| As a major research area,the Wireless Sensor Network(WSN)has been widely used with the development of computer technology,communication network and embedded systems.Wireless sensor network put massive sensor nodes into the target area randomly and supervise the target in the way of self-organization.Those nodes can sense and collect the information of the target,and send those data to users.Node location technology is the base of the application of WSN such as supervision,prediction and recognition.Only a part of the nodes are chosen to install GPS module and set those node as anchor nodes,the other nodes must use the algorithm with high self-organization capability,high connectivity and high location accuracy to determine its location.In this paper,range-based and range-free localization algorithm is mainly discussed.Range-based algorithms measure the information between nodes such as the angle or distance,the localization accuracy is higher than range-free algorithm,and can satisfy the demand of most application.So,the RSSI WSN localization algorithm is deeply studied and some improvements are made on this algorithm because of its large error in the process of ranging and locating the sensor nodes.Firstly,the improved algorithm establish the Kalman filter model and smooth the RSSI signal value use this model to make the received signal strength value close to its real value;And then the reciprocal of the sum of distance between the adjacent anchor nodes is chosen as a weighting factor,and this factor is modified by the measured distance between two nodes at the same time,triangular weighted centroid algorithm is used to locate the coordinate of the unknown nodes.Finally,the average localization error is calculated by locating the adjacent anchor nodes and then used to compensate the coordinate of unknown nodes.The simulation result shows the introduced algorithm can improve the localization accuracy by 5%-6% comparing to RSSI algorithm. |
PDF Full Text Request