基于深度强化学习的可见光定位通信一体化功率分配研究

Research on power allocation of integrated VLPC based on deep reinforcement learning

为了实现定位通信一体化功率分配,提出了一种基于深度强化学习的可见光定位通信(VLPC)一体化系统的功率分配方案。首先,提出了定位通信一体化帧结构设计;其次,利用定位信息实现了信道状态信息的估计,并推导了定位误差的克拉美罗下界(CRLB);再次,阐明了定位精度和通信速率的内在耦合关系;最后,提出了基于深度确定性策略梯度的VLPC动态功率分配方案。仿真结果表明,所提方案可同时实现高精度定位和高速通信。

A power allocation scheme for integrated visible light position and communication(VLPC)system based on deep reinforcement learning was proposed to achieve power allocation for communication positioning integration.First,the frame structure design of integrated VLPC was proposed.Then the channel state information could be estimated by using the positioning information,and the CRLB of the positioning error was derived.Furthermore,the internal coupling relationship between positioning accuracy and communication rate was clarified.On this basis,a dynamic power alloca-tion scheme based on deep deterministic policy gradient was proposed.Simulation results show that the proposed scheme can simultaneously achieve high-precision positioning and high-speed communication.