Research On Ultra-wideband Indoor Positioning Algorithm Based On Ranging

With the advent of the Internet of Things' era,related technologies based on users' location information have been greatly developed and widely used.Location Based Service(LBS)has gradually become a basic service demand in people's work and life.Especially in some complex large indoor venues,people are in urgent need of high-precision location services.Driven by the high demand for location services,indoor positioning technology has been placed at a high-speed developing stage in recent years,and many indoor positioning technologies have emerged.Among them,Ultra-wideband(UWB)has received extensive attention in the field of indoor positioning due to its unique advantages.Therefore,it is important to study the indoor positioning algorithm based on UWB.This paper research the UWB positioning algorithm based on ranging.The outline of this work is as follows:(1)This paper introduces the basic principles of UWB positioning technology,describes the characteristics of UWB channel and simulates and analyzes the standard channel model of IEEE802.15.4a.Based on the channel model,the UWB ranging technology is introduced and different influencing factors contributing to UWB ranging error are analyzed.The article points out that Non Line of Sight(NLOS)is the main reason for the increase of UWB ranging error.According to different implementation principles,the UWB positioning algorithms are divided into two categories: algebraic solution method and fingerprint matching method.The advantages and disadvantages of these two algorithms and their applicable scenarios are compared and analyzed.Then this paper will focus on how to improve these two traditional algorithms.(2)This paper introduces the principle of TDOA positioning algorithm and points out two major deficiencies of the traditional TDOA algorithm.First,the traditional algorithm has a severely degraded positioning accuracy under the condition of large measurement noise,and the robustness is poor.Second,the traditional algorithm assumes that the base station position is accurately known.The base station position error is not considered.In order to solve the above two problems,this paper improves the TDOA algorithm and proposes an iterative localization algorithm based on convex optimization,which effectively solves the shortcomings of the traditional TDOA algorithm,considering both the base station position error and the robustness of the algorithm.Even when the measurement error becomes larger,the performance of the algorithm can still approach the Cramer-Rao Low Bound(CRLB).(3)When using the traditional RSSI-based fingerprint localization algorithm,the RSSI value fluctuates greatly and the fingerprint information is unreliable,resulting in insufficient positioning accuracy.This paper proposes an improved scheme: using the ranging value as the fingerprint amount to construct a fingerprint library,which is more reliable than the traditional fingerprint library.In the process of positioning realization,this paper proposes a fingerprint localization algorithm combining MeanShift and weighted K-nearest neighbors,and carries out experiment in actual scenes.The experimental results show that the average error of algorithm positioning in the LOS and NLOS environments is 0.5077 m and 0.4970 m,respectively.Compared with the traditional K-means algorithm,the proposed fingerprint localization algorithm has higher positioning accuracy and stronger robustness.