Enhanced Network Intrusion Detection Framework Using Deep Convolutional Neural Networks

The Internet of Things (IoT) has transformed human interaction with the natural world by enabling seamless connectivity and communication among smart objects. This interconnection also introduces new security issues, encompassing cyber and physical threats. Protecting network security is crucial, and network intrusion detection systems (NIDS) play a significant role in this effort. In this study, we present a framework that combines principal component analysis (PCA) with convolutional neural network (CNN) techniques to tackle these security issues. Our model effectively differentiates between normal traffic and potential threats by utilizing CNN-1D, CNN-2D, and CNN-3D algorithms trained on the NSL-KDD dataset. The proposed study demonstrates significant enhancements in overall performance compared to previous works. This research provides to the development of NIDS capabilities, enhancing security in the IoT era, and promoting a safer, interconnected world.