Approaches in Anomaly-based Network Intrusion Detection Systems

Anomaly-based network intrusion detection systems (NIDSs) can take into consideration packet headers, the payload, or a combination of both. We argue that payload-based approaches are becoming the most effective methods to detect attacks. Nowadays, attacks aim mainly to exploit vulnerabilities at application level: thus, the payload contains the most important information to differentiate normal traffic from anomalous activity. To support our thesis, we present a comparison between different anomaly-based NIDSs, focusing in particular on the data analyzed by the detection engine to discover possible malicious activities. Furthermore, we present a comparison of two payload and anomaly-based NIDSs: PAYL and POSEIDON.