Research On Localization Optimization Algorithm In Wireless Sensor Network

Wireless sensor network, which has been made viable by the convergence of micro-electro-mechanical systems technology, wireless communications and digital electronics, has received increasing attentions in recent years, for wild applications in many fields. For most applications, data without location information has limited use. Localization has become one of the indispensable supporting techniques in wireless sensor networks. Undoubtedly, the Global Positioning System (GPS) is the most well-known location service in use today. The approach taken by GPS, however, is unsuitable for low-cost, limited power, ad-hoc sensor networks since GPS is based on extensive infrastructure. In wireless sensor network, the basic approach of localization is to use anchor nodes that aware of the positions as references, and other nodes compute their positions through the relationship of distance, angle, or network connectivity to the anchor nodes.This paper focus on the localization optimization methods for WSN, aims to enhance performance of nodes localization ability, to provide reliable location information for other applications. Meanwhile, the optimization methods are both distributed algorithms, so lower energy consumption of position system.The paper fulfills the following work:First, we propose Particle Swarm Optimization (PSO) algorithm to estimate nodes locations, which uses the principle of evolvement and global searching characteristic to find nodes locations. Compared to maximum likelihood estimation method, the simulation results show that PSO method improves the localization accuracy obviously under different ranging errors, as the same time, causing no more communication cost.Second, we propose other localizations system based on GPSO algorithm, which incorporates the advantages of the excellent global searching of the PSO and the local speedy convergence of the gradient method. Experimental simulations show that this proposed method plays better performance than PSO as ranging error increasing, in location precision, convergence speed and robust performance.