Title: Anomaly detection system in 5G networks via deep learning model

Authors: Vikram Sadashiv Gawali; Nihar M. Ranjan

Addresses: Government College of Engineering, Chandrapur, Maharashtra, India ' Department of Information Technology, JSPM's Rajarshi Shahu College of Engineering, Pune, Maharashtra, India

Abstract: Fifth generation (5G) networks are susceptible to a number of attacks that target the 5G platform's major components, including radio communication, user equipment, core and edge networks. Consequently, the aim of this work is to provide a unique feature extraction and detection system for 5G networks. The input data goes through a preparation phase first. The extracted characteristics include statistical and higher order statistical features, technical indicators, raw features, information gain and improved entropy. This procedure is then applied to the pre-processed data. Finally, the detection phase receives the retrieved characteristics, here Hybrid Classifier (HC), including Deep Belief Network (DBN) and Bidirectional Long-Short-Term Memory (Bi-LSTM) is used. To convert detection stage accurately and precisely, the weights of both Bi-LSTM and DBN are optimised using a novel Deer Hunting updated Sun Flower Optimisation (DHSFO) model that hybrids the concept of Sun Flower Optimisation (SFO) and Deer Hunting Optimisation (DHO) algorithm.

Keywords: 5G network; anomaly detection; feature extraction; hybrid classifiers; optimisation.

DOI: 10.1504/IJWMC.2023.131319

International Journal of Wireless and Mobile Computing, 2023 Vol.24 No.3/4, pp.287 - 302

Received: 10 Dec 2021
Received in revised form: 28 Sep 2022
Accepted: 06 Oct 2022
Published online: 06 Jun 2023 *

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