《电讯技术》，北大中文核心期刊，中国科技核心期刊！

首页 期刊视频编委会征稿启事出版道德声明审稿流程读者订阅论文查重联系我们English

引用本文：

Nawaf Q.H.Othman,杨清海,蒋昕沛.基于 DDQN 和 GNN 的分布式无人机路由资源联合优化[J].电讯技术，2026，66（1）： - . [点击复制]
Nawaf Q.H.Othman,YANG Qinghai,JIANG Xinpei.Joint Optimization of Routing and Resource Allocation in Decentralized UAV Networks Based on DDQN and GNN[J].，2026，66（1）： - . [点击复制]

【HTML】【打印本页】【下载PDF全文】【查看/发表评论】【下载PDF阅读器】【关闭】

本文已被：浏览 122次下载 51次	码上扫一扫！
基于 DDQN 和 GNN 的分布式无人机路由资源联合优化
NawafQ.H.Othman,杨清海,蒋昕沛
0 字体:加大+\|默认\|缩小-
(1. 西安电子科技大学通信工程学院，西安 710071;2. 伊利诺伊大学厄本那-香槟分校电气与计算机工程系，伊利诺伊州厄巴纳-香槟 61801;2. 伊利诺伊大学厄本那-香槟分校电气与计算机工程系，伊利诺伊州厄巴纳-香槟 61802;2. 伊利诺伊大学厄本那-香槟分校电气与计算机工程系，伊利诺伊州厄巴纳-香槟 61803)

摘要:

由于干扰及网络拓扑的快速变化，在去中心化的无人机网络中实现路由与资源分配的优化仍面临诸多挑战。提出了一种新颖的框架，融合了双深度Q网络与图神经网络，以实现路由和无线资源分配的联合优化。该框架采用图神经网络对网络拓扑进行建模，并利用双深度Q网络自适应调控路由和资源分配，从而有效解决干扰问题并提高系统性能。仿真结果表明，与传统方法（如目的地最近法、最大信噪比法及基于多层感知器的模型）相比，所提方法吞吐量提升约23.5%，连接概率提高约50%，跳数减少约17.6%，验证了其在动态无人机网络中的有效性。

关键词: 分布式无人机网络资源分配路由算法图神经网络双深度Q网络深度强化学习

DOI：10.20079/j.issn.1001-893x.250208001

基金项目:国家自然科学基金资助项目(61971327);广州市重点科技研究与开发计划(202206030003)

Joint Optimization of Routing and Resource Allocation in Decentralized UAV Networks Based on DDQN and GNN

Nawaf Q.H.Othman,YANG Qinghai,JIANG Xinpei

(1.School of Communication Engineering,Xidian University,Xi’an 710071,China;2.Department of Electric and Computer Engineering,University of Illinois at UrbanaChampaign,Urbana,IL 61801,USA;2.Department of Electric and Computer Engineering,University of Illinois at UrbanaChampaign,Urbana,IL 61802,USA;2.Department of Electric and Computer Engineering,University of Illinois at UrbanaChampaign,Urbana,IL 61803,USA)

Abstract:

Optimizing routing and resource allocation in decentralized unmanned aerial vehicle(UAV) networks remains challenging due to interference and rapidly changing topologies.The authors introduce a novel framework combining double deep Qnetworks(DDQNs) and graph neural networks(GNNs) for joint routing and resource allocation.The framework uses GNNs to model the network topology and DDQNs to adaptively control routing and resource allocation,addressing interference and improving network performance.Simulation results show that the proposed approach outperforms traditional methods such as ClosesttoDestination(c2Dst),MaxSINR(mSINR),and MultiLayer Perceptron(MLP)based models,achieving approximately 23.5% improvement in throughput,50% increase in connection probability,and 17.6% reduction in number of hops,demonstrating its effectiveness in dynamic UAV networks.

Key words: decentralized UAV network resource allocation routing algorithm GNN DDQN DRL