Title: Energy efficient random forest classifier-based secure routing for opportunistic internet of things

Authors: Sanjay K. Dhurandher; Nisha Kandhoul; Isaac Woungang; Han-Chieh Chao

Addresses: Department of Information Technology, Netaji Subhas University of Technology, New Delhi, India; National Institute of Electronics and Information Technology, New Delhi, India ' Department of Information Technology, Netaji Subhas University of Technology, New Delhi, India ' Department of Computer Science, Toronto Metropolitan University, Toronto, Canada ' Department of Artificial Intelligence, Tamkang University, Tamsui, NewTaipei City, Taiwan; Department of Electrical Engineering, National Dong Hwa University, Hualien, Taiwan; Institute of Computer Science and Innovation, UCSI University, Kuala Lumpur, Malaysia

Abstract: Opportunistic internet of things (OppIoT) is a class of opportunistic networks, where the data are transmitted in a broadcast manner and shared among the nodes (i.e., IoT devices and human communities' devices) through opportunistic contact. In such networks, taking into consideration the energy level of each node while performing the routing of data is of utmost importance. This paper proposes an energy-efficient secure routing protocol that uses a random forest classifier (called ESRFCSec) for device behaviour predictions and for protecting the OppIoT network against packet collusion attacks. Through simulations, ESRFCSec is shown to achieve a prediction accuracy of 97.97%. It is also shown to be superior to three benchmark routing schemes in terms of node's residual energy, delivery probability, average latency, number of dead nodes, and number of dropped packets.

Keywords: OppIoT; random forest classifier; energy awareness; space efficient; packet collusion.

DOI: 10.1504/IJAHUC.2024.138743

International Journal of Ad Hoc and Ubiquitous Computing, 2024 Vol.46 No.2, pp.80 - 89

Received: 30 Nov 2023
Accepted: 07 Mar 2024
Published online: 29 May 2024 *

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