1. [Modification of CL-MIMO Codebooks]
IEEE Presentation Submission Template (Rev. 9)
Document Number: IEEE S802.16maint-08/240r2
Date Submitted:
Source:
Sangwoo Lee, Jiyun Seol, Keun Chul Hwang, Mune Joon,Inseok Hawng Voice:
Samsung Electronics
Hassan Yaghoobi
Intel Corporation
Peiying Zhu
Nortel
*<
Venue:
IEEE Session #55
Base Contribution:
Purpose:
For discussion and approval by IEEE Maintenance TG
Notice:
This document does not represent the agreed views of the IEEE 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
Patent Policy:
The contributor is familiar with the IEEE-SA Patent Policy and Procedures:
< and <
Further information is located at < and < >.

2. Purpose of this presentation
Share some views on how to reduce the codebook computational complexity.
Suggest a codebook generation rule based on current 16e codebook.

3. Problem statements Codebook computation example
In current 16e codebook for 2Tx, rank-1 codebook vectors are the first columns of rank-2 codebook matrices  MS’s codebook computation burden can be reduced by calculating only the rank-2 metric
For 4Tx codebooks, rank-1 codebook vectors are not the first columns of rank-2 codebook matrices.  MS’s should calculate the rank-1 and rank-2 metric for rank selection separately.
Rank=1
Rank=2

4. Suggestion on Codebook for Rel.1.x
For 4Tx codebook, we suggest to include the rank-1 codebook vectors in rank-2 codebook matrices.
Current 16e: The codebook generated with HE operation is composed with the last M columns, with which the rank-1 codebook vectors are excluded in rank-2 codebook matrices. Note that rank-1 codebook vectors are the first columns of H(v)
Proposed : For the codebook generated with HE operation, we propose to take the first M columns to include the rank-1 codebooks
Example: Proposed V(4,2,3)
Current 16e: V(4,2,3) = H(V(4,1,3)):,3:4
Proposed : V(4,2,3) = H(V(4,1,3)):,1:2

5. Performance Simulation
Performance of proposed codebook is the same as that of current 16e codebook.
Simulation results (4X2, perfect CE, 10dB Tx SNR)
Mean PD SINR (dB)
Current 16e
Proposed
Difference
(proposed-16e)
3bits V(4,2,3)
11.58dB
11.60dB
+0.02dB

6. Computation Complexity
Assumption: 4X2
Computational Complexity (per codeword)
In one calculation, reduction of complexity is about 31%
Rank=1
Rank=2
Current 16e
Proposed
35 CMs
24 CMs
(31% reduction)
Current 16e: 35 CMs are needed
Rank-1: 11 Complex Multiplication (CMs)= 8 CMs for Hvk + 2 CMs for || ||2 + 1 CMs for .
Rank-2: 24 CMs = 16 CMs for HVk + 6 CMs for VkHHHHVk + 2 CMs for det().
Proposed: Only 24 CMs are needed, since the rank-1 calculation of 1+  *||Hvk||2 can be obtained with (1,1) element of rank-2 matrix of I+  * VkHHHHVk.

7. Conclusions
Codebook computational complexity can be reduced by including the rank-1 codebook vectors in rank-2 codebook matrices without performance degradation.