基于A3C的认知物联网通信干扰消除算法OA北大核心CSTPCD

To address the intelligent needs of spectrum resource interference management,in this study,an Asynchronous Advantage Actor-Critic(A3C)-based intelligent interference elimination algorithm is proposed to tackle the interference problem caused by spectrum resource sharing in a Cognitive Internet of Things(CIoT)communication system.Through learning and optimization of the agent,the algorithm helps Secondary Users(SU)make optimal decisions when affected by interference,thereby improving communication quality and system performance.When the communication quality of the SU is affected by interference,the learning and optimization of the agent enables the SU to choose the action with the lowest interference degree according to the current position information,transmit power,and receive power and interference degree,whereby a reward is offered after performing the action.Intelligent agents attempt different actions to reduce interference and adjust their strategies based on the reward feedback,thereby maximizing the reward function by considering the interference level and signal quality indicators and minimizing the impact of interference on communication quality.The simulation results show that compared with the traditional k-means,Deep Recurrent Q-Network(DRQN),and Deep Q-Network(DQN)optimization algorithms,the A3C-based interference cancellation algorithm has a shorter convergence time,higher execution efficiency,and at least 7％higher throughput performance than the three benchmark methods.This demonstrates that the proposed algorithm can effectively reduce the adverse effects of interference on communication quality.