Research On The Localization Technology Of Wireless Sensor Networks

ABSTRACT:Wireless sensor network is one of the research hotspots of current information technology,and it has been used more and more widely in military and civil fields while developing.Node localization technology is crucial to many applications in wireless sensor networks.With the diversification of wireless sensor network applications and the gradual maturity of localization technology in static network,mobile node localization technology is receiving more and more attentions in recent years.Taking mobile wireless sensor network as a research object and by means of theoretical analysis and simulation experiments,this dissertation is committed to making a more in-depth study of node localization algorithms in wireless sensor networks and trying to improve the efficiency and accuracy of the localization algorithm.In this thesis,current research progress of localization technology in wireless sensor networks is discussed and the existing positioning technology and its characteristics are introduced from aspects of static network and mobile network respectively.Range-based and range-free localization technologies are analyzed at the same time.Focusing on the range-free Monte Carlo localization algorithm,we analyze the problems of low sampling efficiency and improper sampling weights.The improved schemes for the problems are proposed as well.Focusing on the problem of low sampling efficiency of Monte Carlo localization algorithm,an improved Monte Carlo localization algorithm based on historical-anchor ranging is proposed.The improved scheme studies on the sampling process of the algorithm.Combining information of historical anchors and the RSSI-based ranging algorithm,sampling area of the node to be located is further constrained in order to improve the efficiency and accuracy of node sampling.The simulation results show that the improved Monte Carlo method based on historical-anchor ranging can increase the localization accuracy and reduce the attempts of nodes' position sampling.In addition,this thesis proposes an improved Monte Carlo localization algorithm based on ranging weights aiming at solving the problem of improper sampling weights and impractical motion model.Based on the improved Monte Carlo algorithm mentioned above,the RSSI ranging of the anchors and the neighbor nodes is used to provide references for the position sampling weights of the nodes to be located,and the improved motion model is used to further constrain the sampling area in direction.Simulation results show that the Monte Carlo method based on ranging weights further improves the localization accuracy and efficiency.