Least squares predictive transform space-time array processing for adaptive airborne MTI radar

A novel strategy is presented via the application of a multichannel least squares predictive transform (PT) signal modeling procedure. The crux of the technique is the utilization of a multichannel adaptive 'whitening' filter obtained from least squares PT signal modeling to 'pre-whiten' the otherwise angle-Doppler correlated clutter signal prior to detection. The synergistic effect of the prediction and transformation mechanisms provides additional means for reducing both design and implementation complexity while gradually trading off performance-as measured by the degree of clutter cancellation for the application considered. Simulation results demonstrate the utility of the approach in a radar employing an eight element phased array antenna operating in Rayleigh distributed clutter.<>