本文研究一种智能超表面(reconfigurable intelligent surface,RIS)辅助的雷达通信一体化(dual-functional radar and communication,DFRC)系统,在进行多用户通信的同时实现杂波环境下的非视距区域目标探测.为了同时优化雷达和通信性能,本文以雷达接收机的输出信干噪比和通信用户接收信号的均方误差为度量,联合设计DFRC基站的发射波形、RIS反射系数和雷达接收滤波器.由于发射波形和RIS反射系数均受到恒模约束,该问题是一个复杂的非凸约束高阶分式规划问题.为此,本文基于分式规划和交替最小化技术,提出了一种有效的交替优化求解算法.其中,发射波形设计是恒模约束二次函数最小化问题,本文采用二阶黎曼牛顿法快速求解;而RIS反射系数设计是恒模约束四次函数最小化问题,本文利用一致交替方向乘子法技术,通过引入辅助变量将其分解为恒模约束二次函数最小化子问题迭代求解,并采用二阶黎曼牛顿法求解.仿真结果表明,RIS辅助的DFRC系统可以在杂波环境下有效探测非视距目标,且能够显著改善多用户通信性能.
Joint design of constant modular transmit waveform and passive beamforming for RIS-aided dual-functional radar and communication
This paper investigates the reconfigurable intelligent surface(RIS)-aided dual-functional radar and communication(DFRC)system for non-line-of-sight target detection in cluttered environments while performing multi-user communication.To optimize radar and communication performance simultaneously,this paper jointly designs the transmit waveform,RIS reflection coefficients and radar receiving filter for the DFRC base station by using the output signal-to-interferece-plus-noise ratio of the radar receiver and the mean-square error of communication users as the metrics.The proposed problem is a complicated non-convex constrained higher-order fractional programming problem due to the constant-modular constraints on both transmit waveform and RIS reflection coefficients.To tackle this issue,this paper provides an effective alternating optimization solution algorithm based on the fractional programming and alternating minimization techniques.Specifically,the transmit waveform design is a constant-modular constrained quadratic function minimization problem and thus is quickly solved by the second-order Riemannian Newton method(RNM).The RIS reflection coefficient design is a constant-modular constrained quartic function minimization problem.Based on the consensus alternating direction method of multipliers,the quartic problem is decomposed into a series of constant-modular constrained quadratic function minimization sub-problems by introducing auxiliary variables,and then iteratively solved by the RNM.Numerical results show that the proposed RIS-aided DFRC can effectively detect non-line-of-sight targets in cluttered environments and significantly improve the multi-user communication performance simultaneously.
reconfigurable intelligent surfacedual-functional radar and communicationmanifold optimizationRiemannian Newton methodalternating direction method of multipliers
NETL
NSTL
万方数据
