Real time earthquake prediction using cross-correlation analysis & transfer function model

IRAN is located in a region of high seismic potential. This paper presents a real-time method of predicting earthquake before its arrival. Our proposed method is based on cross-correlation calculation of data sensed by Wireless Sensor Network (WSN) as well as, Transfer Function (TF) calculation for the seismic wave propagation path between a location close to the hypo-center and the location where vibration is going to be predicted. The information which was obtained during a long period of continuous monitoring, have been used and deployed for mathematical calculations. This information was relayed to the main server to collect data and store it for future evaluations. We show that there is some consistency between discrepant faults in such a way that releasing energy in one area can lead to another vibration and quake. Our proposed working procedure consists of two algorithms first is named learning algorithm which concentrates on long-term waveform collection, Signal to Noise Ratio (SNR) enhancement, cross-correlation calculations, and finally transfer function calculation during foreshock time, and second is named Prediction Algorithm which calculates the seismic output waveform based on earthquake around hypo-center as input for already determined transfer function. By this approach, we have introduced a novel solution as Earthquake Early-Warning System (EEWS) gives golden time which can preserve human's life and mitigate economical loss by mentioned elaborated process.