Title: Synoptic crow search with recurrent transformer network for DDoS attack detection in IoT-based smart homes

Authors: Abhijeet Ramesh Raipurkar

Addresses: Shri Ramdeobaba College of Engineering and Management, Nagpur, Katol Rd., Gittikhadan, Maharashtra – 440013, India

Abstract: Smart home devices are vulnerable to various attacks, including distributed-denial-of-service (DDoS) attacks. Current detection techniques face challenges due to nonlinear thought, unusual system traffic, and the fluctuating data flow caused by human activities and device interactions. Identifying the baseline for 'normal' traffic and suspicious activities like DDoS attacks from encrypted data is also challenging due to the encrypted protective layer. This work introduces a concept called synoptic crow search with recurrent transformer network-based DDoS attack detection, which uses the synoptic weighted crow search algorithm to capture varying traffic patterns and prioritise critical information handling. An adaptive recurrent transformer neural network is introduced to effectively regulate DDoS attacks within encrypted data, counting the historical context of the data flow. The proposed model shows effective performance in terms of low false alarm rate, higher detection rate, and accuracy.

Keywords: smart homes; internet of things; network security; distributed denial of service; attack detection; crow search algorithm; CSA; recurrent neural network.

DOI: 10.1504/IJWET.2024.142215

International Journal of Web Engineering and Technology, 2024 Vol.19 No.3, pp.330 - 355

Received: 29 Nov 2023
Accepted: 14 Apr 2024
Published online: 14 Oct 2024 *

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