Improved Traffic Filtering for IoT Security

Abstract

This research presents an advanced traffic filtering technique using a Support Vector Machine (SVM) model to enhance IoT network security, achieving 85% accuracy, with high precision (87%) and recall (96%) in identifying legitimate traffic. However, the model struggles with neutral traffic, highlighting the need for further refinement, and offers a scalable solution for real-time dynamic analysis, with future work focusing on improving neutral traffic classification and integrating additional features to enhance performance.