环境感知信息辅助的认知雷达波形参数智能选择

Environment Sensing Information Aided Waveform Parameters Intelligent Selection for Cognitive Radar

现代雷达往往需要在复杂多变的电磁环境中完成多种任务。如何提升雷达的智能化水平,使其能够适应环境变化和任务需求,已成为近年来备受关注的研究课题。本文针对杂波环境下机动目标检测与跟踪的性能优化问题,提出了一种基于环境感知的雷达波形参数智能调度算法。基于最大信噪比准则和最小均方误差准则设计了奖励函数,并利用Q学习与深度Q学习网络进行了训练,通过雷达与环境的交互,充分利用环境中多帧杂波信息,可有效避免由于模糊导致的杂波遮蔽问题,提升目标信噪比和跟踪精度。机载雷达仿真实验结果表明,在杂波环境下对机动目标检测和跟踪过程中,本文提出的环境感知信息辅助的波形智能选择方案可获得比传统启发式算法更高的处理效率和更大的性能改善。

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.