Simulation Research On NLOS Error Recognition,suppression And Location Algorithms Of Self-organizing Network Nodes

The future society is a networked and intelligent society,which is an indispensable trend for technological and social development.In the field of navigation and positioning,the networked perception of time-space information reference sources and the autonomous acquisition of time-space information have become an important manifestation of networked and intelligent radio navigation and positioning.Selforganizing network has unique advantages such as autonomous networking,automatic perception and non-destructive access and exit of network nodes.It is a typical representative of intelligent technology in wireless sensor networks.At present,self-organizing networks are mainly used for non-centralized networking and communication between nodes.However,from the perspective of autonomous navigation and positioning between network nodes,there is still a lack of systematic research.The main problems currently facing include that,for the communication services between network nodes,the existing channel model can meet the requirements of communication quality;But for the positioning service between network nodes,the current self-organizing network channel model cannot identify lineof-sight and non-line-of-sight,let alone suppress non-line-of-sight,which will lead to large deviation between the ranging values.This seriously affects the positioning accuracy.In the background of wireless sensor network and on the basis of wireless positioning theory,the paper discusses how to locate the nodes of self-organizing network in the NLOS propagation environment,focusing on the identification and suppression of NLOS signal and locating nodes.The main research contents and results of this paper are as follows:(1)Based on theory of wireless positioning,this paper studies positioning method model and parameter estimation method based on TOA in wireless sensor network and analyzes the influence of NLOS error on positioning accuracy and four distributions of error.(2)Research the NLOS signal recognition algorithm among self-organizing network nodes.It focuses on the principle of Wylie method and signal detection and recognition method in the error recognition method based on signal arrival parameters.The simulation results show that both can effectively identify NLOS error,but both lack real-time performance and are suitable for single anchor node positioning.Therefore,the recognition method based on polynomial fitting and virtual point density are studied.The focus is on the identification method based on the virtual point density.The analysis results show that the method can effectively identify error and has strong real-time performance in a scene with a bad propagation environment.(3)Research on the NLOS signal suppression algorithm between self-organizing network nodes.The NLOS signal suppression method is systematically analyzed.Aiming at the problem of high algorithm complexity when the number of anchor nodes increases in the existing weighted suppression methods,the minimum residual weighting method is studied.This method suppresses the influence of NLOS signal by selecting the anchor node combination with the smallest residual.Simulation results show that the algorithm can reduce the complexity of existing algorithms and can effectively suppress NLOS signal.(4)Research on positioning technology between nodes in self-organizing network.Based on the recognition and suppression results of NLOS signal,when the observed and measured error conforms to the Gaussian distribution,the Kalman filter algorithm based on motion pattern recognition is studied.The algorithm obtains the positioning results by establishing Kalman filters in different states and weighting the filtering results.The simulation results show that the positioning accuracy of this algorithm is23% higher in the x direction and 16% higher in the y direction than the traditional Kalman filter algorithm.When the observation and measurement error distribution is non-Gaussian,in light of NLOS error,the research considers particle filter algorithm based on residual error.Simulation results show that this algorithm can effectively suppress NLOS error and predict the trajectory of mobile nodes.