A Jamming Decision Method Based on SEEA-TD3 Algorithm

Deep reinforcement learning has been used in electronic warfare. In adversarial situations, jamming decisions are generated by actor network of deep reinforcement learning. Traditional actor network ignores the temporality of radar signal, which leads to the effect of jamming decision is poor. To solve the above problem, this paper proposes a jamming decision method based on SEEA-TD3 algorithm. Combining with a new feature extraction module and squeeze excitation-external attention module, the actor network focuses on the temporal information of radar signal and improves the effectiveness of jamming decision. The simulation results show that the proposed method improves the jamming success rate and optimizes the jamming decision.