基于改进遗传算法的无线传感器网络覆盖优化  

Coverage optimization for WSNs based on improved genetic algorithm

作　　者：荣威 张屹 王帅[2] 陆瞳瞳 RONG Wei;ZHANG Yi;WANG Shuai;LU Tongtong(School of Mechanical Engineering and Rail Transit,School of Intelligent Manufacturing,Changzhou University,Changzhou 213100,China;School of Computer Science and Technology,East China Normal University,Shanghai 200062,China;School of Business,Changzhou University,Changzhou 213100,China)

机构地区：[1]常州大学机械与轨道交通学院智能制造产业学院,江苏常州213100 [2]华东师范大学计算机科学与技术学院,上海200062 [3]常州大学商学院,江苏常州213100

出　　处：《传感器与微系统》2024年第6期141-144,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(51875053)

摘　　要：提出一种基于逆模型引导算法搜索的多目标演化算法(MOEA-OMG),通过对种群的目标空间随机采样,然后利用高斯过程将采样解映射回决策空间,得到包含种群分布信息的试验解,引导算法搜索,利用提出的重组算子将试验解与其他解个体进行组合,产生高质量后代解。将算法应用到解决无线传感器网络(WSNs)覆盖问题,并与传统的几种优化算法进行实验对比,结果表明,所提算法在求解WSNs覆盖问题时,展现出较为明显的性能优势。A multi-objective genetic algorithm based on inverse model guided algorithm search(MOEA-OMG)is proposed,which obtains the experimental solution containing the distribution information of the population by random sampling of objective space of the population,and then mapping the sampled solution back to the decision space by using Gaussian process,guides the algorithm search,and produces high-quality offspring solutions by combining the trival vector solution with the other solution individuals by using the proposed recombination operator.The algorithm is applied to solve the WSNs coverage problem and experimentally compared with several traditional optimization algorithms,and the results show that the designed algorithm based on the inverse model-guided search shows more obvious performance advantages in solving the WSNs coverage problem.

关 键 词：遗传算法 重组算子 逆建模 覆盖优化 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TP18[自动化与计算机技术—计算机科学与技术]