Real-time detection of earthquakes through a smartphone-based sensor network

The Earthquake Network project implements a world-wide smartphone-based sensor network for the detection of earthquakes. The accelerometric sensor onboard each smartphone is used to detect vibrations which are immediately reported to a server. The server analyses the information coming from the entire network and when a quake is detected it is notified to all smartphone users in quasi real-time. In this work we propose and compare two solutions to the detection problem. One solution is based on a likelihood approach and the other is based on filtering.