Scalable Inline Network-Intrusion Detection System with Minimized Memory Requirement

Currently used network-intrusion detection systems (NIDSs) using deep learning have limitations in processing large amounts of data real time. This is because collecting flow information and creating features are time consuming require considerable memory. To solve this problem, a novel NIDS with θ(1) memory complexity for proposed study. Owing to its small requirement, the model can handle numerous concurrent flows. In addition, it uses raw packet as input models, resulting lightweight feature-creation process. For fast detection, classifies received packet, though prone false detection. weakness solved through validation research, high accuracy. Furthermore, real-time possible since intrusion be performed every Inception model. A performance comparison existing methods confirmed an effectively improved lower requirement by 73% 77% on average while maintaining Thus, overcome problems modern deep-learning-based NIDSs.