基于粒子群优化的无线传感器网络定位算法研究

发布时间：2018-07-10 20:13

本文选题：无线传感器网络 + 定位算法　；参考：《昆明理工大学》2015年硕士论文

【摘要】：节点定位技术是无线传感器网络最主要的支撑技术之一,也是研究难点之一在大部分实际应用中,获取精确的节点绝对位置或者节点间的相对位置信息是至关重要的。为了达到精确定位的目的,不仅要从硬件上减小节点间距离估计误差或者测距误差,而且要从定位算法上进行改进,提高算法的定位精度。本文首先讲述了无线传感器网络技术的发展历程,总结了国内外研究现状。研究比较了无线传感器网络节点间测距技术及基于测距的定位算法。最后对粒子群优化算法进行研究和改进,在基于测距的定位模型下对改进算法进行仿真分析。本文对基于改进粒子群优化的无线传感器网络定位算法进行研究。主要的研究工作有以下两点：首先,在有锚节点无线传感器网络中,针对传统RSSI测距模型的缺点,本文采用了一种改进RSSI测距模型,并提出了一种基于线性递减权重的混沌粒子群算法(W-CLSPSO)的定位方法。对基于锚节点选择策略的W-CLSPSO算法与不基于锚节点选择策略的W-CLSPSO算法进行比较,证明了锚节点选择策略在本文仿真实验中的可行性。对W-CLSPSO算法、CLSPSO算法(混沌粒子群算法)、PSO算法、WLS算法(加权最小二乘法)分别在不同噪声指数和锚节点数量情况下的仿真比较,仿真结果说明在定位精度和算法稳定性上,W-CLSPSO算法要优于另外三种算法。其次,针对锚节点缺失的无线传感器网络,本文提出了一种基于TDOA定位模型的改进PSO算法与Taylor算法协同定位的新定位方法。在不同噪声指数下,分别对PSO算法、AsyLnCPSO算法(基于异步学习因子的粒子群优化算法)、SAAPSO算法(基于异步学习因子的自适应权重粒子群优化算法)和Min-max+Taylor算法、SAAPSO算法、SAAPSO+Taylor算法的平均位置误差和均方差两项指标进行仿真比较,仿真结果说明本文提出的改进算法具有更高的定位精度、更小的累积误差和更好的稳定性。
[Abstract]:Node location is one of the most important supporting techniques in wireless sensor networks, and it is also one of the difficulties in most practical applications. In most practical applications, it is very important to obtain accurate absolute position of nodes or relative position information between nodes. In order to achieve the goal of accurate location, not only the distance estimation error or ranging error between nodes should be reduced in hardware, but also the location algorithm should be improved to improve the accuracy of the algorithm. This paper first describes the development of wireless sensor network technology and summarizes the current research situation at home and abroad. This paper studies and compares the ranging technology between nodes in wireless sensor networks and the location algorithm based on ranging. Finally, the particle swarm optimization algorithm is studied and improved, and the improved algorithm is simulated under the location model based on ranging. In this paper, the location algorithm of wireless sensor networks based on improved particle swarm optimization (PSO) is studied. The main research work is as follows: firstly, in wireless sensor networks with anchor nodes, an improved RSSI ranging model is adopted to overcome the shortcomings of the traditional RSSI ranging model. A chaotic particle swarm optimization algorithm (W-CLSPSO) based on linear decreasing weight is proposed. The W-CLSPSO algorithm based on the anchor node selection strategy is compared with the W-CLSPSO algorithm which is not based on the anchor node selection strategy. The feasibility of the anchor node selection strategy in the simulation experiment is proved. The W-CLSPSO (chaotic Particle Swarm Optimization) algorithm and WLS (weighted least squares) algorithm are compared with each other under different noise exponents and the number of anchor nodes. The simulation results show that the W-CLSPSO algorithm is superior to the other three algorithms in the accuracy and stability of the algorithm. Secondly, for wireless sensor networks with missing anchor nodes, this paper proposes a new co-localization method based on TDOA localization model, which is based on improved PSO algorithm and Taylor algorithm. At different noise numbers, The average position error of PSO algorithm (Asynchronous learning factor based particle swarm optimization algorithm) and Min-max Taylor algorithm (adaptive weighted particle swarm optimization algorithm based on asynchronous learning factor) and Min-max Taylor algorithm respectively. The difference and mean variance were compared by simulation. The simulation results show that the improved algorithm has higher positioning accuracy, smaller cumulative error and better stability.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN929.5;TP212.9

【参考文献】