Mobile Wireless Sensor Network Self-localization Algorithm

As a new information acquisition and processing technology, wireless sensor networks have broad application prospects, and many applications are based on node location information, the nodes positioning technologies have become one of the supporting technology. Therefore, the node location technology for wireless sensor networks research has important theoretical significance and application value.Monte Carlo localization algorithm was used for robot positioning first, it Also can be introduced to WSN. The Characteristics is that this algorithm take advantage of mobility to improve algorithm accuracy and also can reduce network positioning cost, it's a effective instrument to get the location of the mobile node.This paper is mainly to study Mobile WSN node localization based on Monte Carlo algorithm. After the profound analysis of the classic positioning system, proposed two methods of localization algorithm based on Monte Carlo for mobile nodes:Relationship Accurate Monte Carlo Localization (RAMCL) and Monte Carlo Accurate localization Boxed(MCAB). Both algorithms take advantage of the communication relationship with the neighbor node , build the sample space, to improve the positioning accuracy. RAMCL is for the localization of underground mobile mine staff, by analyzing the features of people's movement, using Newton interpolation formula to predict motion direction. Then build efficient sample space, improve the extraction efficiency, optimize location selection model. MCAB proposed in this paper can be applied in the open space to localize mobile nodes. It use RSSI model to build the probability model of mobile node, then filter the samples to ensure the sample confidence. Simulation experiment result indicated that, under the condition of same bacon density,node density and node speed, RAMCL and MCAB have a better positioning precision than MCL,MCB.