Research On Key Technologies Of Cooperative Localization In Under-Location Environment

Location services play an important role in many fields such as smart cities,smart factories,and public safety.Traditional non-cooperative wireless positioning technology is prone to lack of information in a complex environment,resulting in poor positioning accuracy or even inability to locate,and it is difficult to provide high-precision and highly available positioning information.The emergence of cooperative localization technology provides new solutions for high-precision and high-reliability positioning in complex environments,and has become a research hotspot in the field of positioning and navigation.Aiming at the main problems of co-location in the under-location environment,this thesis deeply researches key technologies such as coordinated compensation of ranging errors based on spatial correlation,reference node selection based on fuzzy comprehensive evaluation,and time-location coordinated estimation based on gaussian mixture variational transfer.,carried out method innovation and engineering practice,the main research results and innovation points include:1.Aiming at the problem of low ranging accuracy due to factors such as multipath effects under the condition of lack of prior environmental information,the spatial correlation characteristics of multipath errors are analyzed based on the signal indoor delay distribution model,and the coordination of ranging errors based on spatial correlation is proposed.Compensation method,based on the empirical mode decomposition method to decompose the non-stationary ranging information,use the statistical results of the ranging error between neighboring nodes to predict the ranging error of the node to be located,compensate the ranging result,and achieve high accuracy in the indoor complex environment ranging.The test results show that compared with the mainstream received signal strength-signal arrival time fusion ranging compensation method and the ranging compensation method based on the linear fitting of the previous data,the ranging error of the proposed method is reduced by 14%and 23%,respectively.2.Aiming at the problem of poor positioning effect caused by inaccurate location information of a large number of pseudo-anchor nodes in co-location,this thesis first derives the lower bound of the distributed Cramer-Rao lower bound of joint location-time estimation,and analyzes the number of nodes and the accuracy of pseudo-anchor node location information.The influence of multi-dimensional feature information such as degree,distance measurement value on positioning accuracy is proposed.A node selection method based on fuzzy comprehensive evaluation is proposed,a multi-feature joint evaluation strategy based on fuzzy degree membership is constructed,and a node selection framework based on fuzzy comprehensive evaluation is established.Optimizing the positioning nodes effectively improves the positioning accuracy of the system.Experimental results show that,compared with the method based on geometric precision factor and the node selection method based on distance relationship,the positioning accuracy of the proposed method is improved by 12%and 17%,respectively.3.Aiming at the problem of location accuracy degradation caused by the incomplete expression of location features in the message-passing colocation method,the theoretical basis and model of the variational message-passing method are analyzed in detail,and the location information loss caused by probabilistic model fitting is analyzed in detail.The impact of accuracy,the message transfer function based on the gaussian mixture model is derived,the accuracy of location feature expression is improved based on multi-variation information,the joint maximum posterior probability solution based on multi-node coordination is completed,and the limited resources are realized.Accuracy locationtime joint estimation.Experimental results show that the positioning accuracy of the proposed method is improved by 24%and 8%,respectively,compared with the mainstream message passing method based on confidence and non-parametric cooperative positioning method.4.Designed and built a collaborative positioning integrated simulation platform and physical verification system,and incorporated the ranging error compensation method,positioning reference node optimization method,collaborative positioning message transmission method and solution method proposed in this thesis into a unified positioning process and architecture.The theoretical method innovation proposed in this thesis has been integrated testing and experimental verification.The experimental results show that the cooperative localization method proposed in this thesis improves the positioning accuracy by about 33%compared with the message passing method based on confidence,and the positioning accuracy reaches 0.62 meters(1?).