To boost the secrecy rate (SR) of the conventional directional modulation (DM) network and overcome the double fading effect of the cascaded channels of passive intelligent reflecting surface (IRS), a novel active IRS-assisted DM system is investigated in this paper. Aiming to maximize the SR, two power allocation (PA) strategies, called maximizing SR based on fractional programming (FP) (Max-SR-FP) and maximizing SR based on derivative operation (DO) (Max-SR-DO), are proposed by jointly designing the PA factors, beamforming vector, and phase shift matrix of IRS, subject to the power constraint at IRS. The former with higher performance employs the FP and successive convex approximation (SCA) algorithms to design the confidential message PA factor and the total PA factor at the base station, and the SCA algorithm is also utilized to design the beamforming vector and the phase shift matrix of the IRS. The latter with lower complexity adopts the DO, and equal amplitude reflecting (EAR) and general power iterative (GPI) methods to solve them, respectively. The simulation results show that compared with the benchmark PA schemes, both the proposed PA schemes achieve a significant SR performance improvement. Moreover, the SR gap between two proposed schemes decreases gradually with the increases of the number of IRS phase shift element.

翻译：为提升传统定向调制网络的安全速率并克服无源智能反射面级联信道的双重衰落效应，本文研究了一种新型有源IRS辅助的定向调制系统。以最大化安全速率为目标，通过联合设计IRS的功率分配因子、波束成形向量及相移矩阵，提出了两种基于分数规划的Max-SR-FP和基于导数运算的Max-SR-DO功率分配策略。前者性能更优，采用分数规划与逐次凸逼近算法设计基站的机密消息功率分配因子与总功率分配因子，并利用SCA算法优化波束成形向量及IRS相移矩阵；后者复杂度更低，分别采用DO、等幅反射法和通用功率迭代法进行求解。仿真结果表明，与基准功率分配方案相比，所提两种方案均能实现明显SR性能提升，且随着IRS相移单元数量增加，两种方案间的SR差距逐渐缩小。