An Efficient Hybrid Deep Learning Model for Denial of Service Detection in Cyber Physical Systems

Abstract

Security is critical in the Cyber-Physical Systems (CPS) model for smart healthcare networks, and it will likely have a significant impact on the industry, medical and healthcare; and farming-related substructures shortly. Due to an increase in the frequency of security and privacy attacks in present times in healthcare networks, this article addressed a fundamental component of intrusion detection systems (IDS) based on the important parameter security. The limitations of IDS in reacting to cyberattacks as well as in establishing private controls in the field of smart healthcare have motivated this research. An efficient and lightweight deep learning-based CNN-Bidirectional LSTM is proposed for the DDoS detection that uses the features of Convolutional Neural Networks (CNNs) to classify traffic flows as benign and malicious in this study. The results are achieved using Python where four convolutional layers, Maximum Pooling, that ends with the Dense Layer. The hyperparameters used are batch size of 500, epochs 20, number of classes 25, and Relu and softmax pooling activation function along with the softmax