卫星网络中基于双向寻优粒子群优化算法的连接计划设计  

Contact plan design based on bi-directional particle swarm optimization in satellite network

作　　者：戴翠琴[1] 唐煌 郭林峰 DAI Cuiqin;TANG Huang;GUO Linfeng(School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆邮电大学通信与信息工程学院

出　　处：《通信学报》2019年第8期189-199,共11页Journal on Communications

基　　金：国家自然科学基金资助项目(No.61601075);重庆市科委自然科学基金资助项目(No.cstc2016jcyjA0174)~~

摘　　要：针对卫星网络中拓扑时变、连接瞬断及资源受限的问题,提出了一种基于双向粒子群优化(BPSO)的连接计划设计(CPD)方案。首先,通过对资源受限的时变卫星网络分析,建立了基于任务的时间拓展图(TEG)模型;其次,考虑网络拓扑的离散性和节点资源的有限性,通过初始化、编码及修复环节生成卫星网络中的可用连接计划(CP);再次,根据执行任务的特点设计评价函数以区分所生成的可用CP的优劣;最后,根据CP中链路稀疏的特点确定最差位置中待修正的比特,提出BPSO算法不断修正CP以优化空间数据传输性能。仿真结果表明,所提方案能够有效降低任务传输时间,提升任务到达率。Aiming at the problems of time-varying topology,intermittent connection,and constrained resource in satellite network,a contact plan design(CPD)scheme based on bi-directional particle optimization(BPSO)algorithm was proposed.Firstly,the task-based time-expanded graph(TEG)model was constructed through the analysis of resource-constrained time-varying satellite network.Next,available contact plan(CP)in satellite network were generated through initialization,coding and repairing by considering the discreteness of network topology and the limitation of node resources.Then,an evaluation function was designed according to the characteristics of the execution task to distinguish whether the generated available CP was good or bad.Finally,the bits to be corrected in the worst location were determined according to the sparse characteristics of links in CP,and the CP was continuously modified by the proposed BPSO algorithm to optimize the performance of spatial data transmission.The simulation results show that the proposed BPSO-based CPD scheme can effectively reduce the task delivery time and increase the task arrival rate.

关 键 词：卫星网络 连接计划设计 双向寻优粒子群算法 传输时间 到达率 

分 类 号：TN929.5[电子电信—通信与信息系统]