PRIVACY BY DESIGN IN WEARABLE BIOMETRIC SYSTEMS: QUANTIFYING SIGNAL QUALITY AND PRIVACY RISK TRADE-OFFS
Goran Mladenović, Aleksandar Stanković, Arso Vukićević
Abstract: Wearable biometric systems increasingly support cognitive ergonomics and human-system interaction, forcing design-time trade-offs between physiological signal quality and data-protection requirements. This study presents a system-level framework for jointly assessing signal fidelity and privacy risk in multimodal wearable biometric systems. Using a public affective-physiology dataset of 32 participants, signals were processed into a signal-quality index that captures noise robustness, spectral complexity, artifact prevalence, and temporal stability. Privacy risk was quantified as the worst case across complementary attack vectors, namely re-identification, membership inference, and information-theoretic leakage, so that a single high-risk channel rather than an averaged score governs the estimate.
Keywords: Wearable biometrics, Physiological data, Privacy risk, GDPR, Privacy-by-design
DOI: 10.24874/IJQR20.03-01
Recieved: 15.01.2026 Accepted: 01.07.2026 UDC:
Reads: 70 
