Intelligent Detection of Cyber Threats in Wireless Networks Using Machine Learning Algorithms

Abstract

Abstract—Wireless networks have in the recent past emerged as part of a communication system, yet this type of network is prone to many cyber-related attacks due to its open nature. The complexity of attacks will continue to rise, and yet the current systems of detection such as signature-basedsystems and manual monitoring fall short of detecting an attack. This article researches into the application of Machine Learning (ML) algorithms in detecting wireless network cyber threats in an intelligent way to achieve better security to the wireless systems. We suggest an ML-driven solution with both supervised and unsupervised learning to detect frequent and diverse attacks like Denial of Service (DoS) and Man-in-the-Middle (MitM) attacks and attacks that use malicious nodes. We test the performance of different machine learning models, such as accuracy, detection rates and running efficiency by conducting simulation experiments on popularly used datasets. Our findings show that the ML techniques ensure high-level threat detection system as opposed toconventional ones. The suggested system has the potential to be deployed in a real-time large-scale manner; the proposed system offers an effective and smart cybersecurity infrastructure to wireless networks.