Ultra-Wideband (UWB) Positioning Enhancement via Anomaly Detection and Interference Mitigation

Abstract

Ultra-Wideband (UWB) technology, also known as pulse radio technology, provides highly accurate real-time tracking for indoor and outdoor environments, achieving centimeter- or millimeter-level precision. This paper analyzes normal and abnormal data using anchor point distance measurements to develop an accurate positioning model. The research is divided into five tasks: Task 1 preprocesses the data by exporting, cleaning, and filling missing values. Task 2 creates models for normal and abnormal data. Task 3 refines the model with updated anchor point coordinates. Task 4 uses an ensemble learning method to detect signal interference. Task 5 removes interference using the model from Task 4 and applies a Kalman filter to improve positioning accuracy. This method enhances UWB-based systems in dealing with data anomalies and interference.