基于层次混合的高效概率包标记WSNs节点定位算法

发布时间：2018-06-19 22:16

本文选题：无线传感器网络(WSN) + 概率包标记(PPM)　；参考：《电子与信息学报》2014年02期

【摘要】：在利用概率包标记技术对无线传感器网络(WSN)恶意节点的追踪定位中,标记概率的确定是关键,直接影响到算法的收敛性,最弱链,节点负担等方面。该文分析并指出了基本概率包标记(BPPM)和等概率包标记(EPPM)方法的缺点,提出了一种层次式混合概率包标记(LMPPM)算法,可以克服以上算法的不足。该算法对无线传感器网络进行分簇,将每个簇看成一个大的"簇节点",整个网络由一些大的"簇节点"构成,每个"簇节点"内部又包含一定数量的传感器节点。在"簇节点"之间采用等概率包标记法,在"簇节点"内部采用基本概率包标记法。实验分析表明,该算法在收敛性、最弱链方面优于BPPM算法,在节点计算与存储负担方面优于EPPM算法,是在资源约束条件下的一种整体优化。
[Abstract]:In tracking and locating malicious nodes in wireless sensor networks (WSNs) using probabilistic packet marking technique, the determination of marking probability is the key, which directly affects the convergence of the algorithm, the weakest chain, the burden of nodes and so on. This paper analyzes and points out the shortcomings of the basic probability packet marking (BPPM) and the equal probability packet marking (EPPM) methods, and proposes a hierarchical mixed probability packet tagging (LMPPM) algorithm, which can overcome the shortcomings of the above algorithms. In this algorithm, the wireless sensor networks are clustered and each cluster is regarded as a large "cluster node". The whole network is composed of some large "cluster nodes", and each "cluster node" contains a certain number of sensor nodes. The equal-probability packet marking method is used between "cluster nodes" and the basic probability packet marking method is used in "cluster nodes". The experimental results show that the proposed algorithm is superior to BPPM in convergence and weakest chain, and is superior to EPPM in node computation and storage burden. It is a global optimization under resource constraints.
【作者单位】：皖西学院信息工程学院;中国科学技术大学计算机科学与技术学院;解放军陆军军官学院六系;
【基金】：国家自然科学基金青年科学基金(61303209,61302179) 安徽省高等学校省级自然科学研究重点项目(KJ2013A255) 六安市定向委托皖西学院产学研合作项目(2012LWA015)资助课题
【分类号】：TP212.9;TN929.5

【参考文献】