【摘要】 由于电磁信道的开放特征，存在恶意节点对正常数据流实施干扰，阻止合法接收者获得信息，进而截获并篡改数据，因此针对通信抗干扰的研究非常重要。传统的抗干扰方法采用单一抗干扰方式，并不能根据环境自适应改变抗干扰策略，因此难以在复杂通信环境中达到较好的抗干扰效果。为应对这些挑战，研究了基于深度强化学习（Deep Reinforcement Learning,DRL）的抗干扰通信模型，并搭建仿真系统，利用DRL算法实现智能抗干扰决策。仿真结果表明，提出的智能抗干扰决策能够在复杂环境下根据环境选择最优抗干扰方案，有效提升通信质量。更多还原

【Abstract】 Due to the open characteristics of the electromagnetic channel, there is a risk of malicious nodes jamming with the normal data flow, preventing legitimate receivers from obtaining information, and then intercepting and tampering with the data, which makes the research of communication anti-jamming fairly important. Conventional anti-jamming methods use a single anti-jamming strategy and cannot adaptively make changes according to the environment, so it is difficult to achieve a better anti-jamming effect in complex communication environments. In order to solve these challenges, this paper studies the antijamming communication model based on DRL(Deep Reinforcement Learning) and builds a simulation system to achieve intelligent anti-jamming decision-making by using the DRL algorithm. Simulation results indicate that the proposed intelligent anti-jamming decision-making can choose the optimal anti-jamming scheme in complex environments, which effectively enhances the communication quality.更多还原