Institute of Communication and Computer Systems

This work presents a numerical investigation of the applicability of the product of formaldehyde (CH 2 O) and hydroxyl radicals (OH), denoted as CH 2 O ⋅ OH, as a marker for the local heat release rate (HRR) in CH 4 flames over a broad range of boundary conditions. Both laminar freely propagating one-dimensional and premixed counter-flow CH 4 flames were simulated using several detailed reaction mechanisms. The analysis systematically varied the oxidizer composition (from air to pure oxy-fuel), equivalence ratio ( Φ ), inlet temperature, pressure, and strain rate. The correlation between the CH 2 O ⋅ OH profiles and the HRR was quantified using the Pearson correlation coefficient. Overall, the study confirms CH 2 O ⋅ OH as a practical and reliable HRR marker in CH 4 flames, particularly under high-temperature and high-pressure oxy-fuel conditions. The results show that in slightly rich CH 4 -air flames (about Φ = 1.5 ), CH 2 O ⋅ OH provides an accurate estimate of the HRR. In oxygen-enriched and pure oxy-fuel flames, the strongest correlation shifts to ultra-rich conditions (about Φ ≈ 3.0 ). Increasing pressure enhances the correlation at lower equivalence ratios, whereas preheating weakens the correlation locally but expands the overall range of validity. Although increasing strain rate generally degrades the correlation, CH 2 O ⋅ OH remains a robust marker under technically relevant turbulent conditions (strain rates below 10,000 s − 1 ), maintaining Pearson correlation coefficients of approximately R ≈ 0.9 even close to extinction. This demonstrates the strong feasibility of using CH 2 O ⋅ OH for HRR estimation in highly turbulent flames. The demonstrated robustness of CH 2 O ⋅ OH highlights its suitability as an HRR marker, providing valuable insights for experimental diagnostics. • CH 2 O ⋅ OH confirms as robust HRR marker across a wide range of conditions including variations in oxidizer content, equivalence ratio, preheat temperature, pressure, and strain. • For CH 4 -air flames, the highest correlation between CH 2 O ⋅ OH and HRR occurs at slightly rich mixtures ( Φ ≈ 1.5), while in CH 4 /O 2 flames, the strongest correlation shifts to ultra-rich conditions ( Φ ≈ 3.0). • Elevated pressure generally improves the correlation at leaner mixtures but narrows the range of high correlation, whereas preheating shifts the optimal equivalence ratio to richer conditions and broadens the applicability range. • Even under high strain rates (up to 10,000 s − 1 , relevant for turbulent flames), CH 2 O ⋅ OH maintains a strong correlation (R ≈ 0.9) with HRR, highlighting its suitability for turbulent combustion diagnostics. • The results of this study are intended to support and guide experimental diagnostics by identifying reliable marker regimes.

Steels remain among the most widely used structural and engineering materials in modern infrastructure, energy systems, and industrial facilities. Their long-term reliability depends critically on the early detection of corrosion damage and microstructural degradation. This review surveys recent advances in two complementary families of non-destructive evaluation (NDE) methods: magnetic techniques, including magnetic Barkhausen noise (MBN), magnetic flux leakage (MFL), eddy current testing (ECT), and magnetic hysteresis analysis; and electrochemical methods including electrochemical impedance spectroscopy (EIS), linear polarization resistance (LPR), scanning vibrating electrode technique (SVET), and electrochemical noise (EN). Recent progress in sensor miniaturization, signal processing algorithms, and multi-technique integration is reviewed. Particular attention is given to the sensitivity of these methods to microstructural changes reported in the literature, including carbide dissolution, phase transformations, temper embrittlement, and sensitization in stainless steels, as well as to the conditions under which such sensitivity has been demonstrated. The potential synergy between magnetic and electrochemical monitoring is discussed as a possible pathway toward more robust, condition-based maintenance frameworks. Challenges related to field deployment, environmental interference, calibration, and data interpretation are identified, and future directions—including machine learning-assisted analysis and multi-physics sensor arrays—are outlined.

The growing complexity of cyber threats coupled with the widening cybersecurity knowledge gap presents new challenges in the organisational security domain of the business sector. This paper introduces SentinelSphere, an innovative platform that redesigns cybersecurity defense by integrating advanced threat detection with interactive cybersecurity awareness education, creating a unified approach to building organisational cyber resilience. SentinelSphere employs an Enhanced Deep Neural Network model with specialised feature engineering to significantly reduce false positives while maintaining high detection accuracy across diverse attack vectors. The system includes a Traffic Light System (TLS) to transform complex threat intelligence into intuitive visual indicators, serving simultaneously as an operational tool for security professionals and an educational interface for non-technical users. This paper further presents a Large Language Model that delivers real-time, context-aware cybersecurity guidance and training. This conversational AI agent operates efficiently on standard enterprise hardware, making advanced security education accessible without requiring specialised infrastructure. SentinelSphere is validated using industry-standard datasets, achieving enterprise-grade performance in threat detection with a 94% F1 score and 69.5% reduction in false positives compared to baseline models. The system successfully processed nearly 11 million security events in 30 minutes, demonstrating scalability for enterprise deployment. This work contributes to the cybersecurity field by demonstrating that effective defense requires not just technological sophistication but also systematic enhancement of human security awareness.

Edge computing is poised to become a cornerstone of the emerging 6G landscape, where an ever-growing class of ultra-low-latency applications must be served close to the user. Despite its promise, real-world deployments remain nascent, with large-scale implementations anticipated by both Communication and Digital Service Providers (CSPs/DSPs) within the following years. Consequently, strategic edge network design is essential not only to maximize performance, but also to avoid redundant investments that can lead to an increased sum of Capital and Operational Expenditures (CAPEX/OPEX). In this work, we address a tri-fold problem: (i) the selection of deployment locations, (ii) the configuration of devices at the chosen location sites, and (iii) the assignment of the projected workload. Our objective is formulated as a weighted combination of the edge infrastructure’s establishment cost, the expected cumulative workload latency and the total expected energy consumption in the operational phase. To capture the spatial and temporal variability of demand, we solve the assignment subproblem over distinct snapshots, each representing a unique workload projection. We first present a Mixed Integer Linear Programming (MILP) formulation that yields the optimal solution; however, due to its computational intractability, we propose a novel adaptation of the Multi-Objective Evolutionary Algorithm by Decomposition (MOEA/D), with an embedded heuristic algorithm to assist in the chromosome fitness calculation. This method leverages the similarity among neighboring subproblems in a multi-objective framework to efficiently approximate the underlying Pareto frontier. In the experiments, the proposed method is contrasted with a sophisticated single-objective Rollout approach. Our results demonstrate the benefits of adopting a multi-objective algorithm in terms of performance, stability and interpretability across different scalarized subproblems. The proposed framework offers a practical intent-based decision support tool for edge infrastructure providers, weighing CAPEX against operating objectives ahead of initial deployment.

Abstract Stroke remains a leading cause of morbidity and mortality worldwide, necessitating rapid and accurate diagnosis to improve patient outcomes. The gold standard in stroke diagnosis relies on a costly and time-consuming combination of clinical evaluation and neuroimaging techniques, while also being associated with misdiagnoses due to mimicking conditions. This highlights the need for improving the current healthcare system towards faster, more accurate, and low-resource diagnostics. In this direction, this study presents the development of a novel artificial intelligence-driven decision support system for stroke diagnosis using machine learning techniques to fuse standard EHR variables (electronic health records) with information derived from extracellular vesicles (EVs) in blood plasma, which hold promise as diagnostic markers due to their association with stroke pathophysiology. As such, flow cytometry was employed to measure blood samples from 140 patients suffering from stroke or stroke-mimicking conditions. The derived EV-features, including statistical characteristics of their diameters’ distribution and concentration, were integrated with demographical, clinical, and biochemical patient data to form a comprehensive feature set. By utilizing a decision tree learning model, the proposed system achieved a classification weighted accuracy of 86%. This achievement indicates the approach’s potential for integrating cutting-edge photonics-based signals with standard HER information to improve clinical decision-making in emergency settings and for aiding treatment strategies in stroke care.

Breath analysis is a non-invasive diagnostic method that offers insights into both physiological and pathological conditions. Exhaled breath contains volatile organic compounds, which act as biomarkers for disease detection, allowing for the monitoring of treatments and the tailoring of medicine to individuals. Recent advancements in chemical sensing, mass spectrometry, and spectroscopy have improved the ability to identify these biomarkers; however, traditional statistical approaches often struggle to handle the complexities of breath data. Artificial intelligence (AI) and machine learning (ML) have revolutionized breath analysis by uncovering intricate patterns among volatile breath markers, enhancing diagnostic precision, and facilitating real-time disease identification. Despite significant progress, challenges remain, including issues with data standardization, model interpretability, and the necessity for extensive and varied datasets. This study reviews the applications of ML in analyzing breath volatile organic compounds, highlighting methodological shortcomings and obstacles to clinical validation. A thorough literature review was performed using the PubMed and Scopus databases, which included studies that focused specifically on the role of machine learning in disease diagnosis and incidence prediction via breath analysis. Among the 524 articles reviewed, 97 satisfied the specified inclusion criteria. The selected studies applied ML techniques, fell within the scope of this review, and emphasize the potential of ML models for non-invasive diagnostics. The findings indicate that traditional ML methods dominate, while ensemble methods are on the rise, and deep learning (DL) techniques (especially CNNs and LSTMs) are increasingly used for classifying respiratory diseases. Techniques for feature selection (such as PCA and ML-based methods) were frequently implemented, though challenges related to explainability and data standardization persist. Future studies should focus on enhancing model transparency and developing methods to further integrate AI into the clinical setting to facilitate early disease detection and advance precision medicine.

The growing demand for responsive and scalable computing services across cloud, fog, and edge resources calls for intelligent orchestration strategies that can adapt to dynamic workloads and network conditions. Efficient service orchestration across the computing continuum remains a key challenge in distributed systems, especially when optimizing network-centric metrics such as end-to-end delay, total traffic, and maximum link utilization. In this paper, we propose a multi-objective Deep Q-Network (DQN)-based framework for dynamic service orchestration that learns to deploy services in the best network locations optimising the use of network resources and meeting application demands. Unlike traditional heuristic as well as static Q-learning approaches [1] that also suffer from exponential state space growth due to combinatorial placement possibilities, our method leverages function approximation through deep reinforcement learning, significantly reducing the dimensionality and improving scalability. Our solution continuously learns from dynamic network conditions and manages to cope with varying workload intensities enabling adaptive service deployment in the network considering multiple network-related objectives, including total network traffic, maximum link utilization, and average communication delay. We evaluate the approach on synthetic and real-world topologies with medium and high-load scenarios, demonstrating its capacity to adapt to demand changes, achieving superior performance in minimizing network congestion and latency while reducing the computational complexity associated with large state spaces. The results highlight the effectiveness of DQN in orchestrating services in large-scale, dynamic network and compute infrastructures, making it a promising solution for future computing continuum systems.

Study introduces an advanced methodology designed to optimally determine the locations of electric vehicle (EV) charging stations within smart distribution grids. Recognizing the strategic importance and limited availability of these stations, the research addresses their substantial impact on network performance indicators. Scheme minimize a composite objective function that accounts for installation expenses, energy usage costs, and deviations of system voltage. Constraints include AC power flow requirements, operational limitations of EVs and their charging infrastructures, and inherent network capacities. A distinctive aspect of this approach is the utilization of Vehicle-To-Grid (V2G) technology, facilitating bidirectional power flow and enabling both active and reactive power exchanges with the grid. Validation of the proposed strategy was conducted on a 33-bus radial distribution system. The findings reveal that strategically positioned EV charging stations across network feeders significantly improve voltage profiles and reduce energy losses. Coupling these stations with microturbine sources further enhances voltage stability and reduces overall energy expenditures. Study demonstrates that careful planning and integration of EV charging infrastructure yield tangible improvements in both grid reliability and economic efficiency.