2026· ITEGAM- Journal of Engineering and Technology for Industrial Applications (ITEGAM-JETIA)· Vol 12, pp. 382-399· 0 citations
TL;DR
This work introduces an explained, risk-conscious, and secure-by-design autonomic architecture of large scale distributed computing systems defining dynamic cloud-edge environments that is able to implement transparent, resilient and scalable autonomous decision-making which is appropriate to mission critical distributed infrastructures.
Abstract
This work introduces an explained, risk-conscious, and secure-by-design autonomic architecture of large scale distributed computing systems defining dynamic cloud-edge environments. The main issue it is expected to fix is the shortage of transparency, uncertainty quantification, and combined cybersecurity of the traditional black-box AI-driven distributed systems that restricts trust, scalability, and resiliency. In order to address these limitations, the suggested methodology combines some structured data preprocessing (Z-score filtering, entropy and variance extraction), probabilistic risk modeling, explainable AI based on SHAP, and Kubernetes KEIO-enabled adaptive orchestration into a single layered framework. Experimental analysis on cloud anomaly data sets shows good predictive accuracy of 91.1 percent, recall of 86 percent as well as precision of 79 percent as represented in the confusion matrix of 3333 true negative, 1000 true positive, 261 false positive, and 160 false negative. Risk scores are centred on 0.0, with most of the risk scores falling within the range of - 0.3 to 0.3, which is stable with respect to calibration. System availability indicates 94% of uptime, Mean Time to Recovery (MTTR) of 3.45 and 48% of scalability efficiency, which all indicate high uptime and efficient fault management. SHAP analysis reveals that power consumption (0.28) and memory usage (0.27) had the largest contributions to risks. In general, the architecture is able to implement transparent, resilient and scalable autonomous decision-making which is appropriate to mission critical distributed infrastructures.
Microservice and cloud-native architectures have expanded the software attack surface at a pace that outstrips the adaptation of traditional risk assessment methods. Conventional vulnerability management remains centralized, static, severity-oriented, and disconnected from access control—characteristics that are ill-suited to distributed systems, where risk is shaped more by context and connectivity than by isolated weaknesses. This paper proposes a conceptual framework for distributed, AI-assisted risk assessment designed to address this gap. The framework synthesizes four bodies of literature—zero-trust architecture, microservice and cloud-native security, vulnerability analytics, and software risk management—and identifies precisely where each falls short. A six-dimensional taxonomy then structures the design space of distributed risk assessment, classifying approaches by signal source, granularity, analytical technique, risk model, zero-trust integration point, and temporality, and revealing a combination not yet explored by existing work. To fill this void, the paper introduces a four-layer framework: lightweight per-service agents collect multi-source evidence; AI-assisted analytics estimate exploit likelihood and detect anomalies; a graph-aware aggregation layer propagates risk across the service dependency graph; and an integration layer supplies a continuous per-service risk score to the zero-trust policy engine. An illustrative scenario demonstrates how a contextually significant but low-severity vulnerability would be surfaced and contained. As a conceptual contribution without experimental validation, the paper concludes by outlining key open challenges—data, robustness, performance, consistency, explainability, and evaluation—that must be addressed before the framework can be operationalized.
Daoquan Zhou, Xiongsheng Yi· AI and Data Science Journal· 0 citations
Software-defined networking (SDN) offers flexibility and scalability but also introduces new cybersecurity challenges. We propose a scalable and transparent intrusion detection system (IDS) for SDNs using machine learning models that balance accuracy and computational efficiency. Our results show that the decision tree model achieves 99.6% accuracy with minimal missed attacks and lower computational costs compared with random forests. The training time increased with the dataset size, but the prediction time remained stable, even with 24,000 samples. Our simulation results showed that the training complexity of Random Forest increases with the number of samples, whereas the calculation time for inference is comparatively low–a finding that makes Random Forest particularly suitable for real-time detections in Software-Defined Networking (SDN) environments. By using the SHAP-based explainability analysis, we explain the most important traffic characteristics, such as the length of the packets and the destination port, which are used to decide the decision process of the model. This methodological approach ensures high recall detection without compromising performance, thus providing a high-performance and efficient solution for SDN security.
Unknown authors· Journal of Computing & B...· 0 citations
Experimental evaluation demonstrates up to 95% detection accuracy, a 50% reduction in response latency, and scalability to over 100,000 IoT devices without performance degradation, highlighting the suitability of SC-ARS for deployment in smart cities, industrial IoT, and decentralized critical infrastructures where trust, transparency, and real-time responsiveness are essential.
S. Bassey, B. Stephen, Emediong Bassey Obot et al.· E3S Web of Conferences· 0 citations
This study presents a C*-algebraic framework for optimizing intrusion mitigation in Internet of Things (IoT)networks by integrating mathematical models for cyberattack propagation with optimization-based security strategies.Theoretical results demonstrate that the spectral radius of the attack operator ρ(A) governs the recovery of IoTnetworks under attack, where ρ(A) < 1 ensures system recovery, and ρ(A) ≥ 1 leads to persistent or growing attackimpact. The framework combines blockchain-based trust, AI-driven intrusion detection systems (IDS), and Zero-TrustArchitecture (ZTA) to provide a multi-layered, adaptive defence system. Unlike probabilistic models that simplifyattack dynamics, this approach rigorously models threats using bounded linear operators, thereby offering scalabilityand robustness. Optimization ensures computational efficiency, making the model suitable for resource-constrained IoTenvironments, with the operator norm and the spectral radius acting as key constraints. Validation on real-worlddatasets such as CIC-IoT2023, UNSW-NB15, and BoT-IoT revealed that the AI-IDS models achieved near-perfectperformance, while the unified model integrating blockchain, IDS, and ZTA showed an accuracy of 51.0
Mustapha Danjuma Suleiman· International Journal of Mat...· 0 citations
The rapid adoption of distributed intelligent technologies such as cloud computing, Internet of Things (IoT), artificial intelligence, and edge computing has significantly transformed modern digital infrastructures. While these technologies enable scalability, automation, and real‑time data processing, they also introduce complex cybersecurity challenges. Traditional perimeter‑based security models are no longer sufficient to protect dynamic and distributed environments where users, devices, and services interact continuously across heterogeneous networks. This paper proposes a Dynamic Trust‑based Zero Trust Architecture (DTZTA), an adaptive cybersecurity framework designed to enhance security and trust in distributed intelligent systems. The proposed framework integrates zero‑trust access control, dynamic trust evaluation, artificial intelligence driven anomaly detection, and automated response mechanisms. A multi‑layered architecture ensures continuous authentication, real‑time risk assessment, and intelligent threat mitigation. Experimental evaluation using intrusion detection datasets demonstrates that the proposed framework improves threat detection accuracy, reduces false positives, and enhances system scalability compared with traditional security models such as perimeter security, role‑based access control, and conventional intrusion detection systems. The DTZTA framework therefore provides an effective security solution for next‑generation distributed computing environments.
P. Yasodha· International Journal of Res...· 0 citations
Permissioned blockchain systems have emerged as a cornerstone for enterprise-grade distributed applications due to their controlled participation, high throughput, and deterministic consensus protocols. However, existing security mechanisms in such systems are still mostly static, based on pre-defined rules and deterministic validation logic that are not sufficient against evolving adversarial behaviours such as insider threats, transaction manipulation and stealthy anomaly patterns. This paper proposes a novel Explainable Artificial Intelligence (XAI) driven adaptive pre-validation framework, in which an intelligent dynamic decision-making layer is introduced before the blockchain transaction commitment. The framework proposes the usage of unsupervised anomaly detection (Isolation Forest), supervised ensemble classification (Random Forest), dynamic trust score, and explainability mechanisms (Shapley value-based attribution) for transparent and accountable transaction validation. Unlike traditional approaches, the system evaluates transactions in real time using behavioural patterns, prior trust, and contextual anomalies. The proposed framework provides a scalable, transparent, and adaptive security enhancement for enterprise blockchain environments by integrating explainable decision-making into the transaction validation process.
S. S, S. S, A. M et al.· 2026 7th International Confe...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
