Research On TOA Wireless Location Algorithm Based On NLOS Error Identification And Mitigation

With the speedy development of wireless communication network and Internet of Things(IoT),location-based service(LBS)and its applications are becoming more and more popular.Whether people are indoors or outdoors,the requirements for positioning accuracy are getting higher and higher.However,there are many obstacles in the intensive urban architecture and complex indoor environment,resulting in serious damage to the positioning accuracy.Among them,the most serious one is non-line of sight(NLOS)interference.So,how to effectively identify NLOS errors and eliminate them has become a hot topic of current research.This paper focuses on the NLOS error identification and elimination algorithms,which first identify the NLOS signals and then eliminate them.Since the NLOS signals are recognized and discarded,and only the line-of-sight(LOS)signals are used for positioning,the positioning accuracy will be greatly improved.Through the review and research of the literature,this paper analyzes the current development situation of NLOS identification and mitigation technologies.And this paper also summarizes the domestic and international research of NLOS elimination in mixed LOS/NLOS environment that aims to improve the positioning accuracy.The NLOS error elimination technology can be divided into two groups:the first group directly suppresses the NLOS error.The second group first identifies the NLOS signals,then discards them and only uses the LOS signals for positioning.The specific work of this paper is as follows:Firstly,this paper introduces the NLOS error propagation model and uses the residual weighting algorithm and its improved algorithm to analyze the influence of NLOS error on the positioning accuracy.Through the comparative simulation analysis of the traditional RWGH algorithm and its improved algorithms,it is found that although the three algorithms can alleviate the NLOS error to some extent,they cannot achieve a good mitigation effect in the case of more NLOS base stations and larger NLOS error.Secondly,the NLOS identification algorithms based on signal parameters and signal detection method is discussed in this paper.The signal feature-based likelihood ratio test identification algorithm and non-parametric machine learning identification algorithm for UWB positioning system that are widely used in indoor scenes and the NLOS identification algorithms based on maximum likelihood and Neyman-Person criterion that are applied in wider range of scenes are studied.The simulation results show that these algorithms can effectively identify the NLOS signals in specific scenarios.Then,this paper uses Cayley-Menger determinant to design a NLOS identification and mitigation algorithm in a wireless sensor network positioning system,referred to as CNIM.Firstly,all anchors are grouped and NLOS detection algorithm is carried out for each group to detect whether there exist NLOS signals in the combination of anchors.Then the specific LOS and NLOS anchors are identified by traversing the detection results of all groups.Finally,only use LOS anchors to estimate the sensors location in combination with LS/RWGH localization algorithms.Simulations are conducted to investigate the performance of CNIM in comparison with other conventional algorithms.The simulation results demonstrate that the proposed algorithm has good anti-NLOS characteristic and can maintain a good positioning accuracy under mixed LOS/NLOS environment.At last,this paper summarizes the research content and points out the inadequacies and improvements.