应答式干扰下基于Q学习算法的跳频系统信道调度方法

摘要

In response to effect of reactive jamming on frequency hopping( FH) communication,a mode of cognitive radio frequency hopping system's the channel scheduling based on the Q-learning algorithm is proposed. The model uses the method of machine learning to find out the most reasonable strategy of avoi-ding jamming according to the real-time evaluation rewards on hopping behaviors of the cognitive system, and realizes adapting to the jamming eventually. The performance of the application in response to the reac-tive jamming is simulated on Simulink and the results show that it can significantly reduce the bit error rate ( BER) of FH system less than one percent which is nearly close to the one in unjammed condition.%针对应答式干扰对跳频通信的影响,提出了基于Q学习算法的认知无线电跳频系统信道调度模型.该模型根据认知系统对跳频行为的实时回报评估值,运用机器学习的方法寻找出最合理的规避干扰策略,最终达到适应干扰的目的.运用Simulink对该干扰方式下的算法应用性能进行了仿真验证,结果表明该算法能够降低跳频系统此干扰下的误比特率到1%以下,基本接近未受干扰下的误比特率.