Urban Seismology for Anonymous Monitoring of Urban Activities
Gang FANG#+, Yunyue, Elita LI, Enhedelihai NILOT, Yumin ZHAO, Arhur CHENG
National University of Singapore, Singapore

Traditional urban activity monitoring based on cameras, although very effective with recent advances in computer vision, raises severe privacy concerns. Urban seismic monitoring, which records vibrations caused by human activity with geophones, has its natural attribute of anonymity. Here, we present a case study demonstrating the use of seismic method to monitor urban activities with less privacy intrusion. We analyze the seismic signals recorded by a group of wireless geophones deployed on the campus of National University of Singapore for the past year. Based on the spectral analysis of the seismic data, we characterize the seismic signals induced by different kinds of human activities, including motor traffic, air traffic and foot traffic. For road vehicles, we observe strong energy across a wide frequency band (0-200 Hz) on their seismic spectrograms, based on which we propose an automatic motor traffic counting algorithm. For airplanes, we observe unique Doppler phenomena in the spectrogram, from which we estimate flight parameters such as flight path, height, and speed. For pedestrians, we observe spiky vibrations caused by footsteps, from which we estimate the cadence information. After analyzing the seismic monitoring data, we observe a clear correlation between the traffic activities leaving the NUS campus and the severity of the government’s restrictions during Circuit Breaker and different phases of opening. Our study suggests that the anonymity of seismic method enables high resolution monitoring of urban activity, which provide open and useful information for evaluating the effectiveness of public policies, monitoring local human aggregation, and optimizing the use of public facilities.