Network Intrusion Detection Using GAN and Resnet Optimized with Glowworm Optimization

With the increasing complexity and sophistication of cyber-attacks, the development of effective Intrusion Detection System (IDS), has become crucial for ensuring Network Security. In the current era of information technology, organizations have become heavily dependent on computer networks as a means of sharing digital information, including files, data, and programs. As a result, the need to protect a network against intrusion and malicious attacks have become a viable requirement. Network Intrusion Detection Systems (NIDS) are placed within a network strategically to passively examine the traffic traversing the devices on which they are installed and report any suspicious activities. In this research, we present a novel intrusion detection system based on deep learning, employing Generative Adversarial Networks (GANs) with Resnet and Glowworm optimization (GRGSO) to detect and prevent both internal and external threats. The proposed approach has undergone experimental evaluation, resulting in exceptional outcomes on KDDcup 99 dataset with an accuracy surpassing 99.2.