Excel Engineering College

Mobile ad hoc networks (MANETs) are wireless networks ideally designed for applications such as specific outdoor events, communication in areas without wireless infrastructure, crisis situations, natural disasters, and military operations, as they do not need preexisting network infrastructure and can be rapidly deployed. Security is a major problem in ad hoc networks, and extensive research has concentrated on lowering energy consumption in order to extend node a long lifespan. To overcome these challenges, this work proposes an energy-efficient and secure routing framework ASAO-DST for MANETs, addressing the dual challenge of prolonging network lifetime while ensuring trustworthy route selection. The proposed method uses Improved Fuzzy C-Means (IFCM) clustering, which efficiently organizes network nodes into clusters by enhancing membership functions for optimal formation, while Cluster Head (CH) selection is optimized through Electric Eel Foraging Optimization (EEFO), which emulates the foraging behavior of electric eels to determine the most energy-efficient CH. CH selection in EEFO utilizes CH selection because its multi-phase foraging model simultaneously satisfies the criteria of residual energy, mobility, and connectivity, which cannot be simultaneously optimized with a single-objective energy-aware heuristic, thus prolonging the network lifetime compared to conventional clustering techniques. Trust evaluation is performed using Dempster–Shafer Theory (DST) to integrate several evidence sources for evaluating node dependability. The routing path selection utilizes the Adaptive Snow Ablation Optimizer (ASAO), which replicates snow melting processes to determine optimum and secure paths while preserving Quality of Service (QoS) under fluctuating circumstances. In MANETs, Improved Fuzzy C-Means (FCM) clustering organizes network nodes into efficient clusters by refining membership functions for better formation, while CH selection is optimized using EEFO, which mimics the foraging behavior of electric eels to identify the most energy-efficient CH. The trust evaluation is conducted using DST to combine multiple evidence sources for assessing node reliability. For routing path selection, the ASAO is employed, which simulates snow melting patterns to identify optimal and secure routes, maintaining QoS under dynamic conditions. The suggested approach empirically compares the suggested approach to existing methods based on energy consumption 30 mJ, throughput 0.96 Mbps, end-to-end delay 2.03063 s, network lifetime 6100 rounds indicating LND criterion for the 100-nodes, packet delivery rate 99.8

Salt Gradient Solar Ponds (SGSPs) offer a promising solution for sustainable thermal energy storage. This study introduces a novel trapezoidal SGSP enhanced with coal cinder as a porous medium and a double-glass cover to reduce convective heat loss and improve thermal retention. The research bridges a gap in existing literature by experimentally and numerically evaluating the impact of porous media on SGSP performance. Two configurations-a conventional pond (SGTSP-C) and a coal-cinder-enhanced version (SGTSP-CC)-were analyzed. The enhanced pond achieved a 26.63 % higher LCZ temperature, reaching 62.45 degrees C, compared to 49.32 degrees C in the baseline setup. Energy and exergy efficiencies improved notably from 9.2 % to 15.4 % and from 0.5 % to 0.94 %, respectively. Validated numerical simulations further confirmed that coal cinder enhances thermal stratification and stability by reducing convective mixing. The findings demonstrate the potential of porous media integration for improving SGSP efficiency and advancing solar thermal energy storage systems.

Aluminum alloy AA5083 is used in aerospace and marine applications due to its better properties like better corrosion resistance, weld quality, and formability. However, its other qualities like poor strength and low wear resistance limit the service. Therefore, the aim of this study is to enhance the wear resistance of AA5083 alloy through the incorporation of reinforcements. AA5083 aluminum alloy composites were produced in this investigation using the stir casting technique, with the addition of TiO2 and ZrO2 nanoparticles. Totally, three composites were made, namely AA5083/TiO2, AA5083/ZrO2, and AA5083/(TiO2-ZrO2), and performance was assessed. The findings indicated that the AA5083/ZrO2 composite exhibited superior hardness at 93 HV and achieved a greatest strength of 318 MPa during tensile assessment. Further, a least wear rate of 0.00398 mm3/m, lower mass loss of 0.0303 g, and the lowest COF value of 0.41 were noted for AA5083/ZrO2 composite. This was attributed to the inclusion of ZrO2 nanoparticles with high hardness than TiO2 particles. This effort is unique in metal matrix compound area as metal oxide nanoparticles have been used as reinforcements. Previous research reports on AA5083 alloy revealed that the majority of investigators concentrated on carbide particles. This work exhibited that the wear rate of AA5083 aluminum alloy was decreased by approximately 30

This research for photo augmentation and classification, four deep learning models VGG InceptionNetV2, InceptionV3, MobileNetV2, and VGG16—are compared. For models, Accuracy, Precision, and F1 Score were used to measure how well they worked. PSNR, Entropy, SSIM, and MSE were some of the standards for picture quality. InceptionV3 made images look much better by lowering the MSE to 905.35, raising the PSNR to 18.56 dB, raising the SSIM to 0.625, and raising the Entropy to 6.40 %. These tests show that the contrast and characteristics of the image were kept. The fact that VGG16 has a higher accuracy (0.925) and F1 Score (0.916) shows that it is better at finding and classifying important features. VGInceptNetV2 greatly reduced false positives, with a maximum precision of 0.922. MobileNetV2 did not do well at augmentation or categorization. This comparison demonstrates that InceptionV3 is better for applications that need high-quality images and VGG16 is better for jobs that need high-accuracy classification. This shows the tradeoff between image quality and classification accuracy. This project enhances AI-driven imaging systems and develops more intelligent visual monitoring and analytical tools to enhance the safety and sustainability of cities and communities, in accordance with Sustainable Development Goals 9 (Industry, Innovation, and Infrastructure) and 11 (Sustainable Cities and Communities).

Hand Gesture Recognition (HGR) plays a crucial role in machine-human interaction for effective user interactive experience in gaming field, Virtual Reality (VR) and Augmented Reality (AR). The conventional HGR model utilizes various deep learning techniques but low quality image, occlusion and extraction of Spatio-temporal features pose a major challenge. In this research work, a novel dual stream Bidimensional Empirical Mode Decomposition (BEMD)- Amodal Depth estimation (ADE) based 2D and 3D- Convolutional Neural Network (CNN) backbone for HGR has been proposed. The BEMD improves the quality of image by removing background and noise using decomposition by generating multiple image components. The ADE masks the occluded portions of the image and try to predict the features present behind the image and gives occlusion aware features. The dual stream architecture has been employed to handle static and dynamic inputs separately by suing two separate pooling layers: Strip pooling for spatial features and Temporal Attentive Pooling for temporal features. The proposed BEMD-ADE based 2D and 3D CNN has been evaluated and achieved 90.44% in Hagrid dataset and 92.85% in NVGesture dataset which is better when compared to existing model.