基于差分演化算法的无线传感器网络能量分配优化——实验研究.docVIP

基于差分演化算法的无线传感器网络能量分 配优化——实验研究 龚文引 ，蔡之华 中国地质大学（武汉）计算机学院，武汉，430074 摘要：能量分配优化是无线传感器网络设计的核心问题之一。一般来说，能量分配优化问题可 以转换成一个带约束的数值优化问题。差分演化算法是一种高效的全局优化算法，基于此考 虑，本文对差分演化算法用于求解能量分配优化问题进行实验研究，以便为进一步研究给出指 导性建议。我们分别对不同变异策略和不同参数设置对算法性能的影响进行实验研究。实验结 果表明，差分演化算法在能量分配优化问题的求解中对参数 CR 和 F 的设置很敏感；但是， 相对与参数设置敏感而言，变异策略对算法的影响较小。 关键词：差分演化算法；无线传感器网络；能量分配优化；实验研究 中图分类号： TP 391. An Empirical Study on Di?erential Evolution for Optimal Power Allocation in WSNs GONG Wen-yin , CAI Zhi-hua School of Computer Science, China University of Geosciences, Wuhan, 430074 Abstract: Optimal power allocation (OPA) is considered to be one of the key issues in designing a wireless sensor network (WSN). Generally, the OPA in WSN can be formulated as a numerical optimization problem with constraints. Di?erential evolution (DE) has been proved to be a very e?cient evolutionary algorithm for global numerical optimization. Based on this consideration, in this paper, an empirical study is conducted to evaluate the performance of DE for the OPA. Di?erent mutation strategies and di?erent parameter settings are used in DE to test the in?uence of them when optimizing the OPA. The sensors in WSN are scaled up to 200. Experimental results indicate that the parameters of CR and F are sensitive to the performance for the OPA. In addition, with the same parameter setting, the strategy selection is stable in DE. Key words: Di?erential evolution; wireless sensor network; optimal power allocation; empirical study. 基 金 项 目： Research Fund for the Doctoral Program of Higher Education Foundation (20090145110007 and 20110145120009)，National Natural Science Foundation of China Foundation 作者简介： Correspondence author (cu)：GONG Wen-yin（1979-），male，lecturer，major research direction：evolutionary algorithms and their applications. CAI Zhi-hua（1964-），male，professor，major research direction：data mining, machine learning, and evolutionary algorithms. -1- 0 Introduction In recent years, wireless sensor networks (WSNs) have gai