A New Software Defined Networks (SDN) in IoTs Based Deep Learning Techniques

Abstract

In this study, a new software defined networks (SDN) in IoTs based on deep learning techniques was implemented using various types of classifiers such as DNN, CNN, GRU, LSTM RNN and SDN Ryu controller. The system was able to handle high-dimensional and complex data by using NSL-KDD dataset, and was able to detect unknown intrusions that traditional methods may miss. The effectiveness of the proposed model was evaluated by accuracy, precision, recall, F-score, and confusion matrix. Python 3.10 has been used to implementation our system. The proposed system was able to achieve good performance, however, the system's efficacy will be determined by the kind of the data feed and the scale of the issue that is attemped to address. This study highlights the potential of DL-based NIDS with SDN and IoT to detect network intrusions, but also highlights the need for continuous monitoring and updating to ensure that the system remains effective.