Anomalous Network Packet Detection: A Review of Attacks, Datasets and Techniques

Anomalous network packet detection is a dire security concern today. ChatGPT era has emerged as a grave concern for computer scientists and programmers. The method of recognizing diverged network activity from the typical behaviour of network flows is known as anomalous network packet detection. As networks have expanded in terms of complexity and size, detecting outlier occurrences in such huge and complex networks has become crucial in retaining integrity. Detecting threats quickly and accurately determines the success of any network flow. To ensure reliable behaviour and capacity to model features, deep learning models have emerged as promising tools in recent years. Developing an efficient anomaly detection system necessitates understanding the imbalanced datasets, recognizing changing patterns and identifying irregularities. The paper outlines attacks, knowledgeable parameters, techniques and datasets for studying crucial network anomalies by reviewing research for different periods of analysis—before 2000, 2000–2010, 2011–2015, 2016–2018 and 2019–2023. The challenges are addressed, and futuristic options for anomaly detection systems are proposed.