Angular bin compression for joint domain localised (JDL) processor

We propose a modified joint domain localised (JDL) processor that significantly reduces the computational cost. Space-time adaptive processing (STAP) has been proven to be able to suppress clutter efficiently. However due to the size of its covariance matrix, forming the inverse of the covariance matrix is computational expensive. The JDL processor, a reduced dimension version of STAP allows practical implementation through the use of a small dimension covariance matrix (e.g. [9 × 9] matrix when using a 3 × 3 JDL). The propose modified JDL reduces the covariance matrix dimension further to a [3 × 3] matrix by using a tuned discrete Fourier transform.