An intrusion prevention system (IPS) monitors a network's data packets for suspicious activity and tries to take action using specific policies. It acts somewhat like an intrusion detection system that includes a firewall to prevent attacks. It sends an alert to a network or systems administrator when something suspicious is detected, allowing the administrator to select an action to take when the event occurs. Intrusion prevention systems can monitor an entire network, wireless network protocols, network behavior and a single computer's traffic. Each IPS uses specific detection methods to analyze risks.
Depending on the IPS model and its features, an intrusion prevention system can detect various security breaches. Some can detect the spread of malware across a network, the copying of large files between two systems, and the use of suspicious activities such as port scanning. After the IPS compares the issue to its security rules, it logs each event and documents the event's frequency. If the network administrator configured the IPS to perform a specific action based on the incident, the intrusion prevention system then takes the assigned action. A basic alert is sent to the administrator so he or she can respond appropriately or view additional information on the IPS, if necessary.
There are four general types of intrusion prevention systems, including network-based, wireless, network behavior analysis and host-based. A network-based IPS analyzes various network protocols and is commonly used on remote access servers, virtual private network servers and routers. A wireless IPS watches for suspicious activities on wireless networks and also looks for unauthorized wireless networks in an area. Network behavior analysis looks for threats that could take down a network or spread malware and is commonly used with private networks that connect to the Internet. A host-based IPS works on a single system and looks for strange application processes, unusual network traffic to the host, file system modification and configuration changes.
There are three detection methods an intrusion prevention system can use, and many systems use a combination of all three. Signature-based detection works well for detecting known threats by comparing an event to an already documented signature to determine if a security breach has occurred. Anomaly-based detection looks for activity that is abnormal when compared to the normal events that occur on a system or network and is particularly useful for identifying unknown threats. Stateful protocol analysis looks for activity that goes against how a specific protocol is normally used.