【作者】 张浩；

【导师】 裴炳南；

【作者基本信息】 大连大学 ， 计算机科学与技术， 2015， 硕士

【摘要】 无线传感器网络(Wireless Sensor Network,WSN)是通过将微型的传感器节点部署在监测区域,这些节点通过无线通信的方式形成多跳的自组织的网络,用来感知、采集和处理监测区域中对象的信息。而监测信息中的位置信息至关重要,事件发生的位置或获取信息的节点位置是节点监测信息中所包含的重要信息,没有位置信息的监测信息往往是没有意义的。因此,节点定位技术的研究,对WSN的应用起着极其关键的作用。而DV-Hop定位算法是与距离无关的一种算法中最常用的算法,本文主要针对DV-Hop算法进行一些研究和改进。本文针对DV-Hop定位算法所存在的问题,研究了将萤火虫算法(Firefly Algorithm,FA)、压缩感知(Compressed Sensing,CS)理论应用到DV-Hop定位中,以提高其定位性能的若干问题。本文的工作归纳起来可以分为以下几个方面:(1)介绍无线传感器网络定位算法的研究现状,然后介绍了定位算法的基础理论,包括定位算法的基本原理及其分类,最后分别介绍了基于距离和距离无关的两类算法中常见的几种定位算法,并详细的介绍了距离无关的DV-Hop定位算法。(2)定位问题本质上是一个基于距离测量值或路径测量值的优化问题,因此,近年来有研究者基于遗传算法(Genetic algorithm,GA)、粒子群算法(Particle Swarm Optimization,PSO)对DV-Hop算法进行了优化,并取得了一定的成果。而最近提出的萤火虫算法在寻找全局最优解上有着很高的效率,要比GA和PSO有效,在解决优化问题上具有巨大的潜力。因此,在分析了最小二乘法定位的误差来源的基础上,提出了一种基于萤火虫算法优化的DV-Hop定位算法(FDV-Hop)。建立所有的未知节点的定位误差之和最小的数学模型,利用FA算法求解模型的最优解,从而得到节点的最优位置估计,最后仿真对比了FDV-Hop和传统的DV-Hop算法的定位性能。(3)近年来,压缩感知理论是极为热门的研究前沿,在许多领域都引起了瞩目。在一定条件下,CS理论能够以远低于采样定理的采样率对信号进行采样,这将有望解决节点定位中大量数据传输的问题。因此,首先介绍了压缩感知基本理论,并分析了压缩感知理论的三个研究重点,同时阐述了其在节点定位中应用的理论可行性。然后提出了一种基于压缩感知的DV-Hop定位算法(CS DV-Hop),构建测量矩阵和网络模型,并利用已有的重构算法进行重构,最后通过仿真实验和理论分析来对该算法的性能进行分析。更多还原

【Abstract】 Wireless sensor network(WSN) is a multiple hops self-organizing network, which is formed by the wireless communication of some sensor nodes deployed in the detection region.It used to perception 、 acquisition and processing the information of the object in the monitoring area. And the location information in the monitoring information is crucial, the monitoring information tend to be meaningless without location information. Therefore, the research on node location technology is crucial to the application of WSN. The DV-Hop algorithm is a very typical means of WSN without need of ranging.This thesis aiming at the existing problem of DV-Hop localization algorithm, applies the the firefly algorithm(FA) and compressed sensing(CS) theory to location problem to improve the location performance. In a sum, the work of this thesis can be divided into the following aspects:(1) The thesis introduced the research status of localization algorithm for wireless sensor network. And then introduced the theoretical foundation of the localization algorithm including the basic principles and the classification. Finally the thesis introduced several common localization algorithm based on the distance and range-free respectively, at the same time, detailed introduced the DV-Hop algorithm.(2) The problem of location is essentially a optimization problem based on the distance or the measured values of path, therefore, some researchers used genetic algorithms(GA),particle swarm optimization(PSO) to optimized the DV-Hop algorithm for a better position accuracy. The experimental results show that the firefly algorithm is more effective, higher success rate than GA and PSO in the aspect of finding global optimal solution, thus having great potential in solving optimization problem. Therefore, this thesis uses FA to optimize DV-Hop algorithm to get better positioning accuracy. The positioning error of the least-square method was analyzed. And then the thesis introduced the firefly algorithm and proposed a firefly algorithm optimization based DV-Hop localization algorithm(FDV-Hop).Then the mathematical model was built with the smallest error for all of unknown nodes, and the firefly algorithm was used to obtain the optimal solution of the model. And the thesis compared the simulation results of FDV-Hop and the traditional DV-Hop at last.(3) In recent years, the compressed sensing theory is the very popular research front. And it causes attention in many fields. The CS theory can samples the signal with the sampling rate far below the sampling theorem under certain conditions. This is supposed to solve thelarge amounts of data transmission problems in the process of node location. Therefore, the thesis introduced the basic theory of compressed sensing. Then the three research emphases of compressed sensing theory were analyzed. At the same time the thesis expounded the theoretical feasibility of its application in node localization. And then the compressed sensing based DV-Hop location algorithm(CS DV-Hop) was proposed, which including constructs the measurement matrix and network model, and used the existing reconstruction algorithm to obtain the the position of the node. At last the simulation experiment and theoretical analysis was present to analyze the performance of the proposed algorithm.更多还原