IoT-Based Real-Time Structural Health Monitoring for Civil Infrastructure using Edge Computing and Anomaly Detection Algorithms

Abstract

The rapid deterioration of civil infrastructure, including bridges, dams, and high-rise buildings, presents a critical challenge to public safety and economic stability globally. Traditional Structural Health Monitoring systems often rely on manual inspections or centralized cloud-computing frameworks that suffer from high latency, significant bandwidth consumption, and connectivity dependence. This paper proposes a novel framework for real-time Structural Health Monitoring by integrating Internet of Things sensor networks with Edge Computing paradigms and advanced anomaly detection algorithms. By shifting data processing from centralized servers to the edge of the network, we demonstrate the ability to significantly reduce response times to structural anomalies while minimizing data transmission costs. The proposed architecture utilizes lightweight unsupervised learning models deployed directly on edge nodes to identify deviations in vibrational patterns and strain measurements. The results indicate that this decentralized approach maintains high detection accuracy while offering a robust solution for continuous, real-time integrity management of critical infrastructure assets.