Adaptive waveform optimization algorithm based on UWB MIMO radar

Abstract

For the issue of deterioration in detection performance caused by dynamically changing environment in UWB MIMO radar, we propose a novel adaptive waveform design which is aimed to improve the ability of discriminating target and clutter from the radar scene. Firstly, we consider using a Dirichlet process mixture model (DPMM)-based clustering approach to discriminate individual extended targets. Then we apply the minimization mutual-information (MI) strategy between individual targets echoes to select the best waveform from an ensemble of UWB waveforms for transmission. With this approach, the radar system constantly learns about its surroundings and adopts its operational mode accordingly based upon the MI minimization criterion. Finally, simulation results demonstrate that the optimal design method brings an improvement in the target detection ability and target discrimination capability.