Research On Resource Allocation Algorithm For RIS-Assisted D2D Communication Networks

With the rapid deployment and development of 5G,new application scenarios for various types of communications are emerging.The exponential growth in the number of end devices has exacerbated the spectrum resource shortage problem and increased the network load.Key wireless communication technologies such as Device to Device(D2D)communication and Reconfigurable Intelligent Surface(RIS)can effectively alleviate the pressure of wireless spectrum resource shortage.The combination of D2 D and RIS technologies can improve communication systems’ spectrum utilization and network capacity,which have critical practical applications and research values.The interference problem is one of the critical issues that must be faced by D2 D communication,and an effective resource allocation strategy is needed to achieve interference suppression.Most of the current resource allocation strategies are designed based on a fixed and stable wireless propagation environment,which has fundamental limitations.In contrast,RIS can intelligently reconfigure the wireless propagation environment to achieve the enhancement of useful signals as well as the suppression of interfering signals.Therefore,this paper studies the resource allocation problem for RIS-assisted D2 D communication networks,and designs resource allocation strategies for single RIS and multi-RIS distributed-assisted D2 D communication scenarios,respectively,with the following specific work and innovations.The optimization problem of maximizing the data transmission rate of D2 D communication links is established for a single RIS-assisted D2 D communication scenario.Based on the Block Coordinate Descent(BCD)algorithm,the optimization problem is decomposed into three subproblems: channel assignment,user power control,and RIS phase shift optimization,and the subproblems are optimized given the remaining variables in turn.Firstly,the feasible set of multiplexing for each pair of D2 D users is determined based on the constraints,and the maximum weighted bipartite graph matching problem is established with the multiplexing link reachable rate as the weight,and the channel allocation is realized by Kuhn-Munkres(KM)algorithm;secondly,the feasible region of user power allocation is determined by the constraints of cellular user quality of service,and the specific feasible points are obtained after analysis;for the RIS phase shift optimization subproblem The RIS phase shift optimization subproblem is then converted into a quadratic constrained quadratic programming problem by Lagrangian dual transform and quadratic transform methods in turn;finally,each subproblem is optimized iteratively until convergence.The simulation results show that the proposed algorithm can effectively improve the data transmission rate of D2 D links while satisfying cellular users’ service quality.Due to the disadvantage of coverage range in single RIS-assisted scenarios,this paper next establishes a multi-RIS distributed-assisted D2 D communication system model and the optimization problem of maximizing the total data transmission rate of D2 D links and cellular links.First,a channel multiplexing scheme with minimum network interference strength is designed,which is less complex and easy for the base station to realize the channel assignment.After that,the Deep Deterministic Policy Gradient(DDPG)algorithm is used to realize the joint optimization of user power and RIS reflection phase shift.Simulation results show that multi-RIS distributed-assisted D2 D communication can provide superior performance gains compared to single-RISassisted,while validating the effectiveness of the proposed algorithm compared to the benchmark solution.