Human Motion Segmentation via Spatiotemporally Dual-Constrained Density Estimation with Commodity Wi-Fi Device
Open Access
Abstract
In ubiquitous Wi-Fi sensing, human motion interval segmentation is crucial for applications ranging from basic intrusion detection to advanced activity understanding. Existing methods often treat the Channel State Information (CSI) primarily as time series, overlooking its rich information in the spatial and frequency domains. To address this, we propose a training-free motion segmentation method that exploits the spatiotemporal features of CSI. We first analyze the discriminative spatial distributions of the CSI Ratio on the complex plane and construct a spatiotemporally dual-constrained local density estimator to characterize motion-induced perturbations. To overcome subcarrier selection challenges, we introduce a packet-level asymmetric truncation-based fusion algorithm, which yields a feature representation with a pronounced bimodal histogram. This enables the automatic determination of the optimal segmentation threshold based on the distribution characteristics of the truncated density image. Experiments in typical indoor environments demonstrate that the proposed method achieves high accuracy in both motion event detection and interval localization.Keywords
Funding Information
- National Natural Science Foundation of China (62371253)
- Department of Science and Technology of Hunan Province (025JJ70469)
- Education Department of Hunan Province (24B0717)
- Jiangsu Provincial Department of Education (KYCX20_0739)
This publication has 44 references indexed in Scilit:
- Uncertainty-aware risk mapping with passive WiFi and modified Zonal Safety Analysis (mZSA) in BIM for constructionAutomation in Construction, 2026
- EHUNAM, a WiFi CSI-based dataset for human and machine sensingScientific Data, 2025
- Towards scalable indoor positioning systems (IPS): User-centric challenges, methods, and recommendations for user-friendly crowd-powered frameworkInformation Fusion, 2025
- WiFi Channel State Information-based Motion Detection Across Time-Domain2024
- WiFi Sensing for Presence Detection Using Deep Learning2024
- WiFi-based non-contact human presence detection technologyScientific Reports, 2024
- An Unsupervised Approach to Motion Detection Using WiFi Signals2023
- LightSeg: An Online and Low-Latency Activity Segmentation Method for Wi-Fi Sensing2023
- WiFi CSI-based device-free sensing: from Fresnel zone model to CSI-ratio modelCCF Transactions on Pervasive Computing and Interaction, 2021
- A dataset for Wi-Fi-based human-to-human interaction recognitionData in Brief, 2020