Article: Cognitive distraction identification using physiological signals-based enhancing road safety through attributed multi-order graph convolutional network Journal: International Journal of Wireless and Mobile Computing (IJWMC) 2026 Vol.30 No.4 pp.395 - 407 Abstract: Driver distraction, particularly cognitive distraction, is a major cause of road accidents. While visual and manual distractions manifest through observable physical behaviours, the cognitive distraction presents unique detection challenges. The existing methods are not easily scalable due to the high cost of data acquisition devices. In this paper, an Enhancing Road Safety through Attributed Multi-Order Graph Convolutional Network-Based Cognitive Distraction Identification using Physiological Signals (ERD-AMGCN-CDI-PS) is proposed. The ERD-AMGCN-CDI-PS utilises physiological signals from the DEAP and WESAD data sets. The input signals are pre-processing utilising Regularised Bias-Aware Ensemble Kalman Filter (RBAEKF) to remove noise and artifacts. The Synchro-Transient-Extracting Transform (STET) is used to extract visual features from pre-processing signals. These features are given to the Attributed Multi-Order Graph Convolutional Network (AMGCN) to identify cognitive distraction. The ERD-AMGCN-CDI-PS method achieves 19.56%, 10.88% and 19.60% higher accuracy and 19.83%, 11.57% and 19.65% lower False Positive Rate (FPR) over the existing techniques. Inderscience Publishers - linking academia, business and industry through research

Title: Cognitive distraction identification using physiological signals-based enhancing road safety through attributed multi-order graph convolutional network

Authors: P.S. Soumya; S. Mythili

Addresses: Department of Computer Science, Karpagam Academy of Higher Education, Coimbatore, Tamil Nadu, India ' Department of Computer Science, Karpagam Academy of Higher Education, Coimbatore, Tamil Nadu, India

Abstract: Driver distraction, particularly cognitive distraction, is a major cause of road accidents. While visual and manual distractions manifest through observable physical behaviours, the cognitive distraction presents unique detection challenges. The existing methods are not easily scalable due to the high cost of data acquisition devices. In this paper, an Enhancing Road Safety through Attributed Multi-Order Graph Convolutional Network-Based Cognitive Distraction Identification using Physiological Signals (ERD-AMGCN-CDI-PS) is proposed. The ERD-AMGCN-CDI-PS utilises physiological signals from the DEAP and WESAD data sets. The input signals are pre-processing utilising Regularised Bias-Aware Ensemble Kalman Filter (RBAEKF) to remove noise and artifacts. The Synchro-Transient-Extracting Transform (STET) is used to extract visual features from pre-processing signals. These features are given to the Attributed Multi-Order Graph Convolutional Network (AMGCN) to identify cognitive distraction. The ERD-AMGCN-CDI-PS method achieves 19.56%, 10.88% and 19.60% higher accuracy and 19.83%, 11.57% and 19.65% lower False Positive Rate (FPR) over the existing techniques.

Keywords: attributed multi-order graph convolutional network; cognitive distraction; regularised bias-aware ensemble Kalman filter; road safety; synchro-transient-extracting transform.

DOI: 10.1504/IJWMC.2026.154156

International Journal of Wireless and Mobile Computing, 2026 Vol.30 No.4, pp.395 - 407

Received: 02 Dec 2024
Accepted: 30 Jul 2025
Published online: 15 Jun 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Cognitive distraction identification using physiological signals-based enhancing road safety through attributed multi-order graph convolutional network

Keep up-to-date