环境感知信息辅助的认知雷达波形参数智能选择

Multi-function radar can perform various missions simultaneously in a complex and variable electromagnetic environment. How to improve the intelligence level of radars，so that they can adapt to the change of environment and the requirements of tasks，has gradually become a high-profile research. Aiming at the optimization problem of maneuvering target detection and tracking performance in clutter environment，this paper proposes an intelligent scheduling algorithm for radar waveform parameters based on environmental perception. The reward function is designed under the maximum signal-to-noise ratio criterion and the minimum mean square error criterion. And the Q learning and deep Q learning network are used for training. Through the interaction between the radar and the environment，the multi-frame clutter information is fully utilized which can effectively avoid the clutter occlusion problem caused by blur, and improve the SNR and tracking accuracy. The results of the airborne radar simulation experiments show that in the process of maneuvering target detection and tracking in clutter environment, the proposed environment sensing information assisted waveform intelligent selection scheme can achieve better performance with improved processing efficiency than the traditional heuristic algorithms.