Study of joint indoor positioning algorithm based on UWB

Abstract

Aiming at the problem of large measurement error in UWB indoor positioning system, a solution based on two-way time-of-flight ranging method to reduce clock offset error is proposed. Aiming at the motion characteristics of static and dynamic nodes to be tested, this scheme proposes a joint algorithm by detecting the motion state of the nodes to be tested. The static positioning uses the least squares estimation method to obtain the node coordinates and simulates by MATLAB. The coordinate position of the node is displayed. For the dynamic tracking problem, the least square method cannot accurately estimate the node position due to factors such as object occlusion and motion trajectory. For this problem, the extended Kalman filter algorithm is used to track and predict the position of the node to be tested, thus improving positioning effect. The simulation results show that the static positioning error can be controlled within 6%, and the dynamic tracking algorithm is relatively statically positioned and the accuracy is improved by 20%. Experiments show that the algorithm is reliable and effective, and meets the basic needs of indoor positioning.

Keywords

• ultra width-band
• two way-time of flight
• indoor positioning
• extended kalman filter