Each day, new vulnerabilities in software are discovered and new online attacks that bypass traditional security products like firewalls and anti-virus are unleashed. Many intrusions stealthily pass through the firewall and traditional network Intrusion Detection Systems (IDS) are designed for detection and recording of attacks after the fact.
An Intrusion Detection System (IDS) is a network security technology originally built for detecting vulnerability exploits against a target application or computer. Intrusion Prevention Systems (IPS) extended IDS solutions by adding the ability to block threats in addition to detecting them and has become the dominant deployment option for IDS/IPS technologies.
The IPS often sits directly behind the firewall and provides a complementary layer of analysis that negatively selects for dangerous content. Unlike its predecessor the Intrusion Detection System (IDS)—which is a passive system that scans traffic and reports back on threats—the IPS is placed inline (in the direct communication path between source and destination), actively analysing and taking automated actions on all traffic flows that enter the network. Specifically, these actions include:
- Sending an alarm to the administrator (as would be seen in an IDS)
- Dropping the malicious packets
- Blocking traffic from the source address
- Resetting the connection
As an inline security component, the IPS must work efficiently to avoid degrading network performance. It must also work fast because exploits can happen in near real-time. The IPS must also detect and respond accurately, so as to eliminate threats and false positives (legitimate packets misread as threats).
The IPS has a number of detection methods for finding exploits, but signature-based detection and statistical anomaly-based detection are the two dominant mechanisms.
Signature-based detection is based on a dictionary of uniquely identifiable patterns (or signatures) in the code of each exploit. As an exploit is discovered, its signature is recorded and stored in a continuously growing dictionary of signatures. Signature detection for IPS breaks down into two types:
Exploit-facing signatures identify individual exploits by triggering on the unique patterns of a particular exploit attempt. The IPS can identify specific exploits by finding a match with an exploit-facing signature in the traffic stream
Vulnerability-facing signatures are broader signatures that target the underlying vulnerability in the system that is being targeted. These signatures allow networks to be protected from variants of an exploit that may not have been directly observed in the wild, but also raise the risk of false-positives.
Statistical anomaly detection takes samples of network traffic at random and compares them to a pre-calculated baseline performance level. When the sample of network traffic activity is outside the parameters of baseline performance, the IPS takes action to handle the situation.