NOMA-VLC系统中最大化总和速率功率分配方法

A maximum summation-rate power allocation method in NOMA-VLC system

非正交多址接入(Non-Orthogonal Multiple Access,NOMA)被认为是提高无线通信系统频谱效率的一种很有前途的技术。文中将NOMA技术应用于可见光通信(Visible Light Communication,VLC)中,提出了一种基于深度Q网络(Deep Q Network,DQN)强化学习算法的功率分配方案来解决可见光通信系统最大化总和速率优化问题,该方案充分考虑了用户的信道条件,能够提升系统总和速率,可为VLC系统的功率分配问题提供新的思路。仿真结果表明,所提算法比Q学习功率分配算法、增益比功率分配算法、随机功率分配算法拥有更高的总和速率,在用户数小于11的范围内,总和速率平均分别提升了6.28%、12.20%、51.36%。

Non-Orthogonal Multiple Access(NOMA)is considered as a promising technology to improve the spectrum efficiency of wireless communication systems.This paper applies NOMA technology to Visible Light Communication(VLC),and proposes a power allocation scheme based on Deep Q Network(DQN)reinforcement learning algorithm to solve the problem of maximum summation-rate optimization of visible light communication system,which fully considers the channel conditions of users,can improve system performance and provide a new idea for power allocation of VLC system.The simulation results show that the proposed algorithm has a higher summation rate than Q learning power allocation algorithm,gain ratio power allocation algorithm,and random power allocation algorithm,in the range of users less than 11,and the summation rate of the proposed algorithm is increased by 6.28%,12.20%,51.36%,respectively.