As a new type of self-managed data collection network, the wireless sensor network (WSN) integrates the ability of wireless communication, data collection and information disposal. For many application of wireless sensor network, location information is so important that it determine whether the imformation which captured by sensor nodes can be used. So, the technique of nodes localization in wireless sensor network plays an important role. Based on the location imformation of some node, the technique of nodes localization utilizes some schemes to calculate the location imformation of unknown nodes.This study is supported by the project coorperated with a corporation. We research ZigBee technique, IEEE 802.15.4 standard, and localization techniques in wireless sensor networks. Based on these work, a improved localization algorithm is provided, and my major work is as follows.Based on comprehensive analysis of existing localization algorithm in wireless sensor network , this dissertation summaries characteristics of existing localization algorithm, and find out the essence of those algorithm, finally propose a new way to how to classify those algorithm.The existing localization algorithms are simulated and the shortcomings of these algorithms are analyzed. On this basis, a composite localization algorithm is proposed. By comparing the simulation results, it's found that the composite localization algorithm can improve effectively the localization accuracy of nodes.The random distribution of node will produce uncertain factors,which will reduce the localization accuracy of nodes. To solve this problem, a mobile beacon node based on the virtual force model is proposed. The performance of the model has been verified by simulation,and it can improve localization accuracy of composite localization algorithm. |