Taxila, a historic seat of learning and an important archaeological site, is about 30 km north-west of Islamabad and Rawalpindi.HITEC University commenced classes in November 2007 with an intake of 250 students, in affiliation with University of Engineering and Technology, Taxila. The university was granted its own charter in November 2009 by the government of Punjab. The university is sponsored by the Heavy Industries Taxila Education Welfare Trust.
The rapid proliferation of the Internet of Things (IoT) has unlocked unprecedented opportunities across consumer, industrial, medical, and environmental sectors. However, this growth has also led to a significant increase in electronic waste (e-waste), raising concerns about resource depletion and long-term ecological impact. In response, biodegradable and transient antennas (BTAs) have emerged as promising solutions for building sustainable, eco-friendly wireless systems. These antennas are engineered to degrade naturally or dissolve after fulfilling their operational purpose, thereby minimizing their environmental footprint without compromising functional performance. This paper presents a comprehensive review of the design, materials, fabrication techniques, and performance considerations associated with BTAs for IoT applications. The discussion highlights recent advances in biodegradable conductors and substrates, examines critical challenges such as conductivity, mechanical stability, and system-level integration, and proposes innovative strategies to overcome these limitations. In addition, the review explores a range of application domains, including disposable electronics, implantable medical devices, and environmental monitoring platforms, where degradable antenna systems offer unique advantages. Finally, the potential convergence of artificial intelligence (AI) driven IoT ecosystems with green antenna technologies is discussed, outlining future directions for research and development. This work is intended to serve as a foundational resource for engineers, materials scientists, and researchers committed to advancing sustainable electronics and green communication infrastructures.
Reliable and sustainable energy is fundamental to socio-economic development; however, Pakistan faces persistent energy challenges due to rising demand and heavy reliance on costly and environmentally harmful fossil fuels, particularly in Balochistan. Given Gwadar's abundant solar irradiation and wind resources, this study evaluates the feasibility of achieving a net-zero energy city through three renewable-based power system configurations using solar, wind, and battery storage. Model 1, relying solely on solar photovoltaic (PV) cells and batteries, yields a levelised cost of energy (LCOE) of $0.441/kWh, with a total net present cost (NPC) of $182 million. Model 2, integrating solar, wind turbines, and batteries, reduces the LCOE to $0.320/kWh, with an NPC of $133 million. Model 3 incorporates solar, wind, and a diesel generator, offering the lowest LCOE of $0.219/kWh and an NPC of $90.7 million. While Model 1 is the most environmentally sustainable, it has the highest LCOE and NPC. Model 3 ensures the highest reliability with the lowest LCOE and NPC at the expense of environmental impact due to diesel. The payback rate for Model 1 is 6.2 years; for Model 2, it is 7.79 years; and Model 3 has the lowest payback rate of 4.62 years. This study provides a comprehensive techno-economic analysis, identifying optimal solutions for Gwadar's energy requirements, which can be replicated in similar off-grid areas.
Parkinson’s disease (PD) is a degenerative, chronic neurological condition that impairs a person’s ability to move normally. People may experience difficulties with speaking, writing, walking, or performing basic tasks if dopamine-generating neurons in the brain are injured or die. Using traditional techniques for PD analysis is time-consuming and challenging, as the evaluation process is prone to high misclassification rates. Therefore, we proposed a deep learning-based architecture for classifying PD using WiFi signals in this work. The data is generated at the initial stage using WiFi signals. After that, we proposed two deep learning architectures from scratch. The first architecture, named E3-ST transformer, is based on a three-stage encoding scheme, and the second two residual-attention-block-based networks are named PD-RAN2. Both models are trained on generated WiFi signal data, and the hyperparameters are optimized using Bayesian Optimization (BO). In the next phase, trained models are used, and deep features are incorporated, employing a new method termed serial-based attention-weighted. The fused features are finally classified using neural network classifiers. The output is in label classes such as slow walking, fast walking, sitting on a chair, standing still, and FOG episodes. The Medium Neural Network (MN2) classifier achieved the best accuracy of 97.78
Accurate and efficient identification of jute pests is essential to sustaining agricultural yield and preventing fibre-quality degradation in jute (Corchorus spp.) production. Traditional visual inspection is subjective, labour-intensive, and error-prone under variable field conditions, motivating the need for robust automated classification systems. This paper introduces a lightweight deep learning architecture, MLECNet, that integrates squeeze-and-excitation blocks with ViT and a convolutional attention module in a novel way. The proposed architecture is designed based on the information in the image, extracting features at multiple scales and initially fusing the feature maps using a depth-wise concatenation approach. The feature maps are extracted from the squeeze-and-excitation (SE) blocks, inspired by the EfficientNet architecture. The information is refined by the ViT encoder blocks attached to the initial SE blocks. The final information of these blocks is flattened and passed to the attention maps for the extraction of more refined features that classify the jute pests. Experimental evaluations were performed on a 17-class jute pest dataset comprising 17,000 augmented images. The hybrid network, containing 3.7 million learnable parameters and a model size of only 13.4 MB, achieved a classification accuracy of 92.30%, a precision of 92.40%, a recall of 92.30%, and an F1-score of 92.30% using a medium neural network classifier, outperforming individual backbones and prior state-of-the-art models such as DenseNet201, ResNet50, and VGG19. The proposed model also demonstrated superior generalisation and interpretability through comparative ablation studies, confirming the complementary advantages of convolutional, transformer, and attention-based architectures in a unified form. The results establish the proposed fusion-based framework as a computationally efficient, interpretable, and scalable solution for real-world jute pest detection, contributing toward the advancement of AI-driven precision agriculture and sustainable fibre crop management.
Skin cancer represents a major global health issue, and timely and precise detection is essential for enhancing patient outcomes. In recent years, deep learning models have demonstrated exceptional efficacy in numerous computer vision applications, particularly image categorization. Nevertheless, the diagnosis of skin cancer presents challenges due to class imbalance, artifacts such as hair and imprecise lesion margins, and the non-generalizability of current methodologies. This study establishes a novel framework for segmentation and classification incorporating the Parallel Attention Module (PAM). A customized Convolutional Neural Network (CNN) is established for feature extraction, a PAM-UNet for data segmentation, and an attention CNN with PAM for classification, therefore addressing these shortcomings. The PAM, based on the Transformer Attention Module (TAM) and Global Spatial Attention (GSA), enhances features and eliminates artifacts in both segmentation and classification. The proposed framework outperforms the other current methods on the HAM10000 and ISIC2019 datasets by demonstrating improved precision in skin lesion segmentation and an exceptionally high level of classification accuracy. In comparison to previous approaches, it achieves a 99.9% classification accuracy on HAM10000 and a 99.1% classification accuracy on ISIC2019. Interpretation is enhanced by employing Grad-CAM, which provides visual insights into the model’s decision-making process. The proposed framework demonstrates superior precision in skin lesion segmentation and exceptional accuracy in classification, hence surpassing other current methods on the HAM10000 and ISIC2019 datasets.
