Detection of denial-of-service attacks in software-defined networking based on traffic classification using deep learning

In recent years, the increasing popularity of the Internet and its applications has led to significant growth in network users. Subsequently, the number and complexity of cyber-attacks realized against home users, businesses, government organizations, and critical infrastructure have increased significantly. In many cases, it is critical to detect attacks early before significant damage is done to protected networks and systems, including sensitive data. For this purpose, researchers and cyber security experts use software-defined networking technology to defend against cyber-attacks efficiently. Software-defined networking logically separates the control plane from the data plane. This feature enables network programming and blocks network traffic in real-time as soon as the diagnosis of anomalous activity. The main objective of this research is to detect denial-of-service attacks in software-defined networking. The framework of the proposed model includes a data preprocessing process and the implementation of a convolutional neural network structure. After preprocessing the data and building the convolutional neural network model, the training data is used as input to train the convolutional neural network model. According to the evaluation results, the values of precision, accuracy, recall, and F-measure of the proposed model are 98.23, 98.78, 98.42, and 98.32% respectively.