Published Paper Details:

MLSD-IoT (Multi-Layer Stateful DDoS Detection and Mitigation for Intelligent IoT Security) is a security framework designed to identify and mitigate Distributed Denial of Service attacks targeting Internet of Things environments. The rapid growth of interconnected IoT devices has increased network vulnerability to high-volume and low-rate DDoS attacks that can disrupt services, degrade performance, and compromise system availability. This study presents a multi-layer architecture that combines feature selection, machine learning based classification, and stateful traffic analysis for efficient attack detection. Network traffic data are preprocessed, normalized, and analyzed using mutual information based feature selection to identify the most relevant attributes. Multiple machine learning models including Decision Tree, Random Forest, Logistic Regression, Naive Bayes, and Support Vector Machine are evaluated to determine the most effective classifier. The proposed framework performs attack classification, traffic monitoring, and mitigation through layered decision making, improving detection reliability and response efficiency. Experimental evaluation demonstrates high classification accuracy and effective identification of malicious traffic patterns, making the framework suitable for intelligent and scalable IoT network protection.