An inland Wireless Sensor Network system for monitoring seismic activity

Efficient early-warning systems for earthquakes should be able to provide accurate information on the seismic event allowing ample time for precautionary measures. In this paper, an assessment of the feasibility of using a Wireless Sensor Networks (WSNs) for monitoring seismic activity on the island of Mauritius using primary waves (P-waves) is carried out. Mauritius is one of the countries which are highly at risk to natural calamities. A simulation platform is designed and implemented to test the efficiency of a WSN in monitoring seismic activity. The system estimates the location of the hypocenter and the local velocity using time delays in P-wave arrival at each sensing node. Deviations between the estimated and actual quantities are used to study the efficiency of the system. Cross-correlation analysis of the recorded P-wave signals allowed accurate propagation delay estimation up to signal-to-noise ratios (SNRs) as low as -15dB. Time synchronization between the sensing nodes using timing-sync protocol for sensor networks improved the performance of the relocation algorithm by reducing the nodes' clock offsets to within 0.50ms. The proposed system was tested through numerical simulations with various SNRs, sampling frequencies and number of sensing nodes.