Very fast adaptive array processing

We consider the problem of detection of a signal in the presence of some interference plus white gaussian noise. The proposed improved detector rapidly eliminates coherent interference. We present the results of an approximate probability distribution for the signal-to-noise ratio (SNR) of the improved detector. The noise consists of two components, a strong interference component, which has a rank-one or rank-two covariance matrix, plus white noise. Weight vectors are computed using the principal eigenvectors of the estimated covariance matrix of the interference. The statistics of our improved method and the conventional adaptive detector based on the inverse of the estimated covariance matrix are presented. The advantage of the new method over the conventional method is that less observed data are required to produce a desired SNR with high probability. These results extend earlier results of Kirsteins and Tufts which appeared in the 1985 IEEE International Conference on Acoustics, Speech and Signal Processing.