The Research On Motion Node Accurate Location Of Wireless Sensor Network

Recently, as the most basic techonology of getting information: sensor technology, is paid great attention by academic world and industrial circles due to it's significant practical value. Sensor technology, networking and wirless communications promote emergence and development of Wireless Sensor Network(WSN), and WSN expand people's ability to get information from object world and it turns physical information to data. It will provide people with the most realistic and most direct and most effective informat in next generation networks.Paper gives an overview of WSN first, and then the position of wireless sensor is introduced, classified summary of the various positioning algorithm in recent years and representative of the application of the WSN, and analysed the various algorithms advantages and disadvantages. With the development of wireless sensor networks and wireless sensor networks in people's lives and production applications, the previous location algorithm is no longer able to meet people's needs, people need a higher positioning accuracy; in the past, most localization algorithm is static. Localization algorithm for mobile nodes is more and more needed.For the problem of low location algorithm, In the range-based localization algorithm based on RSSI, using BP neural network to improve the localization accuracy of the RSSI algorithm. For the need of localization algorithm for mobile node, in the range-free algorithm, Localization Algorithm for Mobile Nodes in Wireless Sensor Networks Based on Signal Strength and Motion Vector (named as SSMV). Unknown nodes'activities area deployed four anchor nodes, according to the node's signal strength and communication radius to count the approximate area where the unknown node, and then the moving unknown nodes change the motion verctor, which can calculate the othere approximate area. Finally, two polygons intersect to form the new region, and the center is the location of unknown node estimate location.Two kinds of algorithms are simulate in Matlab. Experimental data show that the algorithm based on BP neural network increase low error than Least Square Estimation method, when the distance measurement error increases. In the case of large ranging errors, BP advantages become more pronounced. SSMV algorithm, whether a single node or multiple nodes, location errors are much smaller than the convex programming algorithm. The location of mobile node in the node for the actual project location are very important reference value.