Title: Enhancing network security: a deep learning-based method to detect and diminish attacks

Authors: R.E. Franklin Jino; Arockia Mary Paulsamy; Gobinath Shanmugam; Rajesh Kumar Vishwakarma

Addresses: Department of Information Technology, V.S.B Engineering College, Karudayampalayam, Tamil Nadu 639111, India ' Department of Information Technology, V.S.B Engineering College, Karudayampalayam, Tamil Nadu 639111, India ' Department of Information Technology, V.S.B Engineering College, Karudayampalayam, Tamil Nadu 639111, India ' Department of Electronics and Communication Engineering, Jaypee University of Engineering and Technology, Raghogarh -Vijaypur, Mohanpur, Madhya Pradesh 473226, India

Abstract: In today's world, preventing illegal intrusions into communication networks is an absolute need in order to protect the personal information of users and maintain the integrity of their data. The establishment of intrusion detection systems (IDS) and the improvement of their accuracy have both been shown to benefit from the use of data mining and machine learning techniques. We make use of the well-known AWID3 dataset, which contains traffic from wireless networks. The Krack and Kr00k attacks, which are aimed at the most serious vulnerabilities in the IEEE 802.11 protocols, are the primary focus of our research and development efforts. This success rate was reached by our ensemble classifier. When it came to identifying instances of the Kr00k attack, our neural network-based model had a high accuracy rate of 96.7%, which further emphasised the usefulness of the remedy that we suggested.

Keywords: wireless; IDS; machine learning; Krack; Kr00k; IEEE8021.11.

DOI: 10.1504/IJESDF.2025.148225

International Journal of Electronic Security and Digital Forensics, 2025 Vol.17 No.5, pp.631 - 645

Received: 28 Oct 2023
Accepted: 17 Jan 2024
Published online: 01 Sep 2025 *

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