A Novel Technique to Discover De-Authentication DoS Attacks in 802.11 Wi-Fi Networks

Abstract

Denial of Service (DoS) Attacks in 802.11 networks is mainly caused because of weaknesses of Media Access Layer (MAC). In this article we study about the de-authentication DoS (De-DoS) attack in 802.11 Wi-Fi networks. In De-DoS attack an intruder transmits huge spoofed de-authentication frames to the client(s) which is caused their disconnection. All existing methods to overcome from this De-DoS attack are depends upon protocol alterations, encryption, 802.11 standard updating, hardware and software upgrades which are costly. In this article we proposed a novel Machine Learning (ML) based Intrusion Detection System (IDS) to recognize the De-DoS attack in Wi-Fi network which doesn’t suffer from the above weaknesses. We have utilized number of Machine Learning based classifiers for recognition of De-DoS attack. This facilitates an administrator to decide between wide ranges of classification algorithms. The experiments performed using an in-house test bed shows that the proposed ML based IDS discovers De-DoS attack with precision and recall exceeding 96% mark.

Keywords: De-authentication, DoS, Intrusion Detection System, Machine Learning, Wi-Fi Security, WLAN, 802.11