摘要
电磁环境的复杂多变对军用无线通信系统的抗干扰能力提出了更高的要求,传统的扩跳频抗干扰方式缺乏灵活性,难以抵挡动态变化的干扰信号。从抵抗动态干扰的需求出发,提出了基于改进Q-Learning的信道决策方法。在传统Q-Learning算法的基础上采用动态ε机制的动作选择策略,并设计了一种将误码率和能量消耗相结合的奖励函数。在固定模式干扰、扫频干扰、跟随式干扰下进行验证,仿真结果表明,所提算法能够较快速收敛,降低与干扰发生“碰撞”的可能性,有效地进行了干扰规避,适用于抗干扰动态决策。
The complexity and variability of the electromagnetic environment puts forward higher requirements for the anti-jamming capability of military wireless communication systems,and the traditional spread spectrum hopping anti-jamming method lacks flexibility to resist the dynami-cally changing jamming signals.From the demand of resisting dynamic jamming,a channel deci-sion-making method based on improved Q-Learning is proposed,which adopts the action selec-tion strategy of dynamicεmechanism on the basis of the traditional Q-Learning algorithm,and designs a reward function that combines the bit error rate and energy consumption.The proposed algorithm is validated under fixed mode jamming,frequency sweeping jamming and following jamming.Simulation results show that the proposed algorithm can converge quickly,reduce the possibility of“collision”with jamming,and effectively avoid jamming,which is suitable for anti-jamming dynamic decision-making.
作者
侯艳丽
贾怡霈
崔惠敏
HOU Yanli;JIA Yipei;CUI Huimin(School of Information Science and Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,China)
出处
《电子信息对抗技术》
2025年第5期60-65,共6页
Electronic Information Warfare Technology
基金
河北省重点研发计划项目(21355901D)。
