Qom University of Technology

As Fifth-generation (5G) networks advance towards future 6G technology, they provide enhanced capabilities in terms of data rates, latency, and connectivity. The expanding and heterogeneous landscape further extends the attack surface with increasing vulnerability to a wide range of dynamic cyber threats. Traditional Intrusion Detection Systems, relying on static signature-based methods, face significant challenges in detecting novel attacks in such complex environments. This paper proposes an explainable hybrid AE-XGBoost framework to address these gaps. In this proposed framework, the deep autoencoder serves as the backbone for two primary tasks: learning robust features from high-dimensional traffic data and synthesizing high-quality samples for minority, underrepresented attack classes. It effectively forms a class-balanced dataset. This is further extended to integrate Shapley Additive Explanations (SHAP), which provide needed interpretability. The framework robustness is determined by an extensive evaluation strategy. Extensive experiments are conducted using a 5-fold cross-validation and cross-dataset validation using three different datasets. The experimental results demonstrate the model effectiveness and promising generalization across the different datasets. Across these settings, the proposed IDS achieves F1-scores between approximately 0.96 and 0.999 and AUC values above 0.97, with only minor performance differences between original and synthetic test samples. SHAP-based analysis further reveals that a small subset of 5G- and TCP-level features consistently dominates the predictions, offering actionable explanations for security analysts. This provides a successful, scalable, and reliable security measure that proves effective and appropriate for 6G mobile network implementations in the near future.

This paper presents a hybrid machine learning model to analyze user behavior patterns in educational mobile applications for engagement frequency, session lengths, and interaction events. The new model combines Decision Trees and Neural Networks into a unique system that predicts learning outcomes accurately rather than deviating toward accuracy-guided prediction models. The research data has been collected over the past six months with approximately 10,000 middle school students aged 12-15 using the educational app, all of which contributed to a dataset of 50,000 user sessions. So extensive data collection was, as discussed in the Data Collection section, that it enabled our model to outperform standard methods such as logistic regression, k-NN, and support vector machines at an accuracy of 91 percent. This approach can be applied to app design to personalize learning experiences and teacher training approaches, focusing on the frequency of engagement and range of interaction diversity. The findings will facilitate the development of enhanced, student-centered educational apps, while personalized learning environments will provide useful information to the developers and educators on optimizing use for learning in mobile educational App development.

Breast cancer, a common disease in women globally, is caused by abnormal cell growth in the breast, forming tumors with potential fatality if not detected early. Breast cancer, a major global cause of death, necessitates the development of early-stage diagnostic tests. In this article, the authors show the Enhanced Breast Cancer Histopathology Vectorized Detection (EBCHVD) model that combines deep learning with topological data analysis in order to differentiate breast tissue samples as benign or malignant. For its detailed images, the model takes the Break His 400× histopathology dataset and applies preprocessing techniques followed by a meta-learning ensemble made up of different Convolutional Neural Networks (CNNs) to capture diverse image representations. the performance analysis is benign-malignant of f24 ≈ 0.45, f30 ≈ 0.12, loss of 0.70, accuracy of 0.65, confusion matrix of 86.5%, and ROC curve of 81% calculations.

Multilevel inverters are increasingly finding new applications in renewable energy. However, they have some technical challenges that diversely affect their performance. One of the most important concerns in these inverters is the DC-link capacitor voltage balancing. This paper presents an open-loop Carrier-based Pulse Width Modulation (CBPWM) voltage balance control algorithm. The proposed algorithm presents an excellent inherent voltage balancing even during load changes. For evaluating the proposed control strategy, a 7-level single-phase Modified Diode Clamped Inverter is developed and the method is successfully implemented. The results illustrate that the proposed method can successfully keep the capacitor voltage balanced during load changes. Analytical expressions together with experimental results completely describe the approach. The main novelty of the technique is the introduction of a voltage balancing algorithm for multilevel inverters that eliminates the need for a feedback loop and numerous and costly sensors. This method is suitable for applications which only deal with active power such as solar inverters which inject pure active power into the grid.

Effective thermal management is critical for lithium-ion battery safety and longevity in electric vehicles, particularly under high-discharge conditions. This study investigates a novel hybrid battery thermal management system (BTMS) integrating phase change material (PCM) RT35, liquid cooling, and aluminum fins, assessing the enhancement effects of adding cell-to-cell and cell-to-channel fins alongside pack orientation (θ = 0°, 30°, 60°, 90°). Using computational fluid dynamics (CFD), this study models a cylindrical cell array (18650-type, 3 C/5 C discharge) embedded in PCM, flanked by water-cooling channels. The model varies inclination angles and coolant velocities (0–0.5 m/s) to evaluate buoyancy-driven convection effects on maximum temperature (Tmax) and liquid fraction. Results show that, in the no-fin case, θ = 30° reduces Tmax by ∼6.7 K at 5 C compared to horizontal/vertical setups, driven by enhanced PCM recirculation. Moderate coolant velocities (0.05–0.2 m/s) optimize latent heat utilization, while higher velocities increase parasitic pumping power (∼100 ×) without thermal gains. Cell-to-channel fins reduce Tmax by 9.4 K at θ= 30°, outperforming cell-to-cell fins (as with the maximum 0.4 K reduction at θ=90°), enhancing conduction and convection in a hybrid PCM-liquid BTMS. Unlike prior studies isolating PCM or fixing orientation, this work couples fin enhancements, inclination, and flow, providing design maps for energy-efficient BTMS, guiding next-generation EV battery designs.

• A novel hybrid battery cooling system combines PCM, water, ternary nanofluids, and graded porous media to enhance heat dissipation. • Enhancing thermal conductivity with ternary nanoparticles effectively reduces battery temperature during high-rate discharges. • The proposed system lowers peak battery temperature by 44.47 K at a 5C discharge, enhancing safety and thermal performance. • The optimized design, using graded porous media, achieves excellent temperature uniformity with a maximum cell-to-cell difference of only 0.54 K. This study presents a comprehensive numerical investigation of an advanced hybrid battery thermal management system (BTMS) for lithium-ion batteries. The proposed system integrates passive phase change material (PCM) cooling and active water-based convection, further enhanced by dual ternary nanofluids and porous media. Five configurations are evaluated: (i) passive cooling using pure PCM, (ii) hybrid cooling with PCM and water, (iii) nano-enhanced hybrid using MgO-ZnO-MWCNT in the PCM and Al₂O₃-Cu-Graphene in the coolant, (iv) nano-enhanced hybrid with uniform porous media (ε = 0.85–0.98), and (v) nano-enhanced hybrid with spatially graded porous structures (ε = 0.98 → 0.85 in the x- or y-direction). Simulations are conducted under discharge rates of 1C, 3C, and 5C, with Reynolds numbers ranging from 50 to 300. Results show that the nano-enhanced hybrid reduces the maximum temperature by up to 0.95 K compared to the standard hybrid case, and by more than 10 K compared to the pure PCM configuration. At a 3C discharge rate, increasing the Reynolds number to 300 further reduces the peak temperature by 1.61 K. Uniform porous media improve temperature uniformity, while the graded porous structure offers the best overall thermal performance. At 5C discharge, the optimized system—with nano-enhanced PCM and coolant embedded in graded porous media (ε = 0.98 → 0.85, x-direction)—achieves a maximum cell temperature of 311.75 K and a temperature difference of just 0.54 K, significantly outperforming the passive PCM case (356.22 K and 2.09 K, respectively; ΔT = 44.47 K). The findings demonstrate the high potential of combined nanofluid and porous media strategies for effective and uniform thermal regulation in high-performance EV battery systems.

The potential of synergistic drug combinations in cancer research has been acknowledged for their ability to enhance treatment outcomes. Instead of depending on costly experimental methods, computational techniques have emerged to forecast such combinations. Recent advancements in computational methods have utilized vast datasets comprising chemical, genomic, and pharmacological information. These methods aim to predict drug interactions by analyzing complex biological data and identifying patterns that are indicative of synergistic effects. Here, SynergyImage employs an unsupervised pre-learning technique, ImageMol, to extract features from the chemical structure images of the drugs. Concurrently, gene expression data is transformed into image formats using the DeepInsight method, enabling the extraction of image-based features corresponding to the cancer cell lines through a convolutional neural network. Following a dimensionality reduction step, the framework utilizes a multi-layer perceptron network to predict the synergy scores for various drug combinations. Experimental results show that SynergyImage surpasses leading methods on the O'Neil benchmark datasets, highlighting its capability to predict synergistic drug combinations. In particular, SynergyImage recorded an MSE of 73.402 ± 1.185 and a PCC of 0.83 ± 0.003. These numerical results further highlight the superior performance of SynergyImage compared to other models.