Showing papers by "Muttukrishnan Rajarajan published in 2022"
TL;DR: Deep Contractive Autoencoder based Attribute Learning (DCAE-ZSL) technique as well as an IS method based on Heterogeneous Voting Ensemble was presented in this article .
Abstract: Ransomware attacks are hazardous cyber-attacks that use cryptographic methods to hold victims’ data until the ransom is paid. Zero-day ransomware attacks try to exploit new vulnerabilities and are considered a severe threat to existing security solutions and internet resources. In the case of zero-day attacks, training data is not available before the attack takes place. Therefore, we exploit Zero-shot Learning (ZSL) capabilities that can effectively deal with unseen classes compared to the traditional machine learning techniques. ZSL is a two-stage process comprising of: Attribute Learning (AL) and Inference Stage (IS). In this regard, this work presents a new Deep Contractive Autoencoder based Attribute Learning (DCAE-ZSL) technique as well as an IS method based on Heterogeneous Voting Ensemble (DCAE-ZSL-HVE). In the proposed DCAE-ZSL approach, Contractive Autoencoder (CAE) is employed to extract core features of known and unknown ransomware. The regularization term of CAE helps in penalizing the classifier's sensitivity against the small dissimilarities in the latent space. On the other hand, in case of the IS, four combination rules Global Majority (GM), Local Majority (LM), Cumulative Vote-against based Global Majority (CVAGM), Cumulative Vote-for based Global Majority (CVFGM) are utilized to find the final prediction. It is empirically shown that in comparison to conventional machine learning techniques, models trained on contractive embedding show reasonable performance against zero-day attacks. Furthermore, it is shown that the exploitation of these core features through the proposed voting based ensemble (DCAE-ZSL-HVE) has demonstrated significant improvement in detecting zero-day attacks (recall = 0.95) and reducing False Negative (FN = 6).
9 citations
TL;DR: In this article , a cost-sensitive Pareto ensemble strategy, CSPE-R, was proposed to detect zero-day ransomware attacks by exploiting the unsupervised deep Contractive Auto Encoder (CAE) to transform the underlying varying feature space to a more uniform and core semantic feature space.
Abstract: Abstract Ransomware attacks pose a serious threat to Internet resources due to their far-reaching effects. It’s Zero-day variants are even more hazardous, as less is known about them. In this regard, when used for ransomware attack detection, conventional machine learning approaches may become data-dependent, insensitive to error cost, and thus may not tackle zero-day ransomware attacks. Zero-day ransomware have normally unseen underlying data distribution. This paper presents a Cost-Sensitive Pareto Ensemble strategy, CSPE-R to detect novel Ransomware attacks. Initially, the proposed framework exploits the unsupervised deep Contractive Auto Encoder (CAE) to transform the underlying varying feature space to a more uniform and core semantic feature space. To learn the robust features, the proposed CSPE-R ensemble technique explores different semantic spaces at various levels of detail. Heterogeneous base estimators are then trained over these extracted subspaces to find the core relevance between the various families of the ransomware attacks. Then, a novel Pareto Ensemble-based estimator selection strategy is implemented to achieve a cost-sensitive compromise between false positives and false negatives. Finally, the decision of selected estimators are aggregated to improve the detection against unknown ransomware attacks. The experimental results show that the proposed CSPE-R framework performs well against zero-day ransomware attacks.
8 citations
TL;DR: In this article, the authors proposed a compromise propagation model, alongside a behavioural DDoS model, to explore how dependencies between the grid's networks might influence the scale and impact of DDoS attacks and found that the internal connectedness of a network amplifies the received impact of failures in an external network on which it is dependent.
4 citations
05 Sep 2022
TL;DR: Wang et al. as discussed by the authors designed a privacy-preserving data-aggregation scheme for a microgrid of prosumers that achieves high level of accuracy, thereby benefiting to the management and control of a micro-grid.
Abstract: The concept of a microgrid has emerged as a promising solution for the management of local groups of electricity consumers and producers. The use of end-users' energy usage data can help in increasing efficient operation of a microgrid. However, existing data-aggregation schemes for a microgrid suffer different cyber attacks and do not provide high level of accuracy. This work aims at designing a privacy-preserving data-aggregation scheme for a microgrid of prosumers that achieves high level of accuracy, thereby benefiting to the management and control of a microgrid. First, a novel smart meter readings data protection mechanism is proposed to ensure privacy of prosumers by hiding the real energy usage data from other parties. Secondly, a blockchain-based data-aggregation scheme is proposed to ensure privacy of the end-users, while achieving high level of accuracy in terms of the aggregated data. The proposed data-aggregation scheme is evaluated using real smart meter readings data from 100 prosumers. The results show that the proposed scheme ensures prosumers' privacy and achieves high level of accuracy, while it is secure against eavesdropping and man-in-the-middle cyber attacks.
TL;DR: Continuous authentication offers an intelligent solution to this problem, although its application within IoT is currently in its infancy, and the limitations of sensors, power and processing capabilities present challenges when compared to traditional user devices.
04 Dec 2022
TL;DR: In this article , a framework for open banking API security is proposed, which uses STRIDE model to identify security threats in FinTech integration via Open Banking API and Bayesian Attack Graphs to automate predictions of the most exploitable attack paths.
Abstract: Particularly amid Covid-19, enterprises' digital transformation has rapidly accelerated, making cybersecurity an even bigger challenge. Financial institutions adopt FinTech technologies to advance their service and achieve an enhanced customer experience that creates a competitive edge in the market. FinTech products utilise open banking API services to allow communication between a financial institution and a FinTech provider. However, such an integration introduces significant security concerns. Therefore, financial firms must ensure that a robust API service to protect the bank's infrastructure and its customers' information. To address this concern, we propose a Framework for Open Banking API security that utilises STRIDE model to identify security threats in FinTech integration via Open Banking API and Bayesian Attack Graphs to automate predictions of the most exploitable attack paths.
26 Sep 2022
TL;DR: In this article , the authors proposed a decentralized identity framework for Industrial Internet of Things (IIoT) based on Self-Sovereign Identity (SSI) model, which uses Decentralized Identifiers (DIDs) and Verifiable Credentials (VCs).
Abstract: The fundamental requirement for interaction between digital entities is a secure and privacy-preserving digital identity infrastructure. Traditional approaches rely heavily on centralized architectural components such as Certificate Authorities (CAs) and credential storage databases that have drawbacks like a single point of failure, attack prone honeypot databases and poor scalability. Self-Sovereign Identity (SSI) is a novel decentralized digital identity model that uses Decentralized Identifiers (DIDs) and Verifiable Credentials (VCs). In this work, we propose a novel decentralized identity framework for Industrial Internet-of-Things (IIoT) based on SSI model. The proposed framework is implemented on two blockchain platforms namely Ethereum and Hyperledger Indy to study the underlying overheads.
TL;DR: The security and privacy implications related to water memory are highlighted and the possible countermeasures to effectively handle these potential threats are discussed and a framework to securely store sensitive data on water is presented.
Abstract: The security of IoT devices is a major concern that needs to be addressed for their wide adoption. Users are constantly seeking devices that are faster and capable of holding large amounts of data securely. It is purported that water has memory of its own and the ability to retain memory of the substances that are dissolved into it, even after being substantially and serially diluted. It was also observed in the lab setting that the microscopic pattern of water obtained from the same vessel by different people is unique but can easily distinguish those individuals if the same experiment is executed repeatedly. Furthermore, extensive research is already underway that explores the storage of data on water and liquids. This leads to the requirement of taking the security and privacy concerns related to the storage of data on water into consideration, especially when the real-time collection of data related to water through the IoT devices is of interest. Otherwise, the water memory aspect may lead to leakage of the data and, consequently, the data owners identity. Therefore, this article for the first time highlights the security and privacy implications related to water memory and discusses the possible countermeasures to effectively handle these potential threats. This article also presents a framework to securely store sensitive data on water. The proof-of-concept prototype is implemented and tested over a real-world dataset to analyze the feasibility of the proposed framework. The performance analysis yields that the proposed framework can be deployed once data storage on water is widely used.