Sir Syed University of Engineering and Technology

In this investigation, an artificial intelligence-based neural network is used to estimate solutions for the flow and heat transfer characteristics of Ag-CuO/Water hybrid nanofluid past a rotating stretching sheet. Heat source/sink, convective movement, nonlinear thermal radiation, and chemical reactions are all taken into consideration in the energy and concentration equations. The modelled equations are transformed into ordinary differential equations by implementing a suitable similarity transformation. The Levenberg–Marquardt method is used by an artificial neural network (ANN) for solving equations computationally. The outcomes of training, testing, and validation are examined using regression plots, error histograms, performance charts, and transition state analysis to assess the accuracy of the suggested methodology. An enhancement in the Magnetic (M), Rotation ( ε ), and porosity (K) parameters leads to a rise in temperature and concentration profiles; however, a converse effect has been observed with an increment in the Stretching (λ) and ratio (γ) parameters. Growing the Space-dependent heat source/sink ( A^*) and nonlinear thermal radiation parameter (Rd) appreciates the velocity profiles; nonetheless, it decreases for the stretching ratio parameter (λ). To illustrate the precision of the numerical method used, a comparison table is shown, showing an impressive degree of agreement with earlier reported results. The findings show that the hybrid nanofluid offers the possibility of increased thermal performance in application in engineering applications by dramatically improving the heat transmission properties under the applicable circumstances. The quantitative findings of our research work are that velocity, temperature, and concentration fields exhibit strong parametric sensitivity: λ and γ generally suppress flow, heat, and mass transport, while M, K, Rd, and related parameters enhance boundary-layer thickness, temperature, and concentration but may reduce transfer rates due to Lorentz and radiative effects. Skin friction increases with M, Rc, λ, and related parameters, whereas heat and mass transfer rates decline at higher M and Rd owing to thermal boundary-layer thickening and magnetic damping. The AINN trained via the Levenberg–Marquardt algorithm demonstrated near-perfect accuracy (MSE ≈ 0, R = 1), minimal error dispersion, and robust convergence, validating its effectiveness in modeling complex HNF transport phenomena. The results demonstrate that heat transfer is improved for a strong non-linear thermal radiation parameter by 30

Underwater Acoustic Sensor Networks (UASNs) play a critical role in underwater exploration, yet face challenges like high energy consumption, and uneven load distribution in multi-hop routing. While existing clustering protocols like Anchor Nodes assisted Cluster-based Routing Protocol (ANCRP) have improved energy-efficiency through uniform cluster formation, they still suffer from unequal energy depletion among cluster heads (CHs), particularly in shallower layers. This paper introduces key innovations that advance cluster-based routing in UASNs. We propose an Energy-Adaptive Non-uniform Clustering Protocol (EANCP) with three novel aspects: 1) depth-optimized cluster sizing where deeper CHs manage larger clusters with lower communication demands while shallower CHs handle smaller clusters to prevent early energy depletion; 2) an adaptive beaconing mechanism where transmission rate varies with depth to conserve energy in stable deep regions; and 3) a multi-criteria CH selection strategy considering residual energy, congestion, and distance to sink. The protocol was designed and simulated in Python, leveraging its scientific computing ecosystem (NumPy, SciPy, and Matplotlib) for flexible modeling of underwater acoustic channels and energy consumption. Unlike ANCRP’s uniform approach, EANCP’s depth-aware design addresses the uneven energy consumption in UASNs. Simulations demonstrate significant improvements: a 14% lower energy consumption, 11.5% longer network lifetime, $5-8\% $ higher packet delivery ratio (PDR) and 20% lower delays compared to ANCRP. The protocol particularly excels in large-scale deployment by preventing routing bottlenecks through congestion-aware forwarding. These advancements make EANCP a significant improvement in underwater routing protocols, offering a more robust solution to balancing energy efficiency with reliable communication in harsh underwater environments.

Traffic Sign Recognition (TSR) models based on Deep Learning are highly vulnerable to adversarial perturbations where often imperceptible changes to the input can significantly mislead model predictions and posing serious safety concerns in autonomous driving. This paper presents PGD-PPM, a hybrid defense framework designed to enhance adversarial robustness of TSR models by combining Projected Gradient Descent-based Adversarial Training (PGD-AT) with Pyramid Pooling Module (PPM) integration. The proposed architecture improves multi-scale contextual feature aggregation to resist adversarial perturbations while maintaining high clean accuracy. Four Convolutional Neural Network (CNN) architectures, VGG16, VGG19, ResNet50 and EfficientNetB0 were evaluated on two benchmark datasets GTSRB) and BelgiumTSC datasets under gradient based white-box attacks FGSM, IFGSM and PGD at various perturbation strengths ( $\epsilon =0.1-2.0$ ). The proposed models exhibit significant improvement in clean and adversarial accuracies. For instance, EfficientNetB0 with PGD-PPM achieves the most significant improvement in clean accuracy of 92.57% (up from 87.55%), VGG16 and VGG19 also increase by + 4.29% and + 3.10%, respectively, whereas EfficientNetB0 maintaining robustness with improved adversarial accuracy of 89% and 87% under strong adversarial condition (PGD attack at, is an element of = 0.1, 0.2). However, ResNet50 achieves the highest adversarial accuracy of 90% and 88% under PGD attack at is an element of = 0.1 and 0.2, which is significantly higher than the corresponding baseline model. The experimental results indicates that proposed framework not only possess a strong improvement in adversarial accuracy but it also improves the clean accuracy which mitigates the accuracy-robustness trade-off, contributing to safer and more reliable intelligent systems.

Modern communication networks require efficient spectrum utilization to satisfy the low-latency and high-reliability requirements of next-generation wireless systems. However, the rapid growth of connected devices has intensified traffic congestion, increasing the demand for efficient spectrum sensing in Cognitive Radio Networks (CRNs). The conventional Energy Detection Method (EDM) relies on fixed decision thresholds, limiting its effectiveness under dynamic wireless conditions. This study investigates Swarm Intelligence (SI)-based threshold optimization for spectrum sensing by comparatively evaluating the Artificial Bee Colony (ABC) and Firefly Algorithm (FFA) in a multi-fusion center environment. Three experimental configurations with different swarm population sizes were investigated to evaluate the robustness and consistency of both optimization techniques. The optimized thresholds obtained using ABC and FFA were compared with the conventional EDM in terms of probability of detection (Pd), probability of false alarm (Pf), and average optimization time. The experimental results demonstrate that both swarm intelligence algorithms improve spectrum sensing performance over the conventional EDM by enabling adaptive threshold optimization. Among the investigated approaches, the ABC algorithm achieved a more balanced trade-off between probability of detection and false alarm, whereas the FFA exhibited stronger detection capability at the expense of a slightly higher false alarm rate while requiring a lower average optimization time than the ABC algorithm. These findings demonstrate that swarm intelligence provides an effective optimization framework for adaptive spectrum sensing and supports more reliable dynamic spectrum access in CRNs.

Parallel power system restoration (PPSR) is a quick and efficient way to restore power after a blackout. The main goals of the PPSR are to divide the blackout system into smaller sections, identify connections between these sections, reduce the time needed for restoration, and supply as much load as possible. This paper presents a sectionalizing-based decision-making strategy for PPSR using label propagation algorithm (LPA) and cooperative game theory. Firstly, this paper introduces an improved LPA based on complex network theory to detect groups within the power system for PPSR. The proposed LPA is designed based on the influence of bus labels and the node influence index to address the sectionalizing problem. This provides a clear ranking of nodes across the network, helping to prevent label switching issues seen in traditional LPAs. Secondly, a game theory-based strategy is presented which provides the cooperation between buses and subsystems by evaluating the Shapley value of buses. The method makes recommendations for cooperated subsystems by calculating the real-time Shapley value of the buses. This proposes the dynamical interaction between buses and subsystems with the purpose of improving system restoration stability and efficiency based on a novel cooperative game strategy. Finally, an optimization model for determining the optimal sectionalizing schemes is developed to minimize the cut-sets (i.e., tie-lines between the subsystems) and to minimize the real power exchange (i.e., power flow) between subsystems. The proposed optimization model considers the system topology as well as the operating characteristics (i.e., power flow) before the pre-blackout system. Furthermore, the sectionalizing constraints is applied in the proposed strategy for feasibility verification. The proposed LPA and cooperative game theory-based sectionalizing strategy (SS) is suitable for quickly finding a community division (i.e., optimal sectionalizing scheme) solution after a power system blackout. Finally, case studies on the modified IEEE 9-bus, 39-bus, and 118-bus power systems are performed. The simulation results verify that the proposed strategy can successfully restore the power system and determine the efficient sectionalizing scheme.