A Multimodal Authorization Approach for Integrated Access Control Process

Abstract

Traditional authorization mechanisms such as passwords, smart cards, and conventional biometrics remain vulnerable to spoofing, replay, and social-engineering attacks, particularly in IoT and cloud environments. Physiological biometrics, including electrocardiogram (ECG) and electromyography (EMG) signals, offer stronger resistance to forgery because of their biological origin while also capturing emotional and behavioral context. This study presents AlMuwathiq, an emotion-based multimodal authorization system that integrates ECG and EMG signals to support adaptive access-control decisions. Real physiological signals were collected and labeled using the Self-Assessment Manikin across five emotional states. After signal preprocessing and feature extraction, multiple machine learning models were evaluated, with a focus on explainable learning algorithms. Experimental results indicate that Random Forest and XGBoost achieved the most stable classification performance. The trained models were integrated into a real-time platform where access control authorization decisions are determined by emotional rules. The findings demonstrate that emotion-aware physiological authorization can enhance security, reliability, and context awareness in modern cybersecurity systems. This study thus advances the current state of the art in access control systems, especially in mission-critical infrastructures.