A Motion Intelligence Analysis System Based on Ultra-Wideband

This study proposes a high-precision sports analysis system, to overcome the limitations of existing motion tracking technologies in athlete training. The system integrates Ultra-Wideband (UWB) with SDS-TWR ranging and IMU sensors, achieving centimeter-level accuracy through a hybrid neural network and adaptive Kalman filtering algorithm. It further incorporates IMUbased trajectory compensation for robustness against UWB packet loss. A real-time visualization platform and AHP-TOPSIS evaluation framework provide comprehensive performance analysis. Experimental results demonstrate the system’s superior accuracy and reliability, offering an effective solution for digital sports training and performance assessment. The system integrates: High-precision positioning using SDS-TWR (Symmetric Double-Sided Two-Way Ranging) with TOA (Time of Arrival),achieving centimeter-level accuracy; A hybrid localization algorithm combining neural networks and adaptive Kalman filtering to reduce noise and dynamic errors; IMU-based trajectory compensation to mitigate UWB packet loss; A multi-threaded visualization platform with ECharts for real-time monitoring; An AHP-TOPSIS evaluation framework quantifying athlete performance across $\mathbf{5}$ key metrics.