Cat deep system for intrusion detection in software defined networking

Abstract

The development of software-defined networks (SDN) is the application of the network control, which is more convenient, secure and easy to develop and manage. This paper proposes an intrusion detection system (IDS) in SDN with the developed cat deep system (CDS). The training is done using the deep convolutional neural network (DCNN) with modified-cat swarm optimisation (M-CSO), which is the integration of the stochastic gradient descent (SGD) with the cat swarm optimisation (CSO). The sniffer, detector, and the sensor are the major components of the proposed system. All the packets are inspected with the sniffer to extract the features and these features are used to detect the abnormality using DCNN and then check the boundary to find the presence of attack in the system. The proposed CDS obtains the maximum accuracy, precision, recall, and F1_measure of 0.9203, 0.9498, 0.9277, and 0.9271, respectively, for NSL-KDD dataset.