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

In recent years, the increasing popularity of theInternet and its applications has led to significant growth innetwork users. Subsequently, the number and complexity ofcyber-attacks realized against home users, businesses,government organizations, and critical infrastructure haveincreased significantly. In many cases, it is critical to detectattacks early before significant damage is done to protectednetworks and systems, including sensitive data. For thispurpose, researchers and cyber security experts use softwaredefined networking technology to defend against cyber-attacksefficiently. Software-defined networking logically separates thecontrol plane from the data plane. This feature enablesnetwork programming and blocks network traffic in real-timeas soon as the diagnosis of anomalous activity. The mainobjective of this research is to detect denial-of-service attacksin software-defined networking. The framework of theproposed model includes a data preprocessing process and theimplementation of a convolutional neural network structure.After preprocessing the data and building the convolutionalneural network model, the training data is used as input totrain the convolutional neural network model. According to theevaluation results, the values of precision, accuracy, recall, andF-measure of the proposed model are ۹۸.۲۳, ۹۸.۷۸, ۹۸.۴۲, and۹۸.۳۲% respectively.

Keywords:

Software-defined networking , trafficclassification , denial-of-service attacks , deep learning , securityanomalies.