Test And Optimization Of NLOS Error Suppression Algorithm For UWB Positioning

With the advent of the Internet of Things era,people have higher and higher requirements for the accuracy of location information.With the improvement of positioning technology and short-distance and high-rate communication technologies in wireless communication,as a technology which is emerging soon,Ultra Wide Band(UWB)technology can meet the high precision and low energy consumption of indoor wireless positioning.It also enables the requirements of low complexity.The research on the Direct Path(DP)is one of important parts in the UWB positioning system.We need to evaluate the performance of a positioning system by the difference between the system positioning position and the actual position..The main cause of the distance error is the multipath and non-line of sight in the UWB signal transmission Non-Line-of-Sight phenomenon.The time resolution of the UWB signal is in the nanosecond range,the interference of the multipath effect on the positioning accuracy can be reduced to some extent,but the bandwidth of the system is limited,so there is also a multipath error.The NLOS error is the interference of the UWB signal during the propagation process,causing the distance error caused by the signal propagation to generate additional delay.At this time,the error suppression algorithm is needed to modify or compensate the measurement results,now,there are four classical algorithms,many scholars had a lot of research on the four algorithms.However,most of the suppression algorithms proposed in the research literature stay in the simulation stage.There are very few real engineering applications to prove their actual results.In the actual positioning situation,the environment is generally more complicated,and the NLOS error cannot be effectively eliminated by means of a fixed transmission model,and the real transmission characteristics cannot be effectively simulated by simulation.This paper will use the self-developed UWB positioning actual products to carry out the actual test of various error suppression methods and algorithms of NLOS,NLOS various kinds of exhibition hall environment with less obstacles,complex electromagnetic environment,experimental environment with many walls and load-bearing columns,non-line-of-sight and multi-path environment.After selecting the best algorithm for non-line-of-sight and multi-path error suppression in a certain environment,adding different filtering algorithms to select the optimal combination of error suppression algorithm and filtering algorithm in different environments,analyze and solve the problems encountered in the testing process to further improve the accuracy of UWB positioning.The relatively empty environment represented by the exhibition hall and the electromagnetic environment represented by the experimental area are more complicated.Under the NLOS environment such as wall and bearing column occlusion,the Chan algorithm and Taylor expansion algorithm have better positioning effect.However,based on the huge computational complexity of the Taylor expansion algorithm,the Chan algorithm is preferred.After adding the filtering algorithm,the Kalman filtering algorithm performs well.The actual error of the whole system is about 8cm and 10cm respectively.The walls represented by the office area are thicker.In the completely non-line-of-sight environment with complex building structure,the Taylor algorithm has a relatively large amount of calculation,but its positioning performance is relatively good.Although the positioning error is around 30cm,this positioning effect is acceptable under completely non-line-of-sight conditions.