Intrusion Detection in 5G Cellular Network Using Machine Learning

Ishtiaque Mahmood, Tahir Alyas, Sagheer Abbas, Tariq Shahzad, Qaiser Abbas, Khmaies Ouahada

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

Attacks on fully integrated servers, apps, and communication networks via the Internet of Things (IoT) are growing exponentially. Sensitive devices' effectiveness harms end users, increases cyber threats and identity theft, raises costs, and negatively impacts income as problems brought on by the Internet of Things network go unnoticed for extended periods. Attacks on Internet of Things interfaces must be closely monitored in real time for effective safety and security. Following the 1, 2, 3, and 4G cellular networks, the 5th generation wireless 5G network is indeed the great invasion of mankind and is known as the global advancement of cellular networks. Even to this day, experts are working on the evolution's sixth generation (6G). It offers amazing capabilities for connecting everything, including gadgets and machines, with wavelengths ranging from 1 to 10 mm and frequencies ranging from 300 MHz to 3 GHz. It gives you the most recent information. Many countries have already established this technology within their border. Security is the most crucial aspect of using a 5G network. Because of the absence of study and network deployment, new technology first introduces new gaps for attackers and hackers. Internet Protocol(IP) attacks and intrusion will become more prevalent in this system. An efficient approach to detect intrusion in the 5G network using a Machine Learning algorithm will be provided in this research. This research will highlight the high accuracy rate by validating it for unidentified and suspicious circumstances in the 5G network, such as intruder hackers/attackers. After applying different machine learning algorithms, obtained the best result on Linear Regression Algorithm's implementation on the dataset results in 92.12% on test data and 92.13% on train data with 92% precision.

Original languageEnglish
Pages (from-to)2439-2453
Number of pages15
JournalComputer Systems Science and Engineering
Volume47
Issue number2
DOIs
Publication statusPublished - 2023

Keywords

  • availability
  • confidentiality
  • integrity
  • Intrusion detection system
  • machine learning

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Intrusion Detection in 5G Cellular Network Using Machine Learning'. Together they form a unique fingerprint.

Cite this