Robust transceiver design

1. A method, comprising:

receiving, at a first antenna, estimated channel state information (CSI) for at least one channel between the first antenna and at least one second antenna;

obtaining antenna correlation information associated with the first antenna and the at least one second antenna, respectively;

calculating a set of noise terms associated with random noise for the at least one channel; and

optimizing a configuration of beamformers for the first antenna and the at least one second antenna that maximizes a throughput of all antennas, in accordance with the estimated CSI, the obtained antenna correlation information, and the noise terms.

2. The method of claim 1 , wherein the first antenna is a transmitting antenna, and the at least one second antenna is a receiving antenna.

3. The method of claim 1 , wherein the optimizing includes determining the configuration of the beamformers that maximizes a weighted sum rate of all transmitting-receiving antenna pairs.

4. The method of claim 3 , wherein the optimizing includes determining the configuration of the beamformers that maximizes a weighted sum of lower bounds of mutual information between transmitted and received signals.

5. The method of claim 3 , wherein the optimizing includes determining the configuration of the beamformers that minimizes a weighted sum of mean-square-errors (MSEs) of estimated data streams received at the at least one second antenna, and wherein an MSE of a respective estimated data stream is associated with a corresponding transmit beamformer, a corresponding receive beamformer, the estimated CSI, and the random noise.

6. The method of claim 5 , wherein the optimizing involves a coordinate descent method, in which a beamformer associated with the first antenna and a beamformer associated with the at least one second antenna are updated separately and iteratively.

7. The method of claim 6 , further comprising updating a set of intermediate variables, which are functions of the configuration of the beamformers and the estimated CSI.

8. The method of claim 6 , further comprising repeating the optimizing until convergence of the configuration of the beamformers.

9. The method of claim 1 , wherein optimizing the configuration of the beamformers can be performed asynchronously among transmitting-receiving antenna pairs.

10. A system comprising:

a first antenna;

a processor; and

a non-transitory storage medium storing instructions for execution by the processor,

wherein the processor is configured to:

receive estimated channel state information (CSI) received at the first antenna for at least one channel between the first antenna and at least one second antenna;

obtain antenna correlation information associated with the first antenna and the at least one second antenna, respectively;

calculate a set of noise terms associated with random noise for the at least one channel; and

optimize a configuration of beamformers for the first antenna and the at least one second antenna that maximizes a throughput of all antennas, in accordance with the estimated CSI, the obtained antenna correlation information, and the noise terms.

11. The system of claim 10 , wherein the first antenna is a transmitting antenna, wherein and the at least one second antenna is a receiving antenna.

12. The system of claim 10 , wherein the optimizing includes determining the configuration of the beamformers that maximizes a weighted sum rate of all transmitting-receiving antenna pairs.

13. The system of claim 12 , wherein the optimizing includes determining the configuration of the beamformers that maximizes a weighted sum of lower bounds of mutual information between transmitted and received signals.

14. The system of claim 12 , wherein the optimizing includes determining the configuration of the beamformers that minimizes a weighted sum of mean-square-errors (MSEs) of estimated data streams received at the at least one second antenna, and wherein an MSE of a respective estimated data stream is associated with a corresponding transmit beamformer, a corresponding receive beamformer, the estimated CSI, and the random noise.

15. The system of claim 14 , wherein the optimizing involves a coordinate descent method, in which a beamformer associated with the first antenna and a beamformer associated with the at least one second antenna are updated separately and iteratively.

16. The system of claim 15 , wherein the processor is further configured to update a set of intermediate variables, which are functions of the configuration of the beamformers and the estimated CSI.

17. The system of claim 15 , wherein the the processor is further configured to repeat the optimizing until convergence of the configuration of the beamformers.

18. The system of claim 10 , wherein optimizing the configuration of the beamformers can be performed asynchronously among transmitting-receiving antenna pairs.

19. A network node, comprising:

a CSI-receiving mechanism configured to receive estimated channel state information (CSI) for at least one channel between a first antenna and at least one second antenna;

an obtaining mechanism configured to obtain antenna correlation information associated with the first antenna and the at least one second antenna, respectively;

a noise-term-calculator configured to calculate a set of noise terms associated with random noise for the at least one channel; and

an optimization mechanism configured to optimize a configuration of beamformers for the first antenna and the at least one second antenna that maximizes a throughput of all antennas, in accordance with the estimated CSI, the obtained antenna correlation information, and the noise terms.

20. The node of claim 19 , wherein the first antenna is a transmitting antenna, and the at least one second antenna is a receiving antenna.

21. The node of claim 19 , wherein the throughput is measured by a weighted sum rate of all transmitter-receiver pairs in the system.

22. The node of claim 21 , wherein the optimization mechanism is configured to maximize a weighted sum of lower bounds of mutual information between transmitted and received signals.

23. The node of claim 21 , wherein the optimization mechanism is configured to minimize a weighted sum of mean-square-errors (MSEs) of estimated data streams received at the at least one second antenna, and wherein an MSE of a respective estimated data stream is associated with a corresponding transmit beamformer, a corresponding receive beamformer, the estimated CSI, and the random noise.

24. The node of claim 23 , wherein the optimization mechanism is configured to use a coordinate descent method, which comprises:

receiving updates of receive beamformers from other nodes in the system; and updating the beamformer corresponding to the node using the received updates of the receive beamformers.

25. The node of claim 23 , wherein while optimizing the configuration of beamformers, the optimization mechanism is configured to receive updates of a set of intermediate variables, which are functions of the configuration of the beamformers and the estimated CSI.

26. The node of claim 24 , wherein the optimization mechanism is configured to repeat the optimizing until convergence of the configuration of the beamformers.

27. The node of claim 19 , wherein the optimization mechanism is configured to optimize the configuration of the beamformers associated with the node asynchronously with respect to other nodes.