METHOD AND APPARATUS FOR PRE-CODING FOR A MIMO SYSTEM

• US 20070165738A1
• Filed: 10/25/2006
• Published: 07/19/2007
• Est. Priority Date: 10/27/2005
• Status: Abandoned Application

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Gorokhov, Alexei, Wang, Jibing, Sampath, Hemanth, Kadous, Tamer, Barriac, Gwendolyn
• Application Number:
  US11/552,948
• Publication Number:
  US 20070165738A1
• Time in Patent Office: Days
• US Class Current: 375/267
• CPC Class Codes: H04L 1/0687 Full feedback

Abstract

Systems and methodologies are described that facilitates computing a precoding index which correlates to a precoding matrix within a codebook. According to various aspects, systems and/or methods are described that facilitate computing an effective signal-to-noise ratio (SNR). Such systems and/or methods may further facilitate selecting a precoding matrix and a corresponding precoding index. Such systems and/or methods may still further facilitate employing the precoding matrix in a MIMO wireless communication system.

Claims

1. A method that facilitates computing a precoding index in a wireless communication environment, comprising:

utilizing a per-tile feedback scheme for MIMO precoding;

computing an effective signal-to-noise ratio (SNR) for a precoding matrix and a tile;

selecting the precoding matrix yielding the highest effective SNR; and

employing the precoding matrix and corresponding precoding index in the MIMO wireless communication environment.

2. The method of claim 1, further comprising a codebook related to C={F_j}_j=1^N, where C denotes the codebook, F_j is a matrix within the codebook, and N is an integer of matrices included within the codebook.

3. The method of claim 1, further comprising calculating the precoding index for each tile within the per-tile feedback scheme.

4. The method of claim 3, further comprising a channel matrix that denotes disparate tiles as H_f,1, H_f,2, . . . , H_f,M, where M is a number of tiles in a current assignment and f represents frequency.

5. The method of claim 4, further comprising employing the following metric to select the precoding matrix:

for the i-th tile H_f,i, compute max [trace (F_j^HH^Hf,iH_f,iF_j)].

6. The method of claim 1, further comprising:

computing a post processing SNR; and

converting the post processing SNR to at least one of a constrained capacity with a gap to capacity and an unconstrained capacity with a gap to capacity.

7. The method of claim 1, further comprising:

partitioning a codebook into at least two or more subsets;

partitioning the subset of matrices based at least in part upon distance; and

employing an exhaustive search on a selected subset with the largest signal-to-noise ratio (SNR).

8. A method that facilitates computing a precoding index in a wireless communication environment, comprising:

utilizing an average feedback scheme for MIMO precoding;

computing an average effective signal-to-noise ratio (SNR) for a precoding matrix;

obtaining an averaged channel covariance matrix; and

selecting a precoding matrix from a codebook utilizing at least one of the averaged effective SNR and the averaged channel covariance matrix.

9. The method of claim 8, further comprising a codebook related to C={F_j}_j=1^N, where C denotes the codebook, F_j is a matrix within the codebook, and N is an integer of matrices included within the codebook.

10. The method of claim 8, further comprising computing the average effective signal-to-noise ratio (SNR) that is averaged over at least one of the following:

1) the entire assignment;

     2) at least one tile of the assignment; and

     3) a portion of the bandwidth that is not dependent upon the assignment.

11. The method of claim 8, further comprising sampling at least one of a tile of the assignment and the entire bandwidth to compute the effective SNR.

12. The method of claim 8, further comprising utilizing the following to compute the averaged channel covariance matrix:

R=E(H^HH), where R is the averaged channel covariance matrix.

13. The method of claim 12, further comprising selecting the codebook with at least one of the following:

1) max [trace(F_j^HRF_j)];

     2) max [log det(I+ρ

F_j^HRF_j)], where ρ

is the average SNR; and

     3) maximize the effective SNR by substituting R into a post processing SNR computation.

14. The method of claim 8, further comprising:

partitioning the codebook into at least two or more subsets;

partitioning the subset of matrices based at least in part upon distance; and

employing an exhaustive search on a selected subset with the largest signal-to-noise ratio (SNR).

15. A communication apparatus, comprising:

a memory that retains instructions related to computing a precoding index by calculating an effective SNR for at least one of a per-tile feedback scheme and an average feedback scheme; and

a processor, coupled to memory, configured to evaluate the instructions to employ the precoding index utilizing at least one algorithm, the precoding index correlates to a matrix within a codebook.

16. The communication apparatus of claim 15, further comprising the codebook is related to C={F_j}_j=1^N, where C denotes the codebook, F_j is a matrix within the codebook, and N is an integer of matrices included within the codebook.

17. The communication apparatus of claim 16, further comprising calculating the precoding index for each tile within the per-tile feedback scheme.

18. The communication apparatus of claim 17, further comprising a channel matrix that denotes disparate tiles as H_f,1, H_f,2, . . . , H_f,M, where M is a number of tiles in a current assignment.

19. The communication apparatus of claim 18, further comprising employing the following metric to select the precoding matrix:

for the i-th tile H_f,i, compute max [trace(F_j^HH^Hf,iH_f,iF_j)].

20. The communication apparatus of claim 19, further comprising:

computing a post processing SNR; and

converting the post processing SNR to at least one of a constrained capacity with a gap to capacity and an unconstrained capacity with a gap to capacity

21. The communication apparatus of claim 20, further comprising computing the average effective signal-to-noise ratio (SNR) that is averaged over at least one of the following:

1) the entire assignment;

     2) at least one tile of the assignment; and

     3) a portion of the bandwidth that is not dependent upon the assignment.

22. The communication apparatus of claim 21, further comprising sampling at least one of a tile of the assignment and the entire bandwidth to compute the effective SNR.

23. The communication apparatus of claim 22, further comprising utilizing the following to compute the averaged channel covariance matrix:

R=E(H^HH), where R is the averaged channel covariance matrix.

24. The communication apparatus of claim 23, further comprising selecting the codebook with at least one of the following:

1) max [trace(F_j^HRF_j)];

     2) max [log det (I+ρ

F_j^HRF_j)], where ρ

is the average SNR; and

     3) maximize the effective SNR by substituting R into a post processing SNR computation.

25. The communication apparatus of claim 15, further comprising

partitioning the codebook into at least two or more subsets;

partitioning the subset of matrices based at least in part upon distance; and

employing an exhaustive search on a selected subset with the largest signal-to-noise ratio (SNR).

26. A communication apparatus that facilitates computing a precoding index, comprising:

means for computing an effective signal-to-noise ratio (SNR);

means for selecting a precoding matrix and a corresponding precoding index; and

means for employing the precoding matrix in a MIMO wireless communication system.

27. The communication apparatus of claim 26, further comprising means for computing the average effective signal-to-noise ratio (SNR) that is averaged over at least one of the following:

1) the entire assignment;

     2) at least one tile of the assignment; and

     3) a portion of the bandwidth that is not dependent upon the assignment.

28. The communication apparatus of claim 27, further comprising means for sampling at least one of a tile of the assignment and the entire bandwidth to compute the effective SNR.

29. The communication apparatus of claim 28, further comprising means for calculating an averaged channel covariance matrix with the following:

R=E(H^HH), where R is the averaged channel covariance matrix.

30. The communication apparatus of claim 29, further comprising means for selecting a codebook with at least one of the following:

1) max [trace(F_j^HRF_j)];

     2) max [log det (I+ρ

F_j^HRF_j)], where ρ

is the average SNR; and

     3) maximize the effective SNR by substituting R into a post processing SNR computation.

31. The communication apparatus of claim 26, further comprising a codebook that is related to C={F_j}_j=1^N, where C denotes the codebook, F_j is a matrix within the codebook, and N is an integer of matrices included within the codebook.

32. The communication apparatus of claim 31, further comprising means for calculating the precoding index for each tile within a per-tile feedback scheme.

33. The communication apparatus of claim 32, further comprising a channel matrix that denotes disparate tiles as H_f,1, H_f,2, . . . , H_f,M, where M is a number of tiles in a current assignment.

34. The communication apparatus of claim 33, further comprising means for employing the following metric to select the precoding matrix:

for the i-th tile H_f,i, computer max [trace(F_j^HH^Hf,iH_f,iF_j)].

35. The communication apparatus of claim 26, further comprising:

means for partitioning a codebook into at least two or more subsets;

means for partitioning the subset of matrices based at least in part upon distance; and

means for employing an exhaustive search on a selected subset with the largest signal-to-noise ratio (SNR).

36. A machine-readable medium having stored thereon machine-executable instructions for:

computing an effective signal-to-noise ratio (SNR);

selecting a precoding matrix and a corresponding precoding index; and

employing the precoding matrix in a MIMO wireless communication system.

37. The machine-readable medium of claim 36, further comprising computing an average effective signal-to-noise ratio (SNR) that is averaged over at least one of the following:

1) the entire assignment;

     2) at least one tile of the assignment; and

     3) a portion of the bandwidth that is not dependent upon the assignment.

38. The machine-readable medium of claim 37, further comprising sampling at least one of a tile of the assignment and the entire bandwidth to compute the effective SNR.

39. The machine-readable medium of claim 38, further comprising calculating an averaged channel covariance matrix with the following:

R=E(H^HH), where R is the averaged channel covariance matrix.

40. The machine-readable medium of claim 39, further comprising selecting a codebook with at least one of the following:

1) max [trace(F_j^HRF_j)];

     2) max [log det(I+ρ

F_j^HRF_j)], where ρ

is the average SNR; and

     3) maximize the effective SNR by substituting R into a post processing SNR computation.

41. The machine-readable medium of claim 36, further comprising a codebook that is related to C={F_j}_j=1^N, where C denotes the codebook, F_j is a matrix within the codebook, and N is an integer of matrices included within the codebook.

42. The machine-readable medium of claim 41, further comprising calculating the precoding index for each tile within a per-tile feedback scheme.

43. The machine-readable medium of claim 42, further comprising a channel matrix that denotes disparate tiles as H_f,1, H_f,2, . . . , H_f,M, where M is a number of tiles in a current assignment.

44. The machine-readable medium of claim 43, further comprising employing the following metric to select the precoding matrix:

for the i-th tile H_f,i, compute max [trace(F_j^HH^Hf,iH_f,iF_j)].

45. In a wireless communication system, an apparatus, comprising:

a processor configured to;

ascertain to employ at least one of a per-tile feedback scheme and an average feedback scheme;

select a precoding matrix and a corresponding precoding index; and

employ the precoding matrix in a MIMO wireless communication system.