Abstract: In this paper we look into the deauthentication Denial of Service (De-DoS henceforth) attack in 802.11 Wi-Fi networks. The attack is very serious in nature, as the usage of few system resources can actually disconnect the Wi-Fi clients connected to the network facing immediate disconnection. The primary reason for this attack is the MAC layer vulnerabilities that exist in 802.11 Wi-Fi networks. Many current solutions to deal with De-DoS attack propose usage of digital certificates, using encryption, up-gradation of standards, and other cumbersome solutions which are difficult to deploy and increase the maintenance costs. In De-DoS attack and attacker sends a large number of deauth frames targeting a set of clients. All the client receiving this deauth frames are immediately disconnected from the network. In this paper we propose a Machine Learning (ML) based Intrusion Detection System (IDS) to identify the De-DoS attack in Wi-Fi network. The proposed solution is effective and has high detection rate and accuracy and does not have the problems associated with the existing solutions.
Keywords: Deauthentication DoS, Wi-Fi Security, Intrusion Detection System.
|
DOI:
10.17148/IJARCCE.2018.768
