Leveraging Ubiquitous LTE Signals for Non-Intrusive Wheelchair Detection

Detecting individuals using wheelchairs is crucial for enhancing accessibility, providing personalized assistance, and ensuring safety in various public and private environments. This paper presents the CommSense system, which leverages Channel State Information (CSI) from existing 4G LTE communication signals for accurate wheelchair detection. By using the ubiquitously present 4G signals, the system provides a non-intrusive and cost-effective solution, removing the need for dedicated sensors or additional infrastructure. The methodology includes feature extraction using Principal Component Analysis (PCA) and classification using Support Vector Machine (SVM), which enables effective differentiation between wheelchair users and other individuals. Extensive experimental evaluation was conducted across six diverse environments, demonstrating the system's robustness and adaptability under real-world conditions. The results showed high detection accuracy, ranging from 87% to 97.5%, with the Area Under the ROC Curve (AUC) exceeding 0.96 across all cases, indicating the reliability of the CommSense system as a viable solution for healthcare monitoring and enhanced accessibility services.