Selecting a reduced-rank transformation for STAP-a direct form perspective

Abstract

We introduce the signal-to-interference plus noise ratio (SINR) metric as a tool for selecting a rank reduction transformation that maximizes the output SINR of a direct form space-time adaptive processor. The SINR metric is used to identify the eigenvectors with the greatest impact on the output SINR which become the columns of the rank reduction transformation. The SINR metric method exhibits a graceful degradation in performance as the rank of the transformation approaches one.