Intrusion detection system
An intrusion detection system (IDS; also intrusion prevention system or IPS) is a device or software application that monitors a network or systems for malicious activity or policy violations.[1] Any intrusion activity or violation is typically either reported to an administrator or collected centrally using a security information and event management (SIEM) system. A SIEM system combines outputs from multiple sources and uses alarm filtering techniques to distinguish malicious activity from false alarms.[2]
Not to be confused with intruder detection.IDS types range in scope from single computers to large networks.[3] The most common classifications are network intrusion detection systems (NIDS) and host-based intrusion detection systems (HIDS). A system that monitors important operating system files is an example of an HIDS, while a system that analyzes incoming network traffic is an example of an NIDS. It is also possible to classify IDS by detection approach. The most well-known variants are signature-based detection (recognizing bad patterns, such as malware) and anomaly-based detection (detecting deviations from a model of "good" traffic, which often relies on machine learning). Another common variant is reputation-based detection (recognizing the potential threat according to the reputation scores). Some IDS products have the ability to respond to detected intrusions. Systems with response capabilities are typically referred to as an intrusion prevention system.[4] Intrusion detection systems can also serve specific purposes by augmenting them with custom tools, such as using a honeypot to attract and characterize malicious traffic.[5]
Comparison with firewalls[edit]
Although they both relate to network security, an IDS differs from a firewall in that a conventional network firewall (distinct from a next-generation firewall) uses a static set of rules to permit or deny network connections. It implicitly prevents intrusions, assuming an appropriate set of rules have been defined. Essentially, firewalls limit access between networks to prevent intrusion and do not signal an attack from inside the network. An IDS describes a suspected intrusion once it has taken place and signals an alarm. An IDS also watches for attacks that originate from within a system. This is traditionally achieved by examining network communications, identifying heuristics and patterns (often known as signatures) of common computer attacks, and taking action to alert operators. A system that terminates connections is called an intrusion prevention system, and performs access control like an application layer firewall.[6]
Placement[edit]
The correct placement of intrusion detection systems is critical and varies depending on the network. The most common placement is behind the firewall, on the edge of a network. This practice provides the IDS with high visibility of traffic entering your network and will not receive any traffic between users on the network. The edge of the network is the point in which a network connects to the extranet. Another practice that can be accomplished if more resources are available is a strategy where a technician will place their first IDS at the point of highest visibility and depending on resource availability will place another at the next highest point, continuing that process until all points of the network are covered.[34]
If an IDS is placed beyond a network's firewall, its main purpose would be to defend against noise from the internet but, more importantly, defend against common attacks, such as port scans and network mapper. An IDS in this position would monitor layers 4 through 7 of the OSI model and would be signature-based. This is a very useful practice, because rather than showing actual breaches into the network that made it through the firewall, attempted breaches will be shown which reduces the amount of false positives. The IDS in this position also assists in decreasing the amount of time it takes to discover successful attacks against a network.[35]
Sometimes an IDS with more advanced features will be integrated with a firewall in order to be able to intercept sophisticated attacks entering the network. Examples of advanced features would include multiple security contexts in the routing level and bridging mode. All of this in turn potentially reduces cost and operational complexity.[35]
Another option for IDS placement is within the actual network. These will reveal attacks or suspicious activity within the network. Ignoring the security within a network can cause many problems, it will either allow users to bring about security risks or allow an attacker who has already broken into the network to roam around freely. Intense intranet security makes it difficult for even those hackers within the network to maneuver around and escalate their privileges.[35]
There are a number of techniques which attackers are using, the following are considered 'simple' measures which can be taken to evade IDS:
Development[edit]
The earliest preliminary IDS concept was delineated in 1980 by James Anderson at the National Security Agency and consisted of a set of tools intended to help administrators review audit trails.[39] User access logs, file access logs, and system event logs are examples of audit trails.
Fred Cohen noted in 1987 that it is impossible to detect an intrusion in every case, and that the resources needed to detect intrusions grow with the amount of usage.[40]
Dorothy E. Denning, assisted by Peter G. Neumann, published a model of an IDS in 1986 that formed the basis for many systems today.[41] Her model used statistics for anomaly detection, and resulted in an early IDS at SRI International named the Intrusion Detection Expert System (IDES), which ran on Sun workstations and could consider both user and network level data.[42] IDES had a dual approach with a rule-based Expert System to detect known types of intrusions plus a statistical anomaly detection component based on profiles of users, host systems, and target systems. The author of "IDES: An Intelligent System for Detecting Intruders", Teresa F. Lunt, proposed adding an artificial neural network as a third component. She said all three components could then report to a resolver. SRI followed IDES in 1993 with the Next-generation Intrusion Detection Expert System (NIDES).[43]
The Multics intrusion detection and alerting system (MIDAS), an expert system using P-BEST and Lisp, was developed in 1988 based on the work of Denning and Neumann.[44] Haystack was also developed in that year using statistics to reduce audit trails.[45]
In 1986 the National Security Agency started an IDS research transfer program under Rebecca Bace. Bace later published the seminal text on the subject, Intrusion Detection, in 2000.[46]
Wisdom & Sense (W&S) was a statistics-based anomaly detector developed in 1989 at the Los Alamos National Laboratory.[47] W&S created rules based on statistical analysis, and then used those rules for anomaly detection.
In 1990, the Time-based Inductive Machine (TIM) did anomaly detection using inductive learning of sequential user patterns in Common Lisp on a VAX 3500 computer.[48] The Network Security Monitor (NSM) performed masking on access matrices for anomaly detection on a Sun-3/50 workstation.[49] The Information Security Officer's Assistant (ISOA) was a 1990 prototype that considered a variety of strategies including statistics, a profile checker, and an expert system.[50] ComputerWatch at AT&T Bell Labs used statistics and rules for audit data reduction and intrusion detection.[51]
Then, in 1991, researchers at the University of California, Davis created a prototype Distributed Intrusion Detection System (DIDS), which was also an expert system.[52] The Network Anomaly Detection and Intrusion Reporter (NADIR), also in 1991, was a prototype IDS developed at the Los Alamos National Laboratory's Integrated Computing Network (ICN), and was heavily influenced by the work of Denning and Lunt.[53] NADIR used a statistics-based anomaly detector and an expert system.
The Lawrence Berkeley National Laboratory announced Bro in 1998, which used its own rule language for packet analysis from libpcap data.[54] Network Flight Recorder (NFR) in 1999 also used libpcap.[55]
APE was developed as a packet sniffer, also using libpcap, in November, 1998, and was renamed Snort one month later. Snort has since become the world's largest used IDS/IPS system with over 300,000 active users.[56] It can monitor both local systems, and remote capture points using the TZSP protocol.
The Audit Data Analysis and Mining (ADAM) IDS in 2001 used tcpdump to build profiles of rules for classifications.[57] In 2003, Yongguang Zhang and Wenke Lee argue for the importance of IDS in networks with mobile nodes.[58]
In 2015, Viegas and his colleagues [59] proposed an anomaly-based intrusion detection engine, aiming System-on-Chip (SoC) for applications in Internet of Things (IoT), for instance. The proposal applies machine learning for anomaly detection, providing energy-efficiency to a Decision Tree, Naive-Bayes, and k-Nearest Neighbors classifiers implementation in an Atom CPU and its hardware-friendly implementation in a FPGA.[60][61] In the literature, this was the first work that implement each classifier equivalently in software and hardware and measures its energy consumption on both. Additionally, it was the first time that was measured the energy consumption for extracting each features used to make the network packet classification, implemented in software and hardware.[62]