Detecting IoT Malware with Knowledge Distillation Technique

Abstract

The increasing prevalence and sophistication of Internet of Things (IoT) malware pose significant security threats to IoT devices and networks. Traditional machine learning models for detecting IoT malware are often resource-intensive and computationally expensive, making them unsuitable for deployment on resource-constrained IoT devices. To address this challenge, we propose a novel approach for detecting IoT malware using knowledge distillation, where a teacher model is trained on a large and diverse set of data to provide a compact and efficient student model. We evaluate our approach on two widely used datasets, IoT -23 and Malevis, and achieve promising results, demonstrating the effectiveness of our approach in detecting IoT malware. Our approach can help address the resource constraints and computational challenges of deploying traditional machine learning models on IoT devices. Our study highlights the potential of knowledge distillation as a promising approach for developing lightweight and efficient models for IoT malware detection. Our approach can be extended to detect other types of IoT threats and integrated with existing IoT security systems to provide an additional layer of defense against malware attacks. The future research directions outlined in this paper can help further advance the field of IoT security and enable the development of more effective and efficient approaches for detecting and mitigating IoT malware attacks.

Topics & Concepts