Earthquake detection at the edge: IoT crowdsensing network

TL;DR

This paper proposes an edge-based IoT crowdsensing network for earthquake detection, enabling local processing, fault tolerance, and enhanced privacy, demonstrated through a Raspberry Pi and NodeMCU implementation.

Contribution

It introduces a novel edge computing architecture for earthquake detection that reduces reliance on centralized servers and improves fault tolerance and privacy.

Findings

Successful implementation with Raspberry Pi and NodeMCU

Local processing enhances privacy and fault tolerance

Demonstrated effectiveness of Crowdquake ML model

Abstract

Earthquake Early Warning state of the art systems rely on a network of sensors connected to a fusion center in a client-server paradigm. Instead, we propose moving computation to the edge, with detector nodes that probe the environment and process information from nearby probes to detect earthquakes locally. Our approach tolerates multiple node faults and partial network disruption and keeps all data locally, enhancing privacy. This paper describes our proposal's rationale and explains its architecture. We then present an implementation using Raspberry, NodeMCU, and the Crowdquake machine learning model.