摘要
在无线通信网络环境中,分布式客户端节点在用户隐私保护、数据传输效率、能量利用效率之间较难实现平衡。针对该问题,提出一种结合联邦学习与传统集中式学习的能量与信息管理优化策略。以覆盖性强、适用性广的移动信息采集设备作为学习服务器,将分布分散、资源受限的客户端节点作为学习参与者,通过构建马尔科夫决策模型分析客户端节点在移动信息采集过程中的状态变化和行为模式,同时采用值迭代算法和深度强化学习算法对该模型进行近似求解,获得客户端节点最优的信息传输与能量管理组合策略。仿真结果表明,相比MDP、GRE、RAN策略,该策略的长期效用较高且数据延迟较小,可实现客户端节点在信息传输过程中的数据隐私性、数据可用性与能量消耗之间的最优平衡。
To balance user privacy protection,data transmission efficiency,and energy utilization efficiency of distributed user nodes in wireless communication networks,a federated learning-based strategy for optimizing information transmission and energy management is established.The mobile information collection devices with extended coverage and applicability are deployed as servers,and the distributed user nodes with limited resources are deployed as workers.Then a Markov decision model is built to analyze the status changes and behavior patterns of nodes during mobile information collection.The Markov model is approximately solved by using a value iteration algorithm and deep reinforcement learning algorithm,so an optimal strategy for information transmission and energy management for user nodes is obtained.Simulation results show that compared with MDP,GRE and RAN strategies,the proposed strategy has better long-term utility and less data delay.It can achieve an optimal balance between data privacy,data availability and energy consumption during information transmission of user nodes.
作者
杨文琦
章阳
聂江天
杨和林
康嘉文
熊泽辉
YANG Wenqi;ZHANG Yang;NIE Jiangtian;YANG Helin;KANG Jiawen;XIONG Zehui(School of Computer Science and Technology,Wuhan University of Technology,Wuhan 430063,China;Hubei Key Laboratory of Transportation Internet of Things,Wuhan University of Technology,Wuhan 430063,China;School of Computer Science and Engineering,Nanyang Technological University,Singapore 639798,Singapore;Pillar of Information Systems Technology and Design,Singapore University of Technology and Design,Singapore 487372,Singapore)
出处
《计算机工程》
CAS
CSCD
北大核心
2022年第1期188-196,203,共10页
Computer Engineering
基金
国家自然科学基金(62071343)。
关键词
联邦学习
无线通信网络
信息传输
能量管理
马尔科夫决策过程
深度强化学习
federated learning
wireless communication network
information transmission
energy management
Markov Decision Process(MDP)
Deep Reinforcement Learning(DRL)
