Method and apparatus to provide low cost transmit beamforming for network devices

• US 8,515,359 B2
• Filed: 03/09/2005
• Issued: 08/20/2013
• Est. Priority Date: 03/09/2005
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Intel Corporation
• Inventors: Lin, Xintian E., Li, Qinghua
• Primary Examiner(s): AKINYEMI, AJIBOLA A
• Application Number:
  US11/076,195
• Publication Number:
  US 20060234645A1
• Time in Patent Office: 3,086 Days
• Field of Search: 455/69
• US Class Current: 455/69
• CPC Class Codes: H04B 7/0421 utilizing implicit feedback, e.g. steered pilot signals
• CPC Class Codes: H04B 7/0617 for beam forming
• CPC Class Codes: H04B 7/0619 using feedback from receiving side feedback signaling for adaptive modulation/coding H04L1/0001

Abstract

Techniques and structures for use in generating an approximated beamforming matrix in a MIMO based system are disclosed. The techniques and structures may be used to allow closed loop MIMO beamforming to be performed within a device that does not include singular value decomposition (SVD) circuitry.

Claims

1. A method comprising:

receiving a signal from a remote wireless entity via a MIMO channel, said signal including data that was multiplied by a first beamforming matrix in said remote wireless entity;

generating an approximated beamforming matrix for use in an uplink channel to said remote wireless entity, wherein generating an approximated beamforming matrix includes processing said received signal in a linear filter and a beamforming matrix estimator, wherein the linear filter generates a matrix W that is an inverse of a combined channel matrix HV, wherein H is a channel matrix of a forward channel from the remote wireless entity and V is a first beamforming matrix; and

generating transmit signals for transmission in the uplink channel to said remote wireless entity from multiple antennas using said approximated beamforming matrix.

2. The method of claim 1, wherein:

said linear filter includes a zero-forcing filter.

3. The method of claim 1, wherein:

said linear filter includes a minimum mean square error (MMSE) filter.

4. The method of claim 1, wherein:

generating an approximated beamforming matrix includes processing said inverse of said combined channel matrix using row orthogonalization, a transpose function, and a conjugate function in the beamforming matrix estimator to form said approximated beamforming matrix.

5. The method of claim 1, wherein:

said first beamforming matrix is an exact beamforming matrix.

6. The method of claim 1, wherein:

said linear filter is used in open loop MIMO receive functions.

7. A method comprising:

acquiring an exact beamforming matrix at a first wireless entity for a MIMO channel between said first wireless entity and a second wireless entity;

using said exact beamforming matrix to transmit a first frame from said first wireless entity to said second wireless entity through said MIMO channel;

receiving a response frame at said first wireless entity from the second wireless entity;

generating an approximated beamforming matrix for said MIMO channel using said response frame, wherein generating includes processing said response frame in a linear filter and a beamforming matrix estimator, wherein the linear filter generates a matrix W that is an inverse of a combined channel matrix HV by processing said received signal in said linear filter, wherein H is a channel matrix of a forward channel from the remote wireless entity and V is a first beamforming matrix, said response frame including data that was multiplied by the first beamforming matrix in the second wireless entity; and

using said approximated beamforming matrix to transmit a second frame in an uplink channel from said first wireless entity to said second wireless entity.

8. The method of claim 7, wherein:

said linear filter includes a zero-forcing filter.

9. The method of claim 7, wherein:

said linear filter includes a minimum mean square error (MMSE) filter.

10. The method of claim 7, wherein:

acquiring an exact beamforming matrix includes receiving training data at said first wireless entity, using said training data to determine a channel matrix for said MIMO channel, and performing a singular value decomposition of said channel matrix.

11. The method of claim 7, wherein:

generating an approximated beamforming matrix includes generating an inverse of a combined channel matrix by processing said received response frame in said linear filter.

12. The method of claim 11, wherein:

generating an approximated beamforming matrix includes processing said inverse of said combined channel matrix using row orthogonalization, a transpose function, and a conjugate function in the beamforming matrix estimator to form said approximated beamforming matrix.

13. The method of claim 7, wherein:

said response frame was processed by an approximated beamforming matrix for a reverse direction MIMO channel before being transmitted from said second wireless entity.

14. The method of claim 7, further comprising:

for each new response frame received at said first wireless entity from said second wireless entity, generating another approximated beamforming matrix for said MIMO channel using said new response frame, wherein generating includes processing said new response frame in a linear filter.

15. The method of claim 14, further comprising:

periodically checking for crosstalk in an output matrix of said linear filter; and

generating an exact beamforming matrix using singular value decomposition when crosstalk is detected that exceeds a predetermined level.

16. An apparatus comprising:

a linear filter to process signals received from a wireless MIMO channel, said received signal including data that was multiplied by a first beamforming matrix in a remote wireless entity, said linear filter to output a matrix W during closed loop MIMO operation, wherein the matrix W is an inverse of a combined channel matrix HV that is generated by processing said received signals in said linear filter, wherein H is a channel matrix of a forward channel from the remote wireless entity and V is the first beamforming matrix; and

a beamforming matrix estimator to generate an approximated beamforming matrix for an uplink MIMO channel using said matrix W.

17. The apparatus of claim 16, wherein:

said beamforming matrix estimator includes a row orthogonalizer to orthogonalize the rows of said matrix W, a matrix transpose function to generate a transpose of said matrix W, and a matrix conjugate function to generate a conjugate matrix of said matrix W.

18. The apparatus of claim 16, wherein:

said linear filter includes a zero-forcing filter.

19. The apparatus of claim 16, wherein:

said linear filter includes a minimum mean square error (MMSE) filter.

20. The apparatus of claim 16, wherein:

said linear filter is also operative for use during open loop MIMO operation.

21. The apparatus of claim 16, further comprising:

a transmit subsystem to generate transmit signals for transmission into said return MIMO channel via multiple antennas, said transmit subsystem to receive said approximated beamforming matrix and to use said approximated beamforming matrix to generate said transmit signals.

22. A system comprising:

a plurality of dipole antennas;

a linear filter to process signals received from a wireless MIMO channel by said plurality of dipole antennas, said received signal including data that was multiplied by a first beamforming matrix in a remote wireless entity, said linear filter to output a matrix W during closed loop MIMO operation, wherein the matrix W is an inverse of a combined channel matrix HV that is generated by processing said received signals in said linear filter, wherein H is a channel matrix of a forward channel from the remote wireless entity and V is the first beamforming matrix; and

a beamforming matrix estimator to generate an approximated beamforming matrix for an uplink MIMO channel using said matrix W.

23. The system of claim 22, wherein:

said beamforming matrix estimator includes;

a row orthogonalizer to orthogonalize the rows of said matrix W, a matrix transpose function to generate a transpose of said matrix W, and a matrix conjugate function to generate a conjugate matrix of said matrix W.

24. The system of claim 22, wherein:

said linear filter includes a zero-forcing filter.

25. The system of claim 22, wherein:

said linear filter includes a minimum mean square error (MMSE) filter.

26. An article comprising a non-transitory machine readable storage medium having instructions stored thereon that, when executed by a computing platform, operate to:

receive a signal from a remote wireless entity via a MIMO channel, said signal including data that was multiplied by a first beamforming matrix in said remote wireless entity;

generate an approximated beamforming matrix for use in an uplink channel to said remote wireless entity, wherein operation to generate an approximated beamforming matrix includes operation to cause said received signal to be processed in a linear filter and a beamforming matrix estimator, wherein the linear filter generates a matrix W that is an inverse of a combined channel matrix HV by processing said received signal in said linear filter, wherein H is a channel matrix of the forward channel from the remote wireless entity and V is the first beamforming matrix; and

generate transmit signals for transmission in the uplink channel to said remote wireless entity from multiple antennas using said approximated beamforming matrix.

27. The article of claim 26, wherein:

said linear filter includes a zero-forcing filter.

28. The article of claim 26, wherein:

operation to generate said approximated beamforming matrix includes operation to (a) orthogonalize said inverse of said combined channel matrix;

(b) generate a transpose of said inverse of said combined channel matrix; and

(c) generate a conjugate of said inverse of said combined channel matrix.