 1.1127
0.4 7.1782 1.0856
0.6 8.495 1.0093
0.8 10.0662 1.0268
1.0 11.483 1.065
1.2 13.0718 1.0779
1.4 14.3893 1.1223
1.6
2 4 6 8 10 12 14 16 18 200
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
IoT (in dB)
CD
F
IoT Distribution for different values of
=0
=0.2
=0.4
=0.6
=0.8
=1.0
=1.2
=1.4
MCS distribution
1 2 3 4 5 6 7 8 9 10 11 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
MCS index
Pro
babi
lity
MCS Distribution for different values of
=0=0.2=0.4=0.6=0.8=1.0=1.2=1.4
FER distribution
1 2 3 4 5 6 7 8 9 10 11 0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
MCS index
FE
RFER Distribution for different values of
=0=0.2=0.4=0.6=0.8=1.0=1.2=1.4
CMIMO/SIMO Switch - CMIMO boost 0dB
- Harq is CMIMO - MCS Limitation 4
CMIMO Probability
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
values
CM
IMO
pro
babi
lity
CMIMO probability for different values of
Throughput performance
Gamma
Values
Sector throughput
(in Mbps)
Cell-edge throughput
(in Kbps)Sector SE Cell-Edge SE
0
0.2 3.3664 263.4692 0.8977 0.0703
0.4 4.1204 245.254 1.0988 0.0654
0.6 4.7554 167.1891 1.2681 0.0446
0.8 5.1974 106.6888 1.386 0.0285
1.0 5.2895 78.58 1.4105 0.021
1.2 5.343 52.8565 1.4248 0.0141
1.4 5.3464 45.7547 1.4257 0.0122
1.6 5.2044 43.8935 1.3878 0.0117
User throughput CDF
0 500 1000 1500 2000 2500 3000 3500 40000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
User throughput (in kbps)
CD
FUser throughput distribution for different values of
=0.2=0.4=0.6=0.8=1.0=1.2=1.4=1.6
IoT CDF
Gamma
Values
Mean IoT
(in dB)
IoT Std
(in dB)
0
0.2 6.6655 0.9867
0.4 8.2743 1.0573
0.6 10.049 1.1388
0.8 12.7279 1.1986
1.0 14.8811 1.2758
1.2 16.6125 1.3342
1.4 18.1079 1.3408
1.6 18.967 1.4127
0 5 10 15 20 25 300
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
IoT (in dB)
CD
F
IoT Distribution for different values of
=0.2=0.4=0.6=0.8=1.0=1.2=1.4=1.6
MCS distribution
1 2 3 4 5 6 7 8 9 10 11 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
MCS index
Pro
babi
lity
MCS Distribution for different values of
=0.2=0.4=0.6=0.8=1.0=1.2=1.4=1.6
FER distribution
1 2 3 4 5 6 7 8 9 10 11 0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
MCS index
FE
RFER Distribution for different values of
=0.2=0.4=0.6=0.8=1.0=1.2=1.4=1.6
Backup 2:
Theory Analysis of SMST Algorithm for CSM
Maximum Sector Throughput Method Derivation (1)Initial Modeling
For one MS:
• Channel Loss: the channel loss (include pathloss and fading) form MS to strong BSs can be measured by preamble signal strength, top N: CL0, CL1, CL2, .. CLN, CL0 is the channel loss to home sector.
• NI (Noise plus Interference): Home and neighbor sectors information are modeled as (NI0, NI1, NI2, …, NIN)
When one MS increases Tx PSD, it will bring SE gain and cause SE loss to neighbor sectors (SU-SISO):
)1
1log(
)1log()1log(
Orig
New
OrigNewgain
SINR
SINR
SINRSINRSE
)1log()1log(
)1
1log()
)(1
)(1log(
))(1log())(1log()(
ii
i
i
i
ii
i
i
i
New
Orig
NewOrigloss
NIS
I
NI
I
INIS
NIS
iSINR
iSINR
iSINRiSINRiSE
PSDPSDPSDNI
CLPSDSINR NewOrig 0
0
00 ,/
)1log(
0
00
0
CLPSD
NI
CLPSD
SEgain
Then, total SE loss is:
N
ilossloss iSESE
1
)(
ii CL
PI
In theory, the Maximum Sector Throughput will be got when
0, PSDSESE lossgain
Maximum Sector Throughput Method Derivation (2)Simplification Assumptions
• One virtual neighbor sector: the channel loss to all neighbor sectors are difficult to be accurately measured in real environment, we assume one virtual neighbor sector that accounts for accumulated downlink interferences
Ni i
eambleT
I
eambleTIDL CL
P
CL
PP
~1
Pr,Pr,,
1
~1
1
Ni iI CL
CL
IDL
HDLDL P
PSIR
,
,H
IDL CL
CLSIR
Then, we can get the Maximum Sector Throughput derivation for SU-SISO system:
11
1 DLarg
SIR
SINRNI
NISINR
IH
IetT
1DLarg SIRSINR etT
Gamma is used as the control factor to control the interference to other sectors
Maximum Sector Throughput Method Derivation (3)SU-MIMO Consideration
Nr receive antenna for ABS:
,
,
,
,,
1log
1
( )1
log log 11
New MRCGain
Orig MRC
r tx tx r tx
H H Ant H
r tx r txH Ant
H H Ant H
SINRSE
SINR
N PSD PSD N PSDCL NI CL
N PSD N PSDNI
CL NI CL
, ,
, ,
, ,,
,
,
, ,
,
11
log log1
1/
log 1
1
r I Ant Noise AntIOrig MRC I AntI
Loss Ir I Ant Noise AntNew MRC
I Ant tx I
tx
I I Ant
r I Ant Noise Ant
I Ant tx
N SNR P
SINR NISE
N SNR PSINRNI PSD CL
PSD
CL NI
N SNR P
NI PSD
, ,
, //
r I Ant Noise Ant
I Ant tx I
I
N SNR P
NI PSD CLCL
rAntIrAntH
AntI
AntHHAntI
AntIrAntII
rAntHH
AnttxAntH
HAntHAntI
AntIrAntII
rAnttx
r
HAntH
AntI
AntI
H
IAntIrAnttx
AntI
AntI
H
IAntIr
H
AnttxrAntH
AntI
AntI
AntIr
IH
H
AnttxrAntH
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
H
AnttxrAntH
H
Anttxr
AnttxILossGain
NSIR
SINRNNI
NI
NICLSINR
SINRNNICL
NNICL
PSDSINR
CLNISINR
SINRNNICL
NPSD
N
CLNI
SINR
NI
CL
CLSINRNPSD
SINR
NI
CL
CLSINRN
CL
PSDNNI
NI
SINR
SINRN
CLCL
CL
PSDNNI
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CL
PSDNNI
CL
PSDN
PSDSESESE
1*
*
11*
1**
*1***
1
*
**1**
*1
****1
***1*
*1
/*
1
/
**
/
**1
**
*
)0(0
DL,,
,
,,
,,
,
,,
,,
,,,
,,
,,,
,
,,
,,
,
,
,,,
,,
,,
,,
,,
,
,
,,
,
,
retT N
SIRSINR1
DLarg
Gamma is used as control factor to control the interference to other sectors
Maximum Sector Throughput Method Derivation (4)MU-MIMO Consideration
Four study cases for consideration: • Home sector SU and virtual neighbor sector SU
(already done for SU-MIMO)• Home sector MU and virtual neighbor sector SU • Home sector SU and virtual neighbor sector MU • Home sector MU and virtual neighbor sector MU
Maximum Sector Throughput Method Derivation (5) Home sector MU and virtual neighbor sector SU
H
AnttxrAntH
H
Anttxr
H
AnttxrAntH
H
Anttxr
MRCOrig
MRCNew
MRCOrig
MRCNewGain
CL
PSDNNI
CL
PSDN
CL
PSDNNI
CL
PSDN
SINR
SINR
SINR
SINRSE
,,
,
2
,,
,
2
,
,
,
,
*
**2
1log*
*
1log
1
1log
1
1log*2
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
IAnttxAntI
AntNoiseAntIr
AntI
AntNoiseAntIr
IMRCNew
IMRCOrigI
Loss
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CLPSDNI
PSNRNNI
PSNRN
SINR
SINRSE
/*2
**
/*2
**1
*
*2
1log
/*2
**1
**1
log1
1log
,,
,,
,,
,,
,
,
,,
,,
,
,,
,
,
rAntIrAntH
AntIAntH
r
HAntH
AntI
AntI
H
IAntIrAnttx
AntI
AntI
AntIr
IH
H
AnttxrAntH
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
H
AnttxrAntH
H
Anttxr
AnttxILossGain
NSIR
SINRNNI
NISINR
N
CLNI
SINR
NI
CL
CLSINRNPSD
NI
SINR
SINRN
CLCL
CL
PSDNNI
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CL
PSDNNI
CL
PSDN
PSDSESESE
1*
*
11*
****1
*1
/*
1
/*2
**
/*2
**1
*
*2
*
**2
)0(0
DL,,
,,
,,
,,,
,
,
,,,
,,
,,
,,
,,
,
,
,,
,
,
Maximum Sector Throughput Method Derivation (6) Home sector SU and virtual neighbor sector MU
H
AnttxrAntH
H
Anttxr
MRCOrig
MRCNew
MRCOrig
MRCNewGain
CL
PSDNNI
CL
PSDN
SINR
SINR
SINR
SINRSE
,,
,
,
,
,
,
*
*
1log
1
1log
1
1log
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
IMRCNew
IMRCOrigI
Loss
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
SINR
SINRSE
/
**
/
**1
**2
1log
/
**
/
**1
*1log
1
1log*2
,,
,,
,,
,,
,
,
2
,,
,,
,,
,,
,
,
,
,
AntI
AntI
AntIr
IH
H
AnttxrAntH
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
H
AnttxrAntH
H
Anttxr
AnttxILossGain
NI
SINR
SINRN
CLCL
CL
PSDNNI
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CL
PSDNNI
CL
PSDN
PSDSESESE
,
,
,,,
,,
,,
,,
,,
,
,
,,
,
,
*1
2//*
1
/
**
/
**1
**2
*
*
)0(0
Define , then2/II CLCL
rAntIrAntH
AntIAntH
HAntHAntI
AntIrAntII
rAnttx
HAntHAntI
AntIrAntII
rAnttx
NSIR
SINRNNI
NISINR
CLNISINR
SINRNNICL
NPSD
CLNISINR
SINRNNICL
NPSD
1*
*
11**
2
1
**1**
*2
1*
1
**1**
*1
DL,,
,,
,,
,,,
,,
,,,
Maximum Sector Throughput Method Derivation (7) Home sector MU and virtual neighbor sector MU
Define , then2/II CLCL
H
AnttxrAntH
H
Anttxr
H
AnttxrAntH
H
Anttxr
MRCOrig
MRCNew
MRCOrig
MRCNewGain
CL
PSDNNI
CL
PSDN
CL
PSDNNI
CL
PSDN
SINR
SINR
SINR
SINRSE
,,
,
2
,,
,
2
,
,
,
,
*
**2
1log*
*
1log
1
1log
1
1log*2
IAnttxAntI
AntNoiseAntIr
ItxAntI
AntNoiseAntIr
AntII
tx
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
IMRCNew
IMRCOrigI
Loss
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
SINR
SINRSE
/*2
**
/*2
**1
**4
1log
/*2
**
/*2
**1
**2
1log1
1log*2
,,
,,
,
,,
,
2
,,
,,
,,
,,
,
,
,
,
AntI
AntI
AntIr
IH
H
AnttxrAntH
IAnttxAntI
AntNoiseAntIr
IAnttxAntI
AntNoiseAntIr
AntII
Anttx
H
AnttxrAntH
H
Anttxr
AnttxILossGain
NI
SINR
SINRN
CLCL
CL
PSDNNI
CLPSDNI
PSNRN
CLPSDNI
PSNRNNICL
PSD
CL
PSDNNI
CL
PSDN
PSDSESESE
,
,
,,,
,,
,,
,,
,,
,
,
,,
,
,
*1
2//*
1
/*2
**
/*2
**1
*
*4
*
**2
)0(0
rAntIrAntH
AntIAntH
HAntHAntI
AntIrAntII
rAnttx
HAntHAntI
AntIrAntII
rAnttx
NSIR
SINRNNI
NISINR
CLNISINR
SINRNNICL
NPSD
CLNISINR
SINRNNICL
NPSD
1*
*
11**
2
1
**1**
*2
1*
1
**1**
*1
DL,,
,,
,,
,,,
,,
,,,
Maximum Sector Throughput Method Derivation (8)MU-MIMO Study Summary
rAntIrAntH
AntIAntH N
SIRSINRNNI
NISINR
1*
*
11* DL
,,
,,
Virtual neighbor Sector
SU:
Virtual neighbor Sector MU:rAntIrAntH
AntIAntH N
SIRSINRNNI
NISINR
1*
*
11**
2
1DL
,,
,,
Conclusions:• For MU-MIMO, same power control formula can be applied
• When AMS, MS perform MU/SU switching, the Tx power should not be changed
• The gamma is used to control interference, if neighbor sectors have higher percentage of MU selection, the gamma value can be decreased to reduce interference
retT N
SIRSINR1
DLarg