CLAIMS

We claim:

1. A method comprising:

obtaining an estimate of a receive MIMO channel;

computing a desired fully-digital combiner matrix based on the estimate;

decomposing the fully-digital combiner matrix into an analog combiner matrix and a digital combiner matrix;

decomposing the analog combiner matrix into two constituent matrices, such that each element of the constituent matrices has unit magnitude;

configuring phase rotators in a beamformer having two parallel analog combiners according to elements of the two constituent matrices;

processing received signals using the beamformer and a plurality of radio frequency (RF) chains; and,

processing outputs of the RF chains using the digital combiner matrix.

2. The method of claim 1 wherein the desired fully-digital combiner is W_{FD} , and is based on:

^{s}, where

i = 1 for single user, i = 2 for multi user.

3. The method of claim 1 wherein the constituent matrices are

, and wherein the phase rotators of the two parallel analog combiners are configured according to respectively.

4. The method of claim 1 wherein each element of the two constituent matrices has an angle based on a combination of a corresponding angle of an element of the fully-digital combiner matrix, and an offset angle.

5. The method of claim 4 wherein the offset angle is a positive angle for an element of one constituent matrix of the two constituent matrices, and the offset angle is a negative angle for an element of the other constituent matrix of the two constituent matrices.

6. The method of claim 1 , wherein the beamformer further comprises a summation circuit for combining a first and second intermediate signal vector generated by the respective analog combiner of the two parallel analog combiners.

7. The method of claim 1 wherein a phase of each element of the constituent matrices is based on an inverse cosine function operating on a normalized magnitude of the fully-digital combiner matrix.

8. The method of claim 1 wherein the digital combiner matrix is an arbitrary invertible matrix.

9. The method of claim 1 wherein the digital combiner matrix is a scaled identity matrix given by W_{D} = cl_{Ns}, where c is one half the magnitude of the largest element of the fully-digital combiner matrix, and I_{Ns}is an identify matrix having dimension Ns.

10. An apparatus comprising:

a beamformer having two parallel analog combiners, each of the parallel analog combiners having a plurality of sets of phase rotation circuits configured to generate a respective plurality of outputs as a respective intermediate output vector;

a summation circuit configured to combine the respective intermediate output vectors to generate an analog beam-formed signal vector;

a processor configured to:

obtain a desired fully-digital combiner matrix, and decompose the fully-digital combiner matrix into an analog combiner matrix and a digital combiner matrix;

decompose the analog combiner matrix into two constituent matrices, such that each element of the constituent matrices has unit magnitude;

configuring the sets of phase rotators according to elements of the two constituent matrices;

a plurality of radio frequency (RF) chains, each RF chain configured to process a signal from the analog beam-formed signal vector; and,

a digital combiner circuit configured to apply the digital combiner matrix to outputs of the plurality of RF chains.

1 1. The apparatus of claim 10 wherein each element of the two constituent matrices has an angle based on a combination of a corresponding angle of an element of the fully-digital combiner matrix, and an offset angle.

12. The apparatus of claim 1 1 wherein the offset angle is a positive angle for an element of one constituent matrix of the two constituent matrices, and the offset angle is a negative angle for an element of the other constituent matrix of the two constituent matrices.

13. The apparatus of claim 10 wherein a phase of each element of the constituent matrices is based on an inverse cosine function operating on a normalized magnitude of the fully-digital combiner matrix.

14. The apparatus of claim 10 wherein the digital combiner matrix is an arbitrary invertible matrix.

15. The apparatus of claim 10 wherein the digital combiner matrix is a scaled identity matrix given by W_{D} = cl_{Ns}, where c is one half the magnitude of the largest element of the fully-digital combiner matrix, and I_{Ns}is an identify matrix having dimension Ns.