Machine Learning-Driven IoT Malware Detection Via Network Traffic

Abstract

The continuous evolution of malware threats demands more advanced detection techniques, and artificial intelligence (AI) has emerged as a powerful tool in this area. Traditional security methods often fall short in addressing the increasingly sophisticated tactics used by cybercriminals. Integrating AI with network traffic analysis strengthens cybersecurity by allowing for early detection of malicious activities, providing a more effective defense against potential breaches. Monitoring network traffic is a proven method for identifying suspicious behavior and detecting compromised devices before they inflict serious damage. While some malware is caught by firewalls and other conventional security measures, many threats slip through due to advanced evasion techniques. This project explores the use of ML-driven network traffic analysis to enhance the detection of insider threats, emphasizing the need to establish baseline traffic patterns to distinguish between normal and anomalous network behavior. By understanding what typical activity looks like, deviations that could indicate malicious behavior are easier to detect. Additionally, this project aims to develop a resource-efficient model for IoT malware detection, ensuring the solution is both effective and lightweight for practical use in constrained environments.