Research On Node Localization And Tracking Algorithm For WSN

Wireless sensor networks(WSN) node localization is exchanging information between the unknown node and beacon nodes, to determine the location coordinates of unknown nodes through some location and tracking algorithm, because it has value only when the network node data acquisition and location are bound together. So the study of wireless sensor network node location and tracking technology has great practical significance.The paper analyzes the present situation of research in wireless sensor node localization and target tracking in domestic and foreign, citing some ranging and localization and target tracking algorithms commonly used in WSN, and introduces the Kalman filter(KF) theory and Monte Carlo(MCL) principle, to provide the basis for targeting and tracking theory.When the density of beacon node is large, an adaptive tracking beacon structure scheduling policy is presented in the paper, it effectively reduces the energy consumption; An improved Monte Carlo algorithm(Improved MCL) is used in the paper,it combines MCL beacon box algorithm with a node motion and prediction model based on Newton interpolation algorithm, reducing the sample range; using weighted centroid localization algorithm with special weight to improve positioning accuracy. Simulation results show that adaptive tracking beacons uses half the energy consumption than static clustering beacon and the accuracy of improved MCL algorithm improves 0.5m than beacon box algorithm.When the density of beacon node is small, an adaptive mobile beacon path planning algorithm is established in the paper. It combines scan path planning with virtual force path way to make mobile beacon adjust moving distance and direction in real time by the density of unknown size and achieves higher coverage; The extended Kalman filter(EKF) algorithm is used to optimize the ranging results and it improves ranging stability. Using Taylor expansion weighted least squares positioning algorithm to gradually approaching the solution of equations and obtain higher accuracy. The simulation results show that the coverage of adaptive path up to 86% and the accuracy of Taylor expansion weighted positioning algorithm improves 0.4m than multilateral positioning algorithm.