Adaptive waveform design for multi-target classification in signal-dependent interference

Abstract

The adaptive radar waveform design in signal-dependent interference and the sequential hypothesis testing for multi-target classification problem are jointly addressed in this paper. The radar knowledge of the environment, or specifically speaking, the probability of each hypothesis, is updated according to the observation of every illumination using the Bayesian channel representation. A the mean time, the radar is capable of changing is next transmitted waveform based on the current knowledge of the environment. The signal-to-interference-plus-noise ratio (SINR)-based waveform design under the constraints of limited energy and constant envelope is used in his paper. The weighed sum of the large frequency responses, which incorporates the probability of each hypothesis, is used to design the optimal waveform. Therefore, the adaptive waveform design and the sequential hypothesis testing problem are combined together, which compose a closed-loop operation. Simulation results show that the adaptively changed waveform outperforms the non-adaptive one in the sense that the average illumination number for multi-target classification is reduced and the SINR is larger which is useful for the succeeding information processing.