Beamformed space time code communication with testing spatial signature generation
An apparatus, logic and method are provided to improve beamformed space time code (STC) wireless communication. A first device comprising a plurality of antennas receives signals at the plurality of antennas transmitted from a first antenna of a second device. A testing spatial signature for a second antenna of the second device is computed based on the signals received at the plurality of antennas of the first device from the first antenna of the second device. Using the testing spatial signature and the signals received at the plurality of antennas of the first device from the first antenna of the second device, beamforming weights are computed to be applied to a space time code signal to be transmitted from the first device to the second device via the plurality of antennas of the first device.
1. A method comprising:
at a first device comprising a plurality of antennas, receiving signals transmitted from a first antenna of a second device;
generating a testing spatial signature covariance matrix for a second antenna of the second device based on the signals received at the plurality of antennas of the first device from the first antenna of the second device; and
based on the testing spatial signature covariance matrix and the signals received at the plurality of antennas of the first device from the first antenna of the second device, computing beamforming weights to be applied to a space time code signal to be transmitted from the first device to the second device via the plurality of antennas of the first device.
2. The method of claim 1 , and further comprising transmitting the space time code signal using the beamforming weights from the first device to the second device.
3. The method of claim 1 , wherein receiving comprises receiving signals at the plurality of antennas of the first device from the first antenna of the second device over a time period, and further comprising storing data representing the signals received at each of the plurality of antennas at a plurality of observation times during the time period.
4. The method of claim 3 , wherein storing comprises storing data for a matrix R 1 from the data representing the signals received at the plurality of antennas of the first device from the first antenna of the second device, where R 1 =[r 11 , r 12 , r 13 , . . . r 1L ], and r 12 , . . . , r 1L are L observations of signals received at the plurality of antennas of the first device from the first antenna of the second device, where r 11 is the earliest observation and r 1L is the latest observation during the time period, and wherein generating comprises computing the testing spatial signature covariance matrix R 2 from the matrix R 1 .
5. The method of claim 4 , wherein computing comprises computing the testing spatial signature covariance matrix R 2 from r 11 , r 12 , . . . , r 1L which are time signal observation vectors of matrix R 1 .
6. The method of claim 5 , wherein computing the testing spatial signature is based on less than all of the L observations.
7. The method of claim 6 , wherein computing comprises computing an intermediate testing spatial signature vector r 2 from the latest time signal observation vector r 1L , and computing the testing spatial signature covariance matrix R 2 from the intermediate testing spatial signature vector r 2 .
8. The method of claim 4 , wherein computing the beamforming weights comprises computing a first beamforming vector W 1 for use in transmissions during odd time slots of the space time code signal and a second beamforming vector W 2 for use in transmissions during even time slots of the space time code signal.
9. The method of claim 8 , wherein computing the beamforming weights comprises computing the first beamforming vector W i that is an eigenvector that corresponds to a largest eigenvalue of the matrix R 1 , and computing the second beamforming vector W 2 that is an eigenvector that corresponds to a largest eigenvalue of the testing spatial signature covariance matrix R 2 .
10. The method of claim 8 , wherein computing the beamforming weights comprises adding the matrix R 1 to the testing spatial signature covariance matrix R 2 , and computing the first beamforming vector W i that corresponds to a first dominant eigenvector of a sum matrix resulting from the addition of the matrices R 1 and R 2 and computing the second beamforming vector W 2 that corresponds to a second dominant eigenvector of the sum matrix resulting from the addition of the matrices R 1 and R 2 .
11. The method of claim 8 , and further comprising partitioning power among the plurality of antennas of the first device by normalizing the first beamforming vector W 1 and the second beamforming vector W 2 such that:
W 1 =√{square root over (2)} A·W 1
W 2 =√{square root over (2(1 −A 2 ))}· W 2 , where A is a value between zero and 1.
12. An apparatus comprising:
a plurality of antennas;
a transmitter coupled to the plurality of antennas, the transmitter that is configured to apply beamforming weights to a space time code signal to be beamformed and transmitted from the plurality of antennas to a destination device; and
a controller coupled to the beamforming module that is configured to compute the beamforming weights used by the beamforming module by generating a testing spatial signature covariance matrix for a second antenna of the destination device based on the signals received at the plurality of antennas from a first antenna of the destination device, and to compute the beamforming weights based on the testing spatial signature covariance matrix and the signals received at the plurality of antennas from the first antenna of the destination device.
13. The apparatus of claim 12 , wherein the controller stores data representing signals received at each of the plurality of antennas from the first antenna of the destination device over a time period, wherein the data represents signals received at each of the plurality of antennas at a plurality of observation times during the time period.
14. The apparatus of claim 13 , wherein the controller stores data for a matrix R 1 from the data representing the signals received at the plurality of antennas from the first antenna of the destination device, where R 1 =[r 11 , r 12 , r 13 , . . . r 1L ] and r 11 , r 12 , . . . , r 1L are L observations of signals received at the plurality of antennas of the first device from the first antenna of the destination device, where r 11 is the earliest observation and r 1L is the latest observation during the time period, and computes the testing spatial signature covariance matrix R 2 from the matrix R 1 .
15. The apparatus of claim 14 , wherein the controller computes the testing spatial signature covariance matrix R 2 from r 11 , r 12 , . . . , r 1L which are time signal observation vectors of matrix R 1 .
16. The apparatus of claim 14 , wherein the controller computes the testing spatial signature covariance matrix R 2 based on less than all of the L observations.
17. The apparatus of claim 16 , wherein the controller computes the testing spatial signature covariance matrix R 2 by computing an intermediate testing spatial vector r 2 from the latest time signal observation vector r 1L and computing the testing spatial signature covariance matrix R 2 from the intermediate testing spatial signature vector r 2 .
18. The apparatus of claim 14 , wherein the controller computes a first beamforming vector for use in transmissions during odd time slots of the space time code signal and a second beamforming vector for use in transmissions during even time slots of the space time code signal.
19. The apparatus of claim 18 , wherein the controller computes the first beamforming vector that is an eigenvector that corresponds to a largest eigenvalue of the matrix R 1 , and computes the second beamforming vector that is an eigenvector that corresponds to a largest eigenvalue of the testing spatial signature covariance matrix R 2 .
20. The apparatus of claim 18 , wherein the controller computes the beamforming weights by adding the matrix R 1 to the testing spatial signature covariance matrix R 2 , and computing the first beamforming vector that corresponds to a first dominant eigenvector of a sum matrix resulting from the addition of the matrices R 1 and R 2 and computing the second beamforming vector that corresponds to a second dominant eigenvector of the sum matrix resulting from the addition of the matrices R 1 and R 2 .
21. The apparatus of claim 18 , wherein the controller computes the first and second beamforming vectors W 1 and W 2 so as to partition power among the plurality of antennas by normalizing the first and second beamforming vectors W 1 and W 2 such that:
W 1 =√{square root over (2)} A·W 1
W 2 =√{square root over (2(1 −A 2 ))}· W 2 , where A is a value between zero and 1.
22. A tangible media encoded with instructions for execution by a processor, and when executed operable to cause the processor to:
at a first device comprising a plurality of antennas, generate a testing spatial signature covariance matrix for a second antenna of a second device based on signals received at the plurality of antennas of the first device from a first antenna of the second device; and
based on the testing spatial signature covariance matrix and the signals received at the plurality of antennas of the first device from the first antenna of the second device, compute beamforming weights to be applied to a space time code signal to be transmitted from the first device to the second device via the plurality of antennas of the first device.
23. The tangible media of claim 22 , and further comprising instructions that cause the processor to store data for a matrix R 1 from data representing the signals received at the plurality of antennas of the first device from the first antenna of the second device, where R 1 =[r 11 , r 12 , r 13 , . . . , r 1L ], and r 11 , r 12 , . . . , r 1L are L observations of signals received at the plurality of antennas of the first device from the first antenna of the second device, where r 1,1 is the earliest observation and r 1L is the latest observation during the time period, and wherein the logic for computing the testing spatial signature comprises logic for computing the testing spatial signal covariance matrix R 2 from the matrix R 1 .
24. The tangible media of claim 23 , wherein the instructions that cause the processor to compute the testing spatial signature covariance matrix R 2 comprise instructions that cause the processor to compute the matrix R 2 from r 11 , r 12 , . . . , r 1L which are time signal observation vectors of matrix R 1 .
25. The tangible media of claim 23 , wherein the instructions that cause the processor to compute the testing spatial signature covariance matrix R 2 comprise instructions that cause the processor to compute the testing spatial signature covariance matrix R 2 based on less than all of the L observations.
26. The tangible media of claim 25 , wherein the instructions that cause the processor to compute the testing spatial signature covariance matrix R 2 comprise instructions that cause the processor to compute an intermediate testing spatial vector r 2 from the latest time signal observation vector r 1L , and to compute the testing spatial signature covariance matrix R 2 from the intermediate testing spatial signature vector r 2 .
27. The tangible media of claim 23 , wherein the instructions that cause the processor to compute the beamforming weights comprise instructions that cause the processor to compute a first beamforming vector W 1 for use in transmissions during odd time slots of the space time code signal and a second beamforming vector W 2 for use in transmissions during even time slots of the space time code signal.
28. The tangible media of claim 27 , wherein the instructions that cause the processor to compute the beamforming weights comprise instructions that cause the processor to compute the first beamforming vector W 1 that is an eigenvector that corresponds to a largest eigenvalue of the matrix R 1 , and to compute the second beamforming vector W 2 that is an eigenvector that corresponds to a largest eigenvalue of the testing spatial signature covariance matrix R 2 .
29. The tangible media of claim 27 , wherein the instructions that cause the processor to compute the beamforming weights comprise instructions that cause the processor to add the matrix R 1 to the testing spatial signature covariance matrix R 2 , and computes the first beamforming vector W 1 that corresponds to a first dominant eigenvector of a sum matrix resulting from the addition of the matrices R 1 and R 2 and computes the second beamforming vector W 2 that corresponds to a second dominant eigenvector of the sum matrix resulting from the addition of the matrices R 1 and R 2 .
30. The tangible media of claim 27 , and further comprising instructions that cause the processor to partition power among the plurality of antennas of the first device by normalizing the first beamforming vector W 1 and the second beamforming vector W 2 such that:
W 1 =√{square root over (2)} A·W 1
W 2 =√{square root over (2(1 −A 2 ))}· W 2
where A is a value between zero and 1.