Development of an Intelligent Classifier Model for Denial of Service Attack Detection

• DOI: 10.9781/ijimai.2023.08.003
• Publication Date: 2023-09-07T14:19:29Z
• Authors: Álvaro Michelena; Jose Aveleira-Mata; Esteban Jove; Héctor Alaiz-Moretón; Héctor Quintián; José Luis Calvo-R...
• Keywords: Informática; Soft computing; Technology; Cybersecurity; cybersecurity; 1203.18 Sistemas de Información, Diseño Componentes; machine learning classifier; T; feature extraction; soft computing; mqtt; DoS attacks; IJIMAI; Supervised classifiers; dos attack; Ingeniería de sistemas; supervised learning; Supervised classifiers.; 1207.03 Cibernética; Feature extraction; 1203.06 Sistemas Automatizados de Control de Calidad; MQTT; DoS Attack

The prevalence of Internet Things (IoT) systems deployment is increasing across various domains, from residential to industrial settings.These are typically characterized by their modest computational requirements and use lightweight communication protocols, such as MQTT.However, the rising adoption IoT technology has also led emergence novel attacks, susceptibility these compromise.Among different attacks that can affect main protocols Denial Service (DoS).In this scenario, paper evaluates performance six supervised classification techniques (Decision Trees, Multi-layer Perceptron, Random Forest, Support Vector Machine, Fisher Linear Discriminant Bernoulli Gaussian Naive Bayes) combined with Principal Component Analysis (PCA) feature extraction method for detecting DoS in MQTT networks.For purpose, a real dataset containing all traffic generated network many executed been used.The results obtained several models have achieved performances above 99% AUC.