A signal subspace beamformer

It is shown that it is possible to approximate a maximum-likelihood method, leading to a very high performance subspace algorithm. This has all the advantages of subspace methods (i.e. inexpensive), but with much better threshold performance than existing methods such as MUSIC. The method is applicable to arbitrary array geometry, and is demonstrated on both linear and curved arrays with limited snapshots, multipath, and with correlated (but unknown) noise. The results demonstrate the robustness of the algorithm and its superiority over MUSIC, with no additional cost.<
>