Quantum-Assisted Federated Edge Intelligence With Authenticated Secure Aggregation for Wearable Arrhythmia Detection in IoMT

doi:10.1109/jiot.2026.3703326

Quantum-Assisted Federated Edge Intelligence With Authenticated Secure Aggregation for Wearable Arrhythmia Detection in IoMT

2026-06-12

0

PRE

AI

DOI：10.1109/jiot.2026.3703326

Original

Original request for help

Save

Abstract

En 中文

Continuous wearable electrocardiogram monitoring has transformed cardiac assessment into an Internet of Medical Things problem. Yet, existing learning frameworks remain fragmented—addressing robustness, scalability, and cryptographic security separately while neglecting cross-layer deployment constraints. This article presents a unified cross-layer architecture that jointly designs lightweight wearable representation learning, edge-assisted variational quantum classification, Byzantine-resilient federated optimization, and standardized postquantum authenticated model exchange within a single IoT framework. The proposed system treats communication limits, latency budgets, and adversarial stability as first-class design variables. We establish nonconvex convergence guarantees under robust aggregation, PAC-style generalization bounds under heterogeneous client distributions, and explicit upper bounds on adversarial attack success growth. Evaluation on MIT-BIH with cross-dataset validation on PTB-XL demonstrates consistent macro-<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$F1$ </tex-math></inline-formula> gains over centralized and conventional federated baselines while preserving calibration quality and bounded communication cost. The results substantiate a deployment-aware secure cardiac intelligence architecture for next-generation IoT healthcare systems, supported by controlled comparisons against parameter-matched classical heads and measured cryptographic/system overheads. While the current evaluation is conducted under simulated NISQ and resource-constrained IoMT conditions, the framework is designed to support future validation on physical quantum and wearable-edge platforms.

Keywords:

Calibration

ECG arrhythmia detection

federated learning

Internet of Medical Things

postquantum cryptography

robust aggregation

variational quantum classifier

Quantum-Assisted Federated Edge Intelligence With Authenticated Secure Aggregation for Wearable Arrhythmia Detection in IoMT

Abstract

Journal

Organization