Artificial Intelligence for Cyber Security: Performance Analysis of Network Intrusion Detection

Attack-aware Security Function Management

This paper presents an overview of the technologies and the methodologies used in Network Intrusion Detection and Prevention Systems (NIDPS). Intrusion Detection and Prevention System (IDPS) technologies are differentiated by types of events that IDPSs can recognize, by types of devices that IDPSs monitor and by activity. NIDPSs monitor and analyze the streams of network packets in order to detect security incidents. The main methodology used by NIDPSs is protocol analysis. Protocol analysis requires good knowledge of the theory of the main protocols, their definition, how each protocol works.Keywords: Intrusion Detection and Prevention System, Protocol Analysis, Sensor, Signature, State(ProQuest: ... denotes formulae omitted.)1 IntroductionIncreasing size and complexity of the Internet and Intranet networks have led to increasing number of vulnerabilities that could be exploited. Thus, the internal and external attacks on the information systems are increasing at an alarming rate. Also, these are becoming more severe and sophisticated. The attackers find ingenious ways to bypass the security controls and to compromise the security and the well functioning of the information systems. They are motivated by financial, political, and military objectives. In this context, defending wide area networks from malicious traffic, unauthorized access to systems involves many problems.In security information systems Network Intrusion Detection and Prevention Systems (NIDPS) are important tools to detect possible incidents and also, to attempt to stop them in real time. Due to changing attacks, intrusion detection methodologies and technologies continuously evolve, adding new detection capabilities, to avoid detection. They must adapt to new forms of malware, to the public networks, increased traffic.2 Concepts of Intrusion DetectionAn intrusion is a successful action to gain access to an information system, to compromise it or to make it unavailable. This is possible due to the presence of vulnerability in the target system that can be exploited by a motivated intruder.Intrusion Detection and Prevention is the process of monitoring the information systems by sensors or agents and analyzing the collected information to detect and to attempt to stop the attacks in real time, identifying vulnerabilities, the violation of security policies or standard security practices.An Intrusion Detection and Prevention System (IDPS) is a tool that monitors information systems, collects, analyzes information, and initiates responses when an intrusion is detected.Intrusion Detection Systems (IDSs) mainly work as defensive mechanisms. They only alert the system administrators that an incident has occurred. Intrusion Prevention Systems (IPSs) can take some actions to attempt to stop the attack, such as breaking the connection or modifying the firewall rules to deny access to the intruder. The response of the classic IDS can be slow if the system administrator is busy while the response of the IPS is automatic. An architecture that uses together IPS and IDS technologies is the best solution for defense in depth.Conceptually, a generic IDPS consists of modular components. It mainly has the following components: monitoring system, storage, analyzer, and responder.* Monitoring system - monitors and logs the events in a computer system or network;* Storage - stores information, called audit record, about suspicious activities or intrusions; also, the security policies used in analysis are stored;* Analyzer - uses different analysis methodologies to detect the incidents;* Responder - the response mechanism of incidents.The IDPSs could be classified as:* By detection methodology [12], [18]:- misuse-based detection- anomaly-based detection- stateful protocol analysis* By activity [12]:- network-based- wireless-based- network behavior analysis- host-based* By behavior on detection:- passive- active* By collection and analysis frequency:- continuous- periodicThe detection methodologies describe the characteristics of the analyzer. …

Intrusion detection and prevention systems (IDPS) are security systems that are used to detect and prevent security threats to computer systems and computer networks. These systems are configured to detect and respond to security threats automatically there by reducing the risk to monitored computers and networks. Intrusion detection and prevention systems use different methodologies such as signature based, anomaly based, stateful protocol analysis, and a hybrid system that combines some or all of the other systems to detect and respond to security threats. The growth of systems that use a combination of methods creates some confusion when trying to choose a methodology and system to deploy. This paper seeks to offer a clear explanation of each methodology and then offer a way to compare these methodologies.

This paper presents Intrusion Detection Systems (IDS), Intrusion Detection and Prevention Systems (IDPS) and their classification emphasizing on the use of neural networks in IDS. Contemporary IDS usually include both signature verification and anomaly detection approaches realized by rule-based expert system and statistical module correspondingly. Neural networks may be used mainly as additional module to the statistical module to better recognize the user behavior. User behavior may be represented as frequency pattern of users command history. The paper presents an example of user profile vector for Unix-based platforms.

Although Software-defined networking (SDN) is a promising architecture that simplifies network management and control, it also faces security problems that may affect the whole network. Hence, protecting strategies, such as intrusion detection and prevention system (IDPS), are in need in the SDN context. The potential of machine learning-based solutions can become the motivation of cut-edge deep learning-based intrusion detection system that can leverage the centralized control and view of the controller to secure the underlying infrastructure. However, performing additional IDPS functions in the controller, which needs to process enormous traffic amounts, can overload this component, and slow down the network. This paper introduces an approach of Big Data analysis for intrusion detection system in SDN, named BIDSDN to enhance the classification perfor-mance with a massive amount of network traffic data. Specifically, we leverage Apache Spark to deploy the distributed deep learning – based detector to reduce the processing time on complex algorithms. The experiments conducted on CICIDS2018 dataset with distributed cluster prove the efficacy in tackling the Big Data-related issues in the large-scale network like SDN.

The paper presents an analytical model to study the performance and availability of queueing systems with finite queue and two service phases. The first phase has the exponential distribution of service time, while the second one has the hyperexponential distribution. The analytical results obtained are verified using discrete-event simulation. A few numerical examples for varying the service rates and arrival rates are given. The results presented in the paper can be used for analysis of the intrusion detection and prevention systems (IDPS).

Distributed Denial-of-Service (DDoS) attack has been a “nightmare” for cloud. A countermeasure is to establish an Intrusion Detection and Prevention System (IDPS) for cloud. Nevertheless, current IDPSes fail to achieve the detection and prevention in a flexible and lightweight way. In this paper, we propose a novel scheme of IDPS for overcoming the above problem, termed as Auto-scaling IDPS (AsIDPS). AsIDPS is based on Software-Defined Networking (SDN) and Docker container technologies. It first detects abnormal traffic based on the flow statistics collected in SDN switches in real-time. By the SDN controller, the abnormal traffic will be directed to the created Docker containers with Snort running on them for further detection and clean-up. Particularly, the Docker containers can be automatically scaled out or scaled down on demand. The Snort will also deliver an alert to the SDN controller if it detects attack traffic so as to perform a countermeasure if necessary. Benefitting from the flexible network management offered by SDN and the lightweight Docker container, AsIDPS is able to build a flexible and lightweight defense against DDoS attack in cloud. Based on our prototype implementation, we validate the effectiveness of AsIDPS in defending DDoS attack, and also verify its flexibility and lightweight.

The accelerated evolution of Industrial Internet of Things (IIoT) 5.0 brings human-centric automation, hyperconnectivity, and Artificial Intelligence (AI) based real-time data analytics. The growth presents cybersecurity challenges with an increased attack surface and advanced threats. Legacy Intrusion Detection and Prevention System (IDPS) are not scalable and lack the agility to deal with these threats. This chapter discusses AI-based IDPS through machine learning, deep learning, and FL to enable real-time threat detection. This chapter also mentions the importance of Explainable AI, blockchain, and edge AI in improving security. The future directions are quantum computing and light AI models, and this provides a roadmap to secure IIoT 5.0 against future cyber-attacks.

Routing protocols are responsible to enable efficient communication in any networking environment. Mobile Ad Hoc Networks are an infrastructure-less, distributed and peer-to-peer networks. Most of the routing protocols designed in this network are based on the consideration that the nodes in the network are supportive and cooperate for multi-hop communication. This consideration is not authentic in the hostile environment. The nodes violate the routing protocol specifications and drop the packets, and known as malicious nodes. However, reputed intermediate nodes also drop the packets whenever they receive the packets more than their handling capabilities, and known as bottleneck intermediate nodes. In order to overcome the situation, various intrusion detection and prevention systems (IDS) are designed. The aim of the IDS is to mitigate the packet dropping nodes from network, but they do not recognize the packet dropping node is either malicious or reputed. The paper aim is to investigate the performance of the existing IDS in the presence of the reputed packet dropping nodes in the network. Performance evaluation is carried out by network simulator, NS-2. The results show that neglecting the reputed packet dropping nodes by IDS algorithms is a serious problem and it negatively impacts on the network performance.KeywordsMANETIntrusion detection systemBottleneck nodeRoutingSimulation

IT sector is growing day by day, its data and energy costs are also increasing, which needs to access the high-end computing capabilities. To fulfill these requirements, there is a need to shift from traditional computing practices to new computing practices that give access to a broader network, unlimited resources, and enables self-service on demand at a reasonable cost using pay per use method. All these requirements can be fulfilled with cloud computing. Cloud Computing has a number of advantages as well as disadvantages. Security is a major threat to the cloud computing because the data placed on cloud servers is always vulnerable as there are lots of attackers on the internet who tries to compromise the security of the cloud. To overcome this issue, we use security appliances in the form of physical machines or by virtualization. A significant amount of power is consumed by security appliances which not only leads to the high energy costs but also a major threat to the environment as carbon-dioxide is released. So there is a need to find optimal solutions for providing security in cloud computing environment in an efficient manner. In this paper, a comprehensive survey on existing cloud security frameworks has been done. Based upon the limitations on existing frameworks a new framework has been proposed to provide security in virtual networks that is based on Intrusion Detection and Prevention System (IDPS) and its prototype implementation has also been done.

Modern cybersecurity relies on Intrusion Detection and Prevention Systems (IDPS) to detect and mitigate malicious network activity. This study proposes eIDPS, a packet-level IDPS leveraging Machine Learning and the extended Berkeley Packet Filter (eBPF) for real-time threat detection within the kernel. Unlike a flow-based approach, which analyzes aggregated network flows, eIDPS operates at the packet level, enabling faster decision-making while it does not drop entire flows when a malicious packet is detected, allowing for more precise threat mitigation. Experimental evaluation using CIC-IDS2017, a benchmark intrusion detection dataset, demonstrates that eIDPS achieves higher precision and F1-score compared to a flow-based solution identified in the literature, while still operating in real time. This comparison highlights the trade-offs between packetlevel and flow-based intrusion detection, offering insights into the deployment of eBPF-based security solutions.

Chapter 5 - Intrusion Prevention and Detection Systems

The rapid evolution of next-generation networks (NGNs), driven by 5G, IoT, edge computing, and software-defined networking, has introduced new opportunities alongside complex security challenges. Traditional intrusion detection and prevention systems (IDS), built on signature-based and anomaly-based methods, struggle to cope with the scale, heterogeneity, and dynamic threat landscape of NGNs. In response, artificial intelligence (AI) has emerged as a powerful enabler of modern IDPS. This review surveys AI-powered approaches, beginning with classical machine learning methods such as decision trees, support vector machines, and random forests, and then examining deep learning architectures including convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory networks (LSTMs), and autoencoders. It further analyses hybrid frameworks that integrate ensemble learning, federated learning, and meta-learning, as well as specialised methods tailored for SDN, IoT, edge, and cloud/5G environments. Benchmark datasets, including NSL-KDD, CICIDS2017, UNSW-NB15, Bot-IoT, IoT-23, and TON_IoT, are reviewed, highlighting their contributions and limitations. The paper identifies key challenges, including dataset scarcity, generalisation gaps, computational overhead, adversarial robustness, explainability, and privacy. Future directions emphasise the need for realistic NGN datasets, lightweight yet accurate architectures, privacy-preserving and federated frameworks, and integrated detection and prevention mechanisms. Overall, AI-powered IDPS demonstrate significant potential to secure NGNs, but realising this vision will require advances that balance accuracy, efficiency, interpretability, and resilience.

Intrusion Detection and Prevention Systems (IDPSs) play a crucial role in safeguarding online connections against unauthorized access and malicious activities. To enable efficient and effective detection and mitigation, IDPSs must continuously improve their performance due to the constantly developing nature of cyber threats. However, an IDPS is more difficult to use and less reliable when it deals with huge amounts of data. This study aimed to improve the performance of IDPSs by employing optimization algorithms to reduce the data size. Particularly, the Harris Hawks Optimization (HHO) and Whale Optimization Algorithm (WOA) were combined for feature selection. The experimental results showed that the performance of the proposed IDPS was greatly improved by combining the HHO and WOA algorithms. Combining a Random Forest classifier with the suggested HHO/WOA feature selection method achieved very high results in accuracy (99.17%), recall (98.76%), precision (98.76%), and F1-score (98.43%).

Intrusion detection and prevention systems (IDPSs) are at the core of protecting an enterprise's network. In general, IDPSs use pre-defined rules to detect potential attacks. As the size of an organization grows and new types of intrusions appear, the quantity and complexity of the rules also increase. Moreover, IDPSs generate an overwhelming number of logs that are challenging to handle and analyze. For a more effective and integrative analysis and management of the rules and logs, we propose a novel visual analytics tool, Hyperion. Hyperion interactively visualizes rules to help users understand how the IDPS rules are managed and applied to the enterprise's network entities. Hyperion also provides effective visualizations to enable users to visually analyze the type, period, traffic, and frequency of attacks in addition to a traditional count-based timeline visualization. Finally, Hyperion enables users to interactively simulate the effect of a change in parameters of a detection rule. These features can help streamline the security control cycle consisting of rule application, information collection, log analysis, and rule revision.