Title: Aggregate-attention network optimised and graph sample with arithmetic optimisation algorithm fostered IoT device type identification for enhancing IoT security

Authors: Bala Krishnasamy; Sathish Kumar Palani; Renjith Prabhavathi Neelakandan; S. Uma

Addresses: Department of Computer Science and Engineering, School of Computing, Bharath Institute of Higher Education and Research, Chennai – 600073, Tamil Nadu, India ' Computer and Communication Engineering, Rajalakshmi Institute of Technology, Chennai-600124, Tamil Nadu, India ' School of Computer Science and Engineering, Vellore Institute of Technology, Chennai Campus, Chennai, Tamil Nadu, India ' Department of Information Technology, Panimalar Engineering College, Chennai, Tamil Nadu, India

Abstract: Device type identification (DTI) is a significant system to recognise several device types based on internet of things (IoT) management. If an infected IoT device is not isolated from the network for a certain amount of time, it cause cross-contamination and introduce malware in the whole network. To overcome this problem, a graph sample and aggregate-attention network optimised with arithmetic optimisation algorithm fostered IoT device type identification (GrSAgAN-AOA-DTI-IoT) is proposed for enhancing IoT security. The network traffic feature vector contains maximum, minimum, mean, variance, and kurtosis, which are extracted by TF-IDF. These extracting features are supplied to the IoT device type identification phase. The proposed GrSAgAN-AOA-DTI-IoT approach is activated in Python. The GrSAgAN-AOA-DTI-IoT method attains higher accuracy 29.36%, 32.67%, 36.14% and 21.33% and lower computational complex 16.39%.11.39%, 8.36% and 14.31% compared to the existing methods.

Keywords: arithmetic optimisation algorithm; AOA; device type identification; DTI; graph sample and aggregate-attention network; internet of things; IoT; real network traffic datasets.

DOI: 10.1504/IJBIC.2024.140139

International Journal of Bio-Inspired Computation, 2024 Vol.24 No.1, pp.22 - 31

Received: 18 Feb 2023
Accepted: 25 Mar 2024
Published online: 24 Jul 2024 *

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