Izvorni znanstveni članak
https://doi.org/10.32985/ijeces.15.2.2
An Effective Technique to Detect WIFI Unauthorized Access using Deep Belief Network
Rajakumar S
; Professor, Department of Electronics & Communication Engineering, Panimalar Engineering College, Chennai, 600123 India
*
William P
; Assistant Professor & Dean, Research and Development, Department of Information Technology, Sanjivani College of Engineering, SPPU, Pune
Mabel Rose R. A
; Assistant Professor, Computer Science and Engineering, S.A. Engineering College, Poonamallee, Thiruverkadu, Tamil Nadu 600077 India.
Subraja Rajaretnam
; Assistant professor, Department of Electronics and Communication Engineering, Sathyabama Institute of Science and Technology, Jeppiaar Nagar, Chennai 600 119 India
Azhagu Jaisudhan Pazhani A
; Associate professor, Department of Electronics & Communication Engineering, Ramco Institute of Technology, Rajapalayam, Tamil Nadu 626117 India
Ahilan A
; Associate Professor, Department of Electronics and Communication Engineering, PSN College of Engineering and Technology, India
* Dopisni autor.
Sažetak
Network security has grown to be a major concern in recent years due to the popularity and development of Wi-Fi networks. However, the use of Wi-Fi networks is expanding quickly, and so is the number of attacks on Wi-Fi networks. In this paper, a novel WiFi Unauthorized Access Detection System (WUADS) technique has been proposed to detect unauthorized access in the WiFi network. Initially, the Wi-Fi frames are collected from the AWID dataset. The features of the Wi-Fi frame are extracted by using Principal Component Analysis (PCA). Finally, the Deep Belief Network (DBN) is employed for classification into authorized access and unauthorized access. The efficiency of the proposed WUADS technique was evaluated based on the parameters like accuracy, F1score, detection rate, precision, and recall. The performance analysis of the proposed WUADS technique achieves an overall accuracy range of 99.52%. The proposed WUADS method has a high success rate and the quickest attack detection time compared to deep learning techniques like CNN, RNN, and ANN. The proposed WUADS improves the overall accuracy better than 1.12%, 0.1%, and 14.22% comparative analysis of the SAE (Stacked AutoEncoder), WNIDS (wireless Network Intrusion Detection System), and 3D-ID (3 Dimensional-Identification) respectively.
Ključne riječi
Wi-Fi networks; Unauthorized Access Principal Component Analysis; Deep Belief Network; Principal Component Analysis;
Hrčak ID:
314571
URI
Datum izdavanja:
23.2.2024.
Posjeta: 548 *