Feedback for long term beamforming Document Number: C802.16m-09/1427r1 Date Submitted: 2009-07-08 Source: Intel Corporation, Yuan Zhu (yuan.y.zhu@intel.com) Qinghua Li (qinghua.li@intel.com) Alexi Davydov(alexi.davydov@intel.com) Eddie Lin (eddie.x.lin@intel.com) Huaning Niu (huaning.niu@intel.com) Guangjie Li (guangjie.li@intel.com) Yang-seok Choi (yang-seok.choi@intel.com) Hujun Yin (hujun.yin@intel.com ) Doron Ayelet (ayelet.doron@intel.com) Re: Category: AWD comments / Area: Chapter “Comments on AWD 15.3.9 UL-CTRL” 802.16m amendment working document Venue: Base Contribution: Purpose: Discussion and Approval 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 >.

Outlines Introduction Covariance matrix feedback vs. PMI feedback SLS results Conclusions Proposed text

Long term beamforming Long term beamforming Application scenario A single beamforming matrix is applied to whole band over long period. Applying same beamforming matrix over long period is already in 802.16e. Application scenario Highly correlated antenna Low mobility Feedback quantities Quantized covariance matrix Precoding matrix index (PMI) Compute PMI from covariance matrix Compute PMI from channel matrix estimated from midamble

Quantization error reduction Quantization error of PMI is reduced as codebook resolution. 6-bit codebook has smaller error than 4-bit one. Hierarchical codebook of 10-bit resolution can be formed by concatenating 6-bit base codebook and 4-bit differential codebook. Hierarchical codebook is used in same way as base codebook. The complexity of searching hierarchical codebook can be reduced by hierarchical search from base code book to differential code book.

Quantization of beaming direction Compute ideal beamforming matrix and then quantize it.

Covariance matrix vs. PMI Method I Method II Feedback infomation Quantized channel covariance matrix over PMI or MAC message. PMI over FBCH Number of feedback bits 28 bits for 4-Tx antennas, 120 bits for 8-Tx antennas 4-Tx: 6+4 = 10 bits 8-Tx: 4+4+4 = 12 bits Operations MS: measure covariance matrix, quantize principal eigenvector (for CQI estimation). MS: measure covariance matrix, search codebook.

SLS performances, SU-MIMO, AS=15º, PMI feedback Period=8Frames Method I (28 bits) Method II (6 bits) Method II (10 bits) Calibrated antenna array 100% 99.8% 100.3% Uncalibrated antenna array 95.3% 98% Uncalibrated, BS 45o cross-pole antennas 96.5% 98.7% Uncalibrated, BS cross-pole, MS V/H cross-pole 99%

SLS performances, SU-MIMO, AS=3º, PMI feedback Period=8Frames, PUSC 30km/h Method I (28 bits) Method II (6 bits) Method II (10 bits) Calibrated antenna array 100% 99.3% 100.7% Uncalibrated antenna array 93.5% 96.1% Uncalibrated, BS 45o cross-pole antennas 96.4% 98.1% Uncalibrated, BS cross-pole, MS V/H cross-pole 98.4% 99.9%

LLS performance for MU-MIMO

Conclusions Feeding back PMI and Feeding back covariance matrix have almost exactly same performance. Feed back PMI reduces overhead by 3x as compared to feeding back covariance matrix for 4-Tx BS antennas. PMI feedback should be adopted for long term beamforming.

Appendix

General SLS parameters Parameter Names Parameter Values Network Topology 57 sectors wrap around, 10 MS/sector MS Channel ITU PB3km/h or VA30km/h Frame Structure TDD, 5DL, 3 UL Feedback Delay 5ms Inter cell Interference Modeling Real Antenna Configuration 4Tx, 2 Rx, BS antenna spacing 0.5λ Codebook configuration 16m (4,1,6), 16m D(4,1,4) Tx Channel Covariance Matrix 28 bits quantized Long term feedback period 8 frames PMI error free PMI calculation Maximize V’RV System bandwidth 10MHz, 864 data subcarriers Permutation type AMC, 48 LRU or PUSC CQI feedback 1Subband=4 LRU, ideal feedback

Covariance matrix quantization method, AWD Diagonal element: 1 bit for amp {0.6, 0.9} Down triangle elements: 1 bit for amp {0.1, 0.5}, 3 bits for phase {0,π/8,π/4, 3π/8, π /2, 5π/8, 3 π /4, 7π /8}

Covariance matrix vs. measure channel Long term PMI can be computed from measured covariance matrix or channels estimated from midambles. Using covariance matrix delivers better performance for uncalibrated, cross-pole antennas.