A Study Of Mobile Noed Localization Algorithm For Wireless Sensor Network

Wireless Sensor Network(WSN) is being paid more attentions as a new type of wireless technology in recent years. It is widely used in military, transportation, logistics, environmental protection and other fields and has many applications. Automatic Location of the sensor nodes has important significance as one of the key technologies of WSN. Most of the existing algorithm are mainly for static WSN and the sensor nodes' location is fixed. But, the nodes are always moving due to the impact of external environment outside. This article analyses some typical algorithms and points out the shortcomings on mobile nodes' locations. These algorithms do not consider the mobile nodes.Monte Carlo localization(MCL) is designed for mobile nodes specifically, using the information about the node movement effectively. However, the following deficiencies still exist: (1) The communication's radius will be consistent. But it is difficult to be consistent due to limited resources itself and the changes of height outside. (2) The sample area is too large in MCL, as result the accuracy decreases. The RBDMCL algorithm is proposed in this paper for the shortcomings of MCL. The RBDMCL is based on the distance, so it is a good solution to the problem of node communication radius inconsistent. The RBDMCL analyses the movement of nodes and constructs the model of movement using Newton interpolation polynomial. The algorithm reduces the sample areas and improves location accuracy through the direction of the nodes' movement next time. Finally, the RBDMCL algorithm is simulated. It is clear that RBDMCL has higher accuracy and smaller errors by analysing and comparing the results. The RBDMCL is lack of the cost for communications.