Urban Activity Monitoring and Identification Based on Distributed Fiber Optic Sensing Technology

Distributed Acoustic Sensing (DAS) technology offers a high-sensitivity, cost-effective solution for urban activity monitoring and disaster early warning, leveraging existing fiber optic infrastructure. This study deployed a DAS network around Jilin University’s Chaoyang Campus. Utilizing the Stockwell Transform (S-transform) for adaptive time-frequency analysis, the research denoised mixed acoustic signals, overcoming Short-Time Fourier Transform resolution limitations. Root Mean Square (RMS) and Power Spectral Density (PSD) analysis further quantified signal characteristics. The processed data efficiently extracted and classified urban environmental noise sources. Key findings demonstrated DAS’s ability to distinguish human activities, wind speed variations, and rainfall events. Analysis of January data revealed Changchun’s winter climate patterns (frequent strong winds, scarce precipitation) and spatiotemporal human activity signatures. This capability enables long-term, continuous monitoring critical for urban safety and sustainable development. DAS provides real-time data supporting extreme weather warnings (e.g., storms), urban traffic management, and public safety event detection (e.g., crowd gatherings). Crucially, using existing fiber eliminates deploying additional sensors, significantly reducing operational costs. The technology presents a valuable tool for building resilient smart cities. Future research will expand monitoring scenarios, optimize algorithms for complex signal resolution, and explore multi-source data fusion for enhanced urban disaster prediction.