Link-Aware Energy-Frugal Continual Learning for Fault Detection in IoT Networks

TL;DR

This paper presents an energy-efficient, event-driven continual learning framework for IoT fault detection that adapts to wireless conditions and energy constraints, significantly improving inference accuracy.

Contribution

It introduces a novel event-driven communication framework that enables collaborative model updates in IoT networks, optimizing energy use while maintaining high fault detection accuracy.

Findings

Achieves up to 42.8% improvement in inference recall

Outperforms periodic sampling and non-adaptive CL methods

Effective under tight energy and bandwidth constraints

Abstract

The use of lightweight machine learning (ML) models in internet of things (IoT) networks enables resource constrained IoT devices to perform on-device inference for several critical applications. However, the inference accuracy deteriorates due to the non-stationarity in the IoT environment and limited initial training data. To counteract this, the deployed models can be updated occasionally with new observed data samples. However, this approach consumes additional energy, which is undesirable for energy constrained IoT devices. This letter introduces an event-driven communication framework that strategically integrates continual learning (CL) in IoT networks for energy-efficient fault detection. Our framework enables the IoT device and the edge server (ES) to collaboratively update the lightweight ML model by adapting to the wireless link conditions for communication and the available energy budget. Evaluation on real-world datasets show that the proposed approach can outperform both periodic sampling and non-adaptive CL in terms of inference recall; our proposed approach achieves up to a 42.8% improvement, even under tight energy and bandwidth constraint.