An Anti-Jamming Beam and Power Allocation Strategy for Multistatic Radar System via Multi-Agent Deep Reinforcement Learning

Multistatic radar system (MRS) has the potential to achieve excellent detection performance in electronic warfare thanks to the ability of mutual support among different radars. In this paper, we propose a novel joint beam and power allocation method for the MRS in a multiple target tracking scenario, where each target is equipped with a jamming device. We formulate the problem in a multi-agent deep reinforcement learning manner. To solve the proposed problem, we express the hybrid action of each radar with continuous values and employ multi-agent deep deterministic policy gradient (MADDPG) method to solve the joint beam and power allocation strategy efficiently. Numerical simulations show that the MRS achieves better tracking performance with the beam and power strategy provided by MADDPG.