Towards Family-Grouped Hierarchical Federated Learning on Sub-5KB Models: A Feasibility Study of Privacy-Preserving ECG Monitoring for Ultra-Resource-Constrained Wearables

TL;DR

This study introduces Family-Grouped Hierarchical Federated Learning with ultra-compact models for privacy-preserving ECG monitoring on microcontrollers, achieving significant communication reduction and high accuracy.

Contribution

It proposes a novel three-tier federated learning architecture and a tiny CNN-LSTM model suitable for ultra-resource-constrained wearables, demonstrating feasibility through simulation.

Findings

Family-FL reduces communication by 76.7% compared to FedAvg.

Family-FL-Tiny achieves 91.9% accuracy with minimal model size.

Reliable detection of critical arrhythmias demonstrated.

Abstract

Cardiovascular disease remains the leading cause of death worldwide, and early detection of arrhythmias through continuous ECG monitoring on wearable devices can prevent life-threatening events. Federated Learning (FL) enables privacy-preserving collaborative training by keeping raw ECG data on device, yet standard FL incurs prohibitive communication overhead and standard deep learning models cannot fit on ultra-low-power microcontrollers. We propose Family-Grouped Hierarchical Federated Learning (Family-FL), a three-tier architecture that uses the family as a natural privacy boundary for intra-family aggregation before global synchronization. We further design a hardware-constrained Tiny CNN-LSTM architecture with only 669 parameters, INT8-quantized to occupy merely 4.65KB Flash and 2.95KB RAM, meeting the constraints of STC32G12K128-class microcontrollers. Experiments on the MIT-BIH Arrhythmia Database (mean of 5 independent runs with different seeds) demonstrate that Family-FL reduces communication volume by 76.7% compared to FedAvg while maintaining comparable accuracy. Family-FL-Tiny achieves 91.9 +/- 1.2% accuracy with macro-F1 of 0.483 +/- 0.031, reducing total communication to 0.31% of FedAvg. The model achieves reliable ventricular arrhythmia detection (per-class F1 = 0.80), the most clinically critical abnormality for home-based preliminary screening. These results demonstrate the technical feasibility of privacy-preserving federated learning on ultra-resource-constrained microcontrollers through simulation-based evaluation. We honestly discuss limitations: no hardware deployment, single-dataset validation (MIT-BIH, 47 subjects), reduced rare-class sensitivity, and absence of formal differential privacy guarantees.