Research On Localization Algorithm For Wireless Sensor Networks' Node

In wireless sensor networks, the position of the sensors is very important for monitored objects, so determining the positions of sensors is a key problem in many applications. Based on the analysis of several classical algorithms, this thesis focuses on the investigation of distance error, localization coverage and localization error in DV-hop algorithm and centroid algorithm, and then an improved method has been proposed. To reduce the jump frequency of the CC2431 node, a limiting range and eliminating shake filtering method has been employed.Firstly, to solve the problem of the positioning accuracy over depending on the distribution of anchor nodes in centroid algorithm, a weighted-centroid localization algorithm with iterative refinement by multi-nodes collaboration localization algorithm in WSNs is proposed. In this algorithm, weighted-centroid algorithm is used to initialize the node coordinates by requesting 2-hop anchor nodes to increase available anchor nodes, and in order to improve the positioning accuracy, a iteration method of adjusting nodes'coordinates several times by means of calculating the distance between actual coordinates and the estimate coordinates is introduced. The simulation results show that the algorithm has better estimation error and node localization coverage than common weighted-centroid algorithm.Secondly, a multilateral localization algorithm based on beacon selection in WSNs has been proposed to solve the large error of the estimated distance between nodes affected by the node density with DV-HOP algorithm. The algorithm estimates the single hop distance with the node distribution density as parameter, and filters out some anchor nodes with more distance error. The simulation results show that the algorithm improved the node positioning accuracy, and significantly reduced the computational nodes.At the end, limiting range and eliminating shake filtering method has been used in the thesis to optimize the RSSI value of the signal CC2431 node received from CC2430, what was unstable caused by electromagnetic interference, obstacles block, the movement of persons and so on in complicated environment. The experiment results show that this method reduces the jump frequency of the location position and some improvement has been shown in the location error of the unknown nodes.