Biblio

An IDS is a system that helps in detecting any kind of doubtful activity on a computer network. It is capable of identifying suspicious activities at both the levels i.e. locally at the system level and in transit at the network level. Since, the system does not have its own dataset as a result it is inefficient in identifying unknown attacks. In order to overcome this inefficiency, we make use of ML. ML assists in analysing and categorizing attacks on diverse datasets. In this study, the efficacy of eight machine learning algorithms based on KDD CUP99 is assessed. Based on our implementation and analysis, amongst the eight Algorithms considered here, Support Vector Machine (SVM), Random Forest (RF) and Decision Tree (DT) have the highest testing accuracy of which got SVM does have the highest accuracy

From an information-theoretic standpoint, the intrusion detection process can be examined. Given the IDS output(alarm data), we should have less uncertainty regarding the input (event data). We propose the Capability of Intrusion Detection (CID) measure, which is simply the ratio of mutual information between IDS input and output, and the input of entropy. CID has the desirable properties of (1) naturally accounting for all important aspects of detection capability, such as true positive rate, false positive rate, positive predictive value, negative predictive value, and base rate, (2) objectively providing an intrinsic measure of intrusion detection capability, and (3) being sensitive to IDS operation parameters. When finetuning an IDS, we believe that CID is the best performance metric to use. In terms of the IDS’ inherent ability to classify input data, the so obtained operation point is the best that it can achieve.

Industrial control systems (ICSs) have become more complex due to their increasing connectivity, heterogeneity and, autonomy. As a result, cyber-threats against such systems have been significantly increased as well. Since a compromised industrial system can easily lead to hazardous safety and security consequences, it is crucial to develop security countermeasures to protect coexisting IT systems and industrial physical processes being involved in modern ICSs. Accordingly, in this study, we propose a deep learning-based semantic anomaly detection framework to model the complex behavior of ICSs. In contrast to the related work assuming only simpler security threats targeting individual controllers in an ICS, we address multi-PLC attacks that are harder to detect as requiring to observe the overall system state alongside single-PLC attacks. Using industrial simulation and emulation frameworks, we create a realistic setup representing both the production and networking aspects of industrial systems and conduct some potential attacks. Our experimental results indicate that our model can detect single-PLC attacks with 95% accuracy and multi-PLC attacks with 80% accuracy and nearly 1% false positive rate.

In recent years, cyber-attacks on modern industrial control systems (ICS) have become more common and it acts as a victim to various kind of attackers. The percentage of attacked ICS computers in the world in 2021 is 39.6%. To identify the anomaly in a large database system is a challenging task. Deep-learning model provides better solutions for handling the huge dataset with good accuracy. On the other hand, real time datasets are highly imbalanced with their sample proportions. In this research, GAN based model, a supervised learning method which generates new fake samples that is similar to real samples has been proposed. GAN based adversarial training would address the class imbalance problem in real time datasets. Adversarial samples are combined with legitimate samples and shuffled via proper proportion and given as input to the classifiers. The generated data samples along with the original ones are classified using various machine learning classifiers and their performances have been evaluated. Gradient boosting was found to classify with 98% accuracy when compared to other

Industrial Control Systems (ICS) are increasingly facing the threat of False Data Injection (FDI) attacks. As an emerging intrusion detection scheme for ICS, process-based Intrusion Detection Systems (IDS) can effectively detect the anomalies caused by FDI attacks. Specifically, such IDS establishes anomaly detection model which can describe the normal pattern of industrial processes, then perform real-time anomaly detection on industrial process data. However, this method suffers low detection accuracy due to the complexity and instability of industrial processes. That is, the process data inherently contains sophisticated nonlinear spatial-temporal correlations which are hard to be explicitly described by anomaly detection model. In addition, the noise and disturbance in process data prevent the IDS from distinguishing the real anomaly events. In this paper, we propose an Anomaly Detection approach based on Robust Spatial-temporal Modeling (AD-RoSM). Concretely, to explicitly describe the spatial-temporal correlations within the process data, a neural based state estimation model is proposed by utilizing 1D CNN for temporal modeling and multi-head self attention mechanism for spatial modeling. To perform robust anomaly detection in the presence of noise and disturbance, a composite anomaly discrimination model is designed so that the outputs of the state estimation model can be analyzed with a combination of threshold strategy and entropy-based strategy. We conducted extensive experiments on two benchmark ICS security datasets to demonstrate the effectiveness of our approach.

The success of the web and the consequent rise in data sharing have made network security a challenge. Attackers from all around the world target PC installations. When an attack is successful, an electronic device's security is jeopardised. The intrusion implicitly includes any sort of behaviours that purport to think twice about the respectability, secrecy, or accessibility of an asset. Information is shielded from unauthorised clients' scrutiny by the integrity of a certain foundation. Accessibility refers to the framework that gives users of the framework true access to information. The word "classification" implies that data within a given frame is shielded from unauthorised access and public display. Consequently, a PC network is considered to be fully completed if the primary objectives of these three standards have been satisfactorily met. To assist in achieving these objectives, Intrusion Detection Systems have been developed with the fundamental purpose of scanning incoming traffic on computer networks for malicious intrusions.

Internet of Things devices collect and share data (IoT). Internet connections and emerging technologies like IoT offer privacy and security challenges, and this trend is anticipated to develop quickly. Internet of Things intrusions are everywhere. Businesses are investing more to detect these threats. Institutes choose accurate testing and verification procedures. In recent years, IoT utilisation has increasingly risen in healthcare. Where IoT applications gained popular among technologists. IoT devices' energy limits and scalability raise privacy and security problems. Experts struggle to make IoT devices more safe and private. This paper provides a machine-learning-based IDS for IoT network threats (ML-IDS). This study aims to implement ML-supervised IDS for IoT. We're going with a centralised, lightweight IDS. Here, we compare seven popular categorization techniques on three data sets. The decision tree algorithm shows the best intrusion detection results.

As one of the defensive solutions against cyberattacks, an Intrusion Detection System (IDS) plays an important role in observing the network state and alerting suspicious actions that can break down the system. There are many attempts of adopting Machine Learning (ML) in IDS to achieve high performance in intrusion detection. However, all of them necessitate a large amount of labeled data. In addition, labeling attack data is a time-consuming and expensive human-labor operation, it makes existing ML methods difficult to deploy in a new system or yields lower results due to a lack of labels on pre-trained data. To address these issues, we propose a semi-supervised IDS model that leverages Generative Adversarial Networks (GANs) and Adversarial AutoEncoder (AAE), called a semi-supervised adversarial autoencoder (SAAE). Our SAAE experimental results on two public datasets for benchmarking ML-based IDS, including NF-CSE-CIC-IDS2018 and NF-UNSW-NB15, demonstrate the effectiveness of AAE and GAN in case of using only a small number of labeled data. In particular, our approach outperforms other ML methods with the highest detection rates in spite of the scarcity of labeled data for model training, even with only 1% labeled data.

Intelligent connected vehicle platoon technology can reduce traffic congestion and vehicle fuel. However, attacks on the data transmitted by the platoon are one of the primary challenges encountered by the platoon during its travels. The false data injection (FDI) attack can lead to road congestion and even vehicle collisions, which can impact the platoon. However, the complexity of the cellular - vehicle to everything (C-V2X) environment, the single source of the message and the poor data processing capability of the on board unit (OBU) make the traditional detection methods’ success rate and response time poor. This study proposes a platoon state information fusion method using the communication characteristics of the platoon in C-V2X and proposes a novel platoon intrusion detection model based on this fusion method combined with sequential importance sampling (SIS). The SIS is a measured strategy of Monte Carlo integration sampling. Specifically, the method takes the status information of the platoon members as the predicted value input. It uses the leader vehicle status information as the posterior probability of the observed value to the current moment of the platoon members. The posterior probabilities of the platoon members and the weights of the platoon members at the last moment are used as input to update the weights of the platoon members at the current moment and obtain the desired platoon status information at the present moment. Moreover, it compares the status information of the platoon members with the desired status information to detect attacks on the platoon. Finally, the effectiveness of the method is demonstrated by simulation.

With the continues growth of our technology, majority in our sectors are becoming smart and one of its great applications is in agriculture, which we call it as smart farming. The application of sensors, IoT, artificial intelligence, networking in the agricultural setting with the main purpose of increasing crop production and security level. With this advancement in farming, this provides a lot of privileges like remote monitoring, optimization of produce and too many to mention. In light of the thorough systematic analysis performed in this study, it was discovered that Edge-of-things is a potential computing scheme that could boost an artificial intelligence for intelligent remote farm security and intrusion detection pre-alarm system over other computing schemes. Again, the purpose of this study is not to replace existing cloud computing, but rather to highlight the potential of the Edge. The Edge architecture improves end-user experience by improving the time-related response of the system. response time of the system. One of the strengths of this system is to provide time-critical response service to make a decision with almost no delay, making it ideal for a farm security setting. Moreover, this study discussed the comparative analysis of Cloud, Fog and Edge in relation to farm security, the demand for a farm security system and the tools needed to materialize an Edge computing in a farm environment.

Train operation is highly influenced by the rail track state and the surrounding environment. An abnormal obstacle on the rail track will pose a severe threat to the safe operation of urban rail transit. The existing general obstacle detection approaches do not consider the specific urban rail environment and requirements. In this paper, we propose an edge intelligence (EI)-based obstacle intrusion detection system to detect accurate obstacle intrusion in real-time. A two-stage lightweight deep learning model is designed to detect obstacle intrusion and obtain the distance from the train to the obstacle. Edge computing (EC) and 5G are used to conduct the detection model and improve the real-time detection performance. A multi-agent reinforcement learning-based offloading and service migration model is formulated to optimize the edge computing resource. Experimental results show that the two-stage intrusion detection model with the reinforcement learning (RL)-based edge resource optimization model can achieve higher detection accuracy and real-time performance compared to traditional methods.

The main intention of edge computing is to improve network performance by storing and computing data at the edge of the network near the end user. However, its rapid development largely ignores security threats in large-scale computing platforms and their capable applications. Therefore, Security and privacy are crucial need for edge computing and edge computing based environment. Security vulnerabilities in edge computing systems lead to security threats affecting edge computing networks. Therefore, there is a basic need for an intrusion detection system (IDS) designed for edge computing to mitigate security attacks. Due to recent attacks, traditional algorithms may not be possibility for edge computing. This article outlines the latest IDS designed for edge computing and focuses on the corresponding methods, functions and mechanisms. This review also provides deep understanding of emerging security attacks in edge computing. This article proves that although the design and implementation of edge computing IDS have been studied previously, the development of efficient, reliable and powerful IDS for edge computing systems is still a crucial task. At the end of the review, the IDS developed will be introduced as a future prospect.

The Internet has evolved to the point that gigabytes and even terabytes of data are generated and processed on a daily basis. Such a stream of data is characterised by high volume, velocity and variety and is referred to as Big Data. Traditional data processing tools can no longer be used to process big data, because they were not designed to handle such a massive amount of data. This problem concerns also cyber security, where tools like intrusion detection systems employ classification algorithms to analyse the network traffic. Achieving a high accuracy attack detection becomes harder when the amount of data increases and the algorithms must be efficient enough to keep up with the throughput of a huge data stream. Due to the challenges posed by a big data environment, some monitoring systems have already shifted from deep packet inspection to flow-level inspection. The goal of this paper is to evaluate the applicability of an existing intrusion detection technique that performs deep packet inspection in a big data setting. We have conducted several experiments with Apache Spark to assess the performance of the technique when classifying anomalous packets, showing that it benefits from the use of Spark.

Machine learning (ML) models are increasingly being used in the development of Malware Detection Systems. Existing research in this area primarily focuses on developing new architectures and feature representation techniques to improve the accuracy of the model. However, recent studies have shown that existing state-of-the art techniques are vulnerable to adversarial machine learning (AML) attacks. Among those, data poisoning attacks have been identified as a top concern for ML practitioners. A recent study on clean-label poisoning attacks in which an adversary intentionally crafts training samples in order for the model to learn a backdoor watermark was shown to degrade the performance of state-of-the-art classifiers. Defenses against such poisoning attacks have been largely under-explored. We investigate a recently proposed clean-label poisoning attack and leverage an ensemble-based Nested Training technique to remove most of the poisoned samples from a poisoned training dataset. Our technique leverages the relatively large sensitivity of poisoned samples to feature noise that disproportionately affects the accuracy of a backdoored model. In particular, we show that for two state-of-the art architectures trained on the EMBER dataset affected by the clean-label attack, the Nested Training approach improves the accuracy of backdoor malware samples from 3.42% to 93.2%. We also show that samples produced by the clean-label attack often successfully evade malware classification even when the classifier is not poisoned during training. However, even in such scenarios, our Nested Training technique can mitigate the effect of such clean-label-based evasion attacks by recovering the model's accuracy of malware detection from 3.57% to 93.2%.

Security of Internet of Things (IoT) is one of the most prevalent crucial challenges ever since. The diversified devices and their specification along with resource constrained protocols made it more complex to address over all security need of IoT. Denial of Service attacks, being the most powerful and frequent attacks on IoT have been considered so forth. However, the attack happens on multiple layers and thus a single detection technique for each layer is not sufficient and effective to combat these attacks. Current study focuses on cross layer intrusion detection system (IDS) for detection of multiple Denial of Service (DoS) attacks. Presently, two attacks at Transmission Control Protocol (TCP) and Routing Protocol are considered for Low power and Lossy Networks (RPL) and a neural network-based IDS approach has been proposed for the detection of such attacks. The attacks are simulated on NetSim and detection and the performance shows up to 80% detection probabilities.

As the IPv6 protocol has been rapidly developed and applied, the security of IPv6 networks has become the focus of academic and industrial attention. Despite the fact that the IPv6 protocol is designed with security in mind, due to insufficient defense measures of current firewalls and intrusion detection systems for IPv6 networks, the construction of covert channels using fields not defined or reserved in IPv6 protocols may compromise the information systems. By discussing the possibility of constructing storage covert channels within IPv6 protocol fields, 10 types of IPv6 covert channels are constructed with undefined and reserved fields, including the flow label field, the traffic class field of IPv6 header, the reserved fields of IPv6 extension headers and the code field of ICMPv6 header. An IPv6 covert channel detection method based on field matching (CC-Guard) is proposed, and a typical IPv6 network environment is built for testing. In comparison with existing detection tools, the experimental results show that the CC-Guard not only can detect more covert channels consisting of IPv6 extension headers and ICMPv6 headers, but also achieves real-time detection with a lower detection overhead.

The (IoT) paradigm’s fundamental goal is to massively connect the “smart things” through standardized interfaces, providing a variety of smart services. Cyber-Physical Systems (CPS) include both physical and cyber components and can apply to various application domains (smart grid, smart transportation, smart manufacturing, etc.). The Digital Twin (DT) is a cyber clone of physical objects (things), which will be an essential component in CPS. This paper designs a systematic taxonomy to explore different attacks on DT-based CPS and how they affect the system from a four-layer architecture perspective. We present an attack space for DT-based CPS on four layers (i.e., object layer, communication layer, DT layer, and application layer), three attack objects (i.e., confidentiality, integrity, and availability), and attack types combined with strength and knowledge. Furthermore, some selected case studies are conducted to examine attacks on representative DT-based CPS (smart grid, smart transportation, and smart manufacturing). Finally, we propose a defense mechanism called Secured DT Development Life Cycle (SDTDLC) and point out the importance of leveraging other enabling techniques (intrusion detection, blockchain, modeling, simulation, and emulation) to secure DT-based CPS.

The Controller area network (CAN) is the most extensively used in-vehicle network. It is set to enable communication between a number of electronic control units (ECU) that are widely found in most modern vehicles. CAN is the de facto in-vehicle network standard due to its error avoidance techniques and similar features, but it is vulnerable to various attacks. In this research, we propose a CAN bus intrusion detection system (IDS) based on convolutional neural networks (CNN). U-CAN is a segmentation model that is trained by monitoring CAN traffic data that are preprocessed using hamming distance and saliency detection algorithm. The model is trained and tested using publicly available datasets of raw and reverse-engineered CAN frames. With an F\_1 Score of 0.997, U-CAN can detect DoS, Fuzzy, spoofing gear, and spoofing RPM attacks of the publicly available raw CAN frames. The model trained on reverse-engineered CAN signals that contain plateau attacks also results in a true positive rate and false-positive rate of 0.971 and 0.998, respectively.

Intrusion detection for Controller Area Network (CAN) protocol requires modern methods in order to compete with other electrical architectures. Fingerprint Intrusion Detection Systems (IDS) provide a promising new approach to solve this problem. By characterizing network traffic from known ECUs, hazardous messages can be discriminated. In this article, a modified version of Fingerprint IDS is employed utilizing both step response and spectral characterization of network traffic via neural network training. With the addition of feature set reduction and hyperparameter tuning, this method accomplishes a 99.4% detection rate of trusted ECU traffic.

International regulations specified in WP.29 and international standards specified in ISO/SAE 21434 require security operations such as cyberattack detection and incident responses to protect vehicles from cyberattacks. To meet these requirements, many vehicle manufacturers are planning to install Intrusion Detection Systems (IDSs) in the Controller Area Network (CAN), which is a primary component of in-vehicle networks, in the coming years. Besides, many vehicle manufacturers and information security companies are developing technologies to identify attack paths related to IDS alerts to respond to cyberattacks appropriately and quickly. To develop the IDSs and the technologies to identify attack paths, it is essential to grasp normal communications performed on in-vehicle networks. Thus, our study aims to develop a technology that can easily grasp normal communications performed on in-vehicle networks. In this paper, we propose the first message source identification method that easily identifies CAN-IDs used by each Electronic Control Unit (ECU) connected to the CAN for message transmissions. We realize the proposed method by utilizing diagnostic communications and an IDS installed in the CAN (CAN-IDS). We evaluate the proposed method using an ECU installed in an actual vehicle and four kinds of simulated CAN-IDSs based on typical existing intrusion detection methods for the CAN. The evaluation results show that the proposed method can identify the CAN-ID used by the ECU for CAN message transmissions if a suitable simulated CAN-IDS for the proposed method is connected to the vehicle.

The emergence of smart cars has revolutionized the automotive industry. Today's vehicles are equipped with different types of electronic control units (ECUs) that enable autonomous functionalities like self-driving, self-parking, lane keeping, and collision avoidance. The ECUs are connected to each other through an in-vehicle network, named Controller Area Network. In this talk, we will present the different cyber attacks that target autonomous vehicles and explain how an intrusion detection system (IDS) using machine learning can play a role in securing the Controller Area Network. We will also discuss the main research contributions for the security of autonomous vehicles. Specifically, we will describe our IDS, named Histogram-based Intrusion Detection and Filtering framework. Next, we will talk about the machine learning explainability issue that limits the acceptability of machine learning in autonomous vehicles, and how it can be addressed using our novel intrusion detection system based on rule extraction methods from Deep Neural Networks.

In recent years, as an important part of the Internet, web applications have gradually penetrated into life. Now enterprises, units and institutions are using web applications regardless of size. Intrusion detection to effectively identify malicious traffic has become an inevitable requirement for the development of network security technology. In addition, the proportion of deserialization vulnerabilities is increasing. Traditional intrusion detection mostly focuses on the identification of SQL injection, XSS, and command execution, and there are few studies on the identification of deserialization attack traffic. This paper use a method to extracts relevant features from the deserialized traffic or even the obfuscated deserialized traffic by reorganizing the traffic and running the relevant content through simulation, and combines deep learning technology to make judgments to efficiently identify deserialization attacks. Finally, a prototype system was designed to capture related attacks in real-world. The technology can be used in the field of malicious traffic detection and help combat Internet crimes in the future.

This Cognitive radio networks are vulnerable to specific intrusions due to the unique cognitive characteristics of these networks. This DoS attacks are known as the Primary User Emulation Attack and the Spectrum Sensing Data Falsification. If the intruder behavior is not statistically identical to the behavior of the primary users, intrusion detection techniques based on observing the energy of the received signals can be used. Both machine learning-based intrusion detection and sequential statistical analysis can be effectively applied. However, in some cases, statistical sequential analysis has some advantages in dealing with such challenges. This paper discusses aspects of using statistical sequential analysis methods to detect attacks in Cognitive radio networks.

In this work, we discuss data breaches based on the “2012 SocialArks data breach” case study. Data leakage refers to the security violations of unauthorized individuals copying, transmitting, viewing, stealing, or using sensitive, protected, or confidential data. Data leakage is becoming more and more serious, for those traditional information security protection methods like anti-virus software, intrusion detection, and firewalls have been becoming more and more challenging to deal with independently. Nevertheless, fortunately, new IT technologies are rapidly changing and challenging traditional security laws and provide new opportunities to develop the information security market. The SocialArks data breach was caused by a misconfiguration of ElasticSearch Database owned by SocialArks, owned by “Tencent.” The attack methodology is classic, and five common Elasticsearch mistakes discussed the possibilities of those leakages. The defense solution focuses on how to optimize the Elasticsearch server. Furthermore, the ElasticSearch database’s open-source identity also causes many ethical problems, which means that anyone can download and install it for free, and they can install it almost anywhere. Some companies download it and install it on their internal servers, while others download and install it in the cloud (on any provider they want). There are also cloud service companies that provide hosted versions of Elasticsearch, which means they host and manage Elasticsearch clusters for their customers, such as Company Tencent.

The spread of Internet of Things (IoT) devices in our homes, healthcare, industries etc. are more easily infiltrated than desktop computers have resulted in a surge in botnet attacks based on IoT devices, which may jeopardize the IoT security. Hence, there is a need to detect these attacks and mitigate the damage. Existing systems rely on supervised learning-based intrusion detection methods, which require a large labelled data set to achieve high accuracy. Botnets are onerous to detect because of stealthy command & control protocols and large amount of network traffic and hence obtaining a large labelled data set is also difficult. Due to unlabeled Network traffic, the supervised classification techniques may not be used directly to sort out the botnet that is responsible for the attack. To overcome this limitation, a semi-supervised Deep Learning (DL) approach is proposed which uses Semi-supervised GAN (SGAN) for IoT botnet detection on N-BaIoT dataset which contains "Bashlite" and "Mirai" attacks along with their sub attacks. The results have been compared with the state-of-the-art supervised solutions and found efficient in terms of better accuracy which is 99.89% in binary classification and 59% in multi classification on larger dataset, faster and reliable model for IoT Botnet detection.