A Computationally Light Semi-Supervised Learning Framework for Constrained Embedded Platforms

Avirup Roy

doi:10.1109/JIOT.2025.3646121

Return

A Computationally Light Semi-Supervised Learning Framework for Constrained Embedded Platforms

2025-12-18

Avirup Roy

Amit Kumar Bhuyan

Subir Biswas

DOI：10.1109/JIOT.2025.3646121

Original

Original request for help

Save

Abstract

En 中文

This article presents a computationally efficient semi-supervised learning (SSL) framework designed for real-time, on-device learning in resource-constrained Internet of Things (IoT)-based health monitoring systems. The proposed methods focus on minimizing redundant computation and memory usage, which are directly linked to energy consumption in embedded platforms. First, mini-batch K-means (KMs) clustering is employed as an alternative to full-batch clustering, reducing per-iteration computational time and memory footprint, particularly for low to moderate-dimensional datasets. To further enhance efficiency, two unsupervised convergence detection mechanisms, $\boldsymbol {\beta }$ -stop and $\boldsymbol {\zeta }$ -stop are introduced to autonomously halt clustering once model stability is achieved, preventing unnecessary retraining and reducing energy overhead. $\boldsymbol {\beta }$ -stop monitors the stabilization of clustering iterations per learning cycle, while $\boldsymbol {\zeta }$ -stop tracks the rate of cluster growth as a convergence indicator. Experimental evaluations on three representative IoT health monitoring datasets: smart hydration tracking (SHT), human activity detection (HAD), and infant activity detection (IAD) demonstrate that the proposed strategies reduce computational time by up to 45% and CPU memory consumption by up to 30% without compromising classification accuracy. The results confirm the framework’s scalability, energy efficiency, and suitability for reliable, real-time SSL on embedded IoT devices.

Keywords:

Constrained devices

eHealth and mHealth

energy efficient devices

in situ processing

mobile and ubiquitous systems

semi-supervised learning (SSL)