Computer Engineering Commons^™

Enhancing Snort Ids Performance Using Data Mining, Mohammed Ali Almaleki

Theses

Intrusion detection systems (IDSs) such as Snort apply deep packet inspection to detect intrusions. Usually, these are rule-based systems, where each incoming packet is matched with a set of rules. Each rule consists of two parts: the rule header and the rule options. The rule header is compared with the packet header. The rule options usually contain a signature string that is matched with packet content using an efficient string matching algorithm. The traditional approach to IDS packet inspection checks a packet against the detection rules by scanning from the first rule in the set and continuing to scan all …