智能反射面增强的多无人机辅助语义通信资源优化  

Optimization of Resource Allocation for Intelligent Reflecting Surface-enhanced Multi-UAV Assisted Semantic Communication

作　　者：王浩博 吴伟 周福辉[2] 胡冰 田峰[1] WANG Haobo;WU Wei;ZHOU Fuhui;HU Bing;TIAN Feng(School of Communications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;College of Electronic and Information Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China;School of Modern Posts,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003 [2]南京航空航天大学电子信息工程学院,江苏南京211106 [3]南京邮电大学现代邮政学院,江苏南京210003

出　　处：《无线电通信技术》2024年第2期366-372,共7页Radio Communications Technology

基　　金：国家重点研发计划(2020YFB1807602);国家自然科学基金(62271267);广东省促进经济发展专项资金(粤自然资合[2023]24号);国家自然科学基金(青年项目)(62302237)。

摘　　要：无人机(Unmanned Aerial Vehicle,UAV)为无线通信系统提供了具有高成本效益的解决方案。进一步地,提出了一种新颖的智能反射面(Intelligent Reflecting Surface,IRS)增强多UAV语义通信系统。该系统包括配备IRS的UAV、移动边缘计算(Mobile Edge Computing,MEC)服务器和具有数据收集与局部语义特征提取功能的UAV。通过IRS优化信号反射显著改善了UAV与MEC服务器的通信质量。所构建的问题涉及多UAV轨迹、IRS反射系数和语义符号数量联合优化,以最大限度地减少传输延迟。为解决该非凸优化问题,本文引入了深度强化学习(Deep Reinforce Learning,DRL)算法,包括对偶双深度Q网络(Dueling Double Deep Q Network,D3QN)用于解决离散动作空间问题,如UAV轨迹优化和语义符号数量优化;深度确定性策略梯度(Deep Deterministic Policy Gradient,DDPG)用于解决连续动作空间问题,如IRS反射系数优化,以实现高效决策。仿真结果表明,与各个基准方案相比,提出的智能优化方案性能均有所提升,特别是在发射功率较小的情况下,且对于功率的变化,所提出的智能优化方案展示了良好的稳定性。Unmanned Aerial Vehicles(UAV)present a cost-effective solution for wireless communication systems.This article i ntroduces a novel Intelligent Reflecting Surface(IRS)to augment the semantic communication system among multiple UAVs.The system encompasses UAV equipped with IRS,Mobile Edge Computing(MEC)servers,and UAV featuring data collection and local semantic feature extraction functions.Optimizing signal reflection through IRS significantly enhances communication quality between drones and MEC servers.The formulated problem entails joint optimization of multiple drone trajectories,IRS reflection coefficients,and the number of semantic symbols to minimize transmission delays.To address this non-convex optimization problem,this paper introduces a Deep Reinforcement Learning(DRL)algorithm.Specifically,the Dueling Double Deep Q Network(D3QN)is employed to address discrete action space problems such as drone trajectory and semantic symbol quantity optimization.Additionally,Deep Deterministic Policy Gradient(DDPG)algorithm is utilized to solve continuous action space problems,such as IRS reflection coefficient optimization,enabling efficient decision-making.Simulation results demonstrate that the proposed intelligent optimization scheme outperforms various benchmark schemes,particularly in scenarios with low transmission power.Furthermore,the intelligent optimization scheme proposed in this paper exhibits robust stability in response to power changes.

关 键 词：无人机网络 智能反射面 语义通信 资源分配 

分 类 号：TN925[电子电信—通信与信息系统]