An Improved Clutter Filtering and Spectral Moment Estimation Algorithm for Staggered PRT Sequences

Abstract

In the staggered pulse repetition time (PRT) radar, the phase difference between the autocorrelations at lag T1 and T2 is used for velocity estimation. This paper investigates velocity estimates from the autocorrelation at shorter lag, while the longer lag is used to resolve the ambiguity. This velocity estimate is shown to have lower error than the one using the phase difference. This method is preferred in the absence of ground clutter. In a recent paper on the spectral processing of the staggered PRT sequences, a new algorithm was presented that enables the recovery of spectral moments in presence of ground clutter. Although this algorithm recovers velocity over the extended unambiguous interval corresponding to the difference PRT, there is a small bias error in the velocity and spectrum width estimates due to the loss of some of the signal components in the process of filtering the clutter. An enhancement in the algorithm is suggested that enables removal of this bias by reconstructing the lost spectral components before the spectral moments are estimated. The proposed algorithm completely removes the bias error in the velocity and spectral width most of the time thereby significantly improving these estimates. The window function and the number of samples used for processing significantly influence the performance of the clutter filter. The window function effect is explored via simulation, and these results are presented.