Abstract. For indoor positioning, ultra-wideband (UWB) radar comes
to the forefront due to its strong penetration, anti-jamming, and high-precision ranging abilities. However, due to the complex indoor environment
and disorder of obstacles, the problems of diffraction, penetration, and
ranging instability caused by UWB radar signals also emerge, which make it
difficult to predict the noise and leads to a great impact on the accuracy
and stability of the measurement data in the short term. Therefore, the
abnormal value migration of the positioning trajectory occurred in real-time
positioning. To eliminate this phenomenon and provide more accurate results,
the abnormal values need to be removed. It is not difficult to eliminate
abnormal values accurately based on a large number of data, but it is still a
difficult problem to ensure the stability of the positioning system by using
a small number of measurement data in a short time to eliminate abnormal
value in real-time ranging data. Thus, this paper focuses on the
experimental analysis of a UWB-based indoor positioning system. By
repeatedly measuring the range , a large number of measurement data can be
obtained. Using the massive data to train linear regression models, we get
the parameter of the linear model of range data measured with the UWB radar.
Based on the Gaussian function outlier detection, abnormal values are
eliminated, and putting the new range data into the regression model trained
by us, the ranging error is reduced by nearly 50 % compared with the peak
and mean ranging errors in general.