Research on radar jamming resource scheduling based on deep reinforcement learning

Abstract:
In this paper, the autonomous decision-making method of the scheduling of radar jamming resource for remote support jamming aircraft is investigated based on the actual combat requirements using remote support jamming aircraft to cover penetration aircraft. First, the decision-making for the scheduling of radar jamming resources for remote support jamming aircraft is analyzed. Subsequently, a model of radar jamming resource scheduling is built. Next, a decision-making method of the scheduling of radar jamming resource for remote support jamming aircraft is proposed based on the proximal policy optimization (PPO) algorithm. Lastly, for typical combat scenarios, simulation experiments are performed in the simulation environment deduced by electronic countermeasure. As revealed by the experimental results, the PPO algorithm can drive the agent to determine the jamming frequency and release the jamming beam in the process of radar jamming, which has high real-time performance.