Koneru Lakshmaiah Education Foundation

In the design of Wireless Sensor Networks (WSNs), efficient energy consumption and prolonged network lifespan are essential objectives. Cluster-based routing protocols have proven particularly effective in addressing these goals among several methods. Clustering and routing are consistently among the most advanced approaches for extending the network lifetime in WSNs. A significant challenge is selection of an appropriate node to act as the Cluster Head (CH), as this decision directly affects the energy efficiency and network durability. Numerous routing and clustering approaches have been developed which utilizes both single-hop and multi-hop communication structures to optimize energy consumption. Consequently, the design of an efficient and robust routing protocol remains a vital area of research to overcome these constraints and ensure sustainable network performance. Considering the existing limitations, a novel (Rhinopithecus Swarm Optimization based Bobcat Optimization algorithm) RSOBO algorithm is proposed for CH selection and routing in WSNs. The approach is organized into two stages: CH selection and routing. Firstly, CHs are selected through RSOBO algorithm, which considers constraints such as distance, delay, energy, and security. Secondly, the RSOBO algorithm is applied again for trust-aware cluster-based routing which incorporates constraints like trust, link quality, and link lifetime. Comprehensive testing has validated the RSOBO algorithm which demonstrates its high accuracy and higher performance in cluster-based routing. Consequently, the RSOBO algorithm outperforms conventional methods in key measures consisting distance, delay, energy efficiency, security, trustworthiness, link quality, and link lifetime. The RSOBO strategy reached the delay rate of 1.978 × 1e-7s, distance of 39.873m, energy of 0.247J, Link lifetime of 1965 and Link Quality (PDR) of 0.983, respectively. These enhancements provide reliable, energy-efficient, and secure communication in WSNs. Therefore, the proposed RSOBO algorithm is suitable for large-scale IoT applications, environmental monitoring, precision agriculture, industrial monitoring, and disaster management. It is particularly effective for mission-critical WSN applications where prolonged network lifetime and reliable data transmission are essential.

Optical biosensors require high sensitivity, stability and measurement accuracy for trace-level biomolecule detection. Tacrolimus (TAC) is a first-line immunosuppressant with a narrow therapeutic window (5–20 ng/mL), and its blood concentration varies greatly among patients, making real-time and accurate monitoring essential for clinical medication safety after organ transplantation. In this work, a fiber-optic biosensing system integrated with a laser ring cavity is proposed for sensitive, rapid and accurate detection of TAC. A nested dual-tapered multimode fiber-multi-core fiber-multimode fiber (MMF-MCF-MMF) probe was fabricated to strengthen the evanescent field, boost light-matter interaction and improve sensor sensitivity. The nested dual-taper structure was analyzed using RSoft software, and the optimized probe was verified by experimental characterization. An erbium-doped fiber ring laser with a full width at half-maximum (FWHM) of 0.116 nm acts as the narrow-linewidth near-infrared excitation source. ZnCdS quantum dots are immobilized on the probe surface to improve sensing responses, while Apt122 aptamers functionalized on the fiber enable specific molecular recognition of small-molecule TAC. The developed biosensor achieved a sensitivity of 0.447 nm/lg(ng/mL) and a limit of detection (LoD) of 193.8 pg/mL, about 15.85-fold lower than that of the conventional amplified-spontaneous-emission-based sensing system. Selectivity, stability and pH-response tests further confirmed the reliable quantitative detection performance of the sensor. This sensor provides reliable technical support for real-time and precise monitoring of blood TAC concentrations after organ transplantation.

Multi-camera frameworks detect and track aerial vehicles in smart cities, integrating multiple viewpoints to enhance coverage and accuracy. However, challenges such as high computational demands, large-scale data handling, synchronization, and ensuring consistent privacy-preserving operation limit their practical deployment. In order to solve UAV-based aerial vehicle detection problems, this paper has introduced an approach that is called Dual-Domain Unfolding Decker Convolutional Neural Network with Hiking Optimization Algorithm (DDUDCNN-HOA) based on deep learning technology. The system employs images captured by the UAV-Vehicle-Detection-Dataset, which consists of different classes of vehicles and varying environmental situations. The encryption process is done using Orthogonal Arrays Supporting Thumbnail Decryption (OASTD), and this ensures that the aerial images are encrypted securely. The pre-processing technique used is adaptive learning-based bilateral filtering (ALBF), which will be used to eliminate noise and improve edges. The relevant spatiotemporal information is then captured by Short-Time Quaternion Quadratic Phase Fourier Transform (STQQPFT). The detection process is performed with the aid of DDUDCNN that incorporates both Double-Decker Convolutional Neural Network (DDCNN) and Dual-Domain Deep Unfolding Transformer (DDDUT) to analyze multi-level spatial and temporal dependencies. Finally, the Hiking Optimization Algorithm (HOA) refines the model parameters to ensure quick convergence, accurate detection, and high generalization on multiple camera streams. The experimental validation of the proposed DDUDCNN-HOA system shows its effectiveness with an accuracy of 99.70%, a precision of 99.10%, a recall of 97.90%, and an F1 score of 98.50% with only 1.2% false positive rate, thus ensuring an effective UAV-based vehicle detection and tracking mechanism.

Geopolymers have now made their way through as a low-carbon substitute for Portland cement with the increasing demand for construction materials that are both sustainable and thermally resilient. This research revolves around the fly ash–calcium carbide residue (FA–CCR) geopolymer composites and the attempts that are made to optimize their properties and performance comprehensively, i.e. mechanical, thermal, environmental, and economic ones, by applying an experimental and decision-making integrated framework. In total, ten different geopolymer mixes were formulated through CCR content from 0 to 20

Accurate classification of glioma grades from magnetic resonance imaging (MRI) is essential for clinical decision-making in neuro-oncology. Although deep learning performance has been impressive with classical models, they struggle with high-dimensional medical imaging data and generalise poorly beyond their training data, especially in time- and resource-constrained settings. In light of the aforementioned challenges, we propose QuantumMedDx, a hybrid quantum–classical learning framework for classifying gliomas using MRI. The framework combines quantum feature encoding and variational quantum circuits with classical neural inference to improve diagnostic performance. The base model, QImageNet, uses amplitude-based quantum encoding for writing, entanglement-enabled parameterised quantum circuits (EPQCs) as feature extractors, and classical dense layers for classifying HGG and LGG from multimodal MRI slices. We demonstrate the effectiveness of the proposed approach on the BraTS 2021 benchmark dataset using a patient-aware 5-fold cross-validation protocol. Experimental results show that QuantumMedDx achieves accuracies of 94.12