UWB localization algorithm based on BP neural network compensation extended Kalman filter

Abstract

When UWB technology is applied to indoor positioning, there will be a lot of noise interfering with its positioning accuracy. In order to improve the positioning accuracy, the improved extended incremental Kalman filtering algorithm and BP neural network algorithm are proposed to eliminate the positioning errors. In this paper, the extended incremental Kalman filter algorithm is first used to denoise the distance measured by UWB to reduce the errors caused by environment, equipment and other factors, then the classical Chan positioning algorithm is used to get the positioning results of tags, and finally the BP neural network algorithm is used to compensate the positioning results.