一种协同进化的无线传感器网络多播路由算法  

Coevolutionary Multicast Routing Algorithm for Wireless Sensor Networks

作　　者：傅彦铭[1] 周兴 黄保华[1] 张小萍[1] 朱杰夫 FU Yan-ming;ZHOU Xing;HUANG Bao-hua;ZHANG Xiao-ping;ZHU Jie-fu(School of Computer,Electronics and Information,Guangxi University,Nanning 530004,China)

机构地区：[1]广西大学计算机与电子信息学院,南宁530004

出　　处：《小型微型计算机系统》2023年第2期401-407,共7页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61962005)资助

摘　　要：物联网促进了无线传感器网络的发展,同时对无线传感器网络在能量损耗和通信质量保证等方面的性能也提出更高的要求.本文提出一个基于多目标协同进化的多播路由算法(CCMRA)来解决同时优化功耗、时延和丢包率3个目标的无线传感器网络多播路由问题.CCMRA算法的种群个体是目标传感器网络其中一颗多播树,将种群分成两个子种群LP和GP,它们针对不同的最优多播树结构分别采用Local操作和Global操作策略生成下一代多播树种群个体,利用竞争性协同进化机制在两个子种群之间通过选择和融合交换种群之间的信息,经过逐代进化最终得到最优的多播树.上述种群内部和种群之间的协同进化机制,增加了种群多样性和随机性,从而提高算法的搜索能力,使算法能够跳出局部最优,获得良好的收敛精度和速度.本文采用8种不同复杂程度的无线传感器网络场景,引入3种对比算法.实验表明CCMRA在功耗、时延和丢包率3个指标上表现出优良的性能.同时在超体积、反向世代距离和世代距离等多目标指标上的表现也优于对比算法,进一步验证CCMRA算法在收敛性和多样性保持上具有竞争力.The Internet of things promotes the development of wireless sensor networks,and puts forward higher requirements for the performance of wireless sensor networks in terms of energy loss and communication quality assurance.In this paper,a multi-objective coevolution based multicast routing algorithm(CCMRA)is proposed to solve the multicast routing problem of wireless sensor networks(WSNs)which simultaneously optimizes the power consumption,delay and packet loss rate.The individual population of CCMRA algorithm is a multicast tree in the target sensor network.The population is divided into two subpopulations LP and GP.According to different optimal multicast tree structures,they use local operation and global operation strategies to generate the next generation of multicast tree population individuals,and use competitive coevolution mechanism to exchange information between the two subpopulations through selection and fusion,Finally,the optimal multicast tree is obtained through generation by generation evolution.The coevolution mechanism within and between populations increases the diversity and stochasticity of the population,improves the search ability of the algorithm,makes the algorithm jump out of the local optimum,and achieves good convergence accuracy and speed.In this paper,eight different complexity scenarios of wireless sensor networks are adopted,and three comparison algorithms are introduced.Experiments show that CCMRA has excellent performance in power consumption,delay and packet loss rate.At the same time,the performance of CCMRA is also better than that of the comparison algorithm in multi-objective indexes such as hyper volume,inverted generational distance and generational distance,which further verifies that CCMRA algorithm is competitive in convergence and diversity maintenance.

关 键 词：协同进化算法 多目标优化 无线传感器网络 多播路由 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]