Task Allocation Optimization Algorithm for Multitarget ISAR Imaging Based on Deep Reinforcement Learning Under Suppression Jamming

Inverse synthetic aperture radar (ISAR) can perform high-resolution imaging of non-cooperative targets around the clock and in all weather conditions, which has been widely used in the military field. The mechanism of centralized decision-making and multi-station observation enables the radar network to complete the multi-target ISAR imaging task. A reasonable task allocation solution can give full play to the resources flexibility of radar network, enabling it to efficiently complete imaging tasks in complex electromagnetic environments. Therefore, this paper proposes a deep reinforcement learning-based evolutionary method (DRL-EM) for multi-target imaging task allocation in radar network under concomitant suppression jamming. First, a task allocation model for multi-target ISAR imaging for radar network under concomitant suppression jamming is constructed. Then, a task allocation optimization method that integrates the environment understanding mechanism of deep reinforcement learning and the search and optimization capability of evolutionary method is proposed. Numerical experiments have demonstrated that the proposed algorithm has a faster convergence rate, and the resulting final task allocation solution is more reasonable with a higher utility value.