Authors - Md Amirul Islam, Giovanni Stea Abstract - Recent advancements in industry underscore the growing demand for systems that provide both high computational performance and real-time assurances, despite these objectives traditionally being seen as conflicting. To support the complex task of designing such Network-on-Chip (NoC) systems with both performance and Quality of Service (QoS) requirements, frameworks such as ARM MPAM envisage systems with hardware support for resource partitioning and the observation of its effects. It enables new application areas for traditional QoS techniques. Network Calculus (NC) uses transformation operations to model traffic profiles through network nodes. It enables the estimation of the minimum service guaranteed to a flow as it moves through flow-controlled nodes. This paper proposes using Network Calculus to derive insights from traces of traffic samples and to evaluate service curves for QoS validation and worst-case delay analysis in NoC architectures. The derived worst-case performance bound is compared with existing works based on Queueing Theory (QT) and Network Calculus (NC). This comparison demonstrates a significant improvement in the accuracy of the delay bounds over the existing QT and NC approaches.
Authors - Khalid Jaber Almalki Abstract - Blockchain technology offers a new solution to address IoT data accuracy, security, and speed issues in healthcare. This paper details a secure way to communicate, verify, and evaluate healthcare data from IoT devices. The technology protects medical data with strong encryption and agreement. The suggested solution outperforms bitcoin and conventional systems in data security, integrity, scalability, latency, speed, and interoperability. User identification, energy efficiency, network stability, and regulatory compliance top the framework's goals. It's ideal for healthcare applications. The strategy uses statistical analysis and machine learning to make verified data meaningful. This improves healthcare choices. The system improves patient care and operational efficiency, making it a major advance in healthcare data management. This study highlights how blockchain-enhanced IoT devices will transform healthcare data processing, improving patient outcomes and business practices.
Authors - Altaf Raja, AKhil Pandey, Vishal Shrivastava, Mohit Mishra, Sangeeta Sharma Abstract - Grocery inventory control is an essential thing of every day existence that often leads to inefficiencies such as over-buying, forgetting critical objects, or meals wastage. traditional methods rely on human reminiscence and guide tracking, which might be at risk of mistakes. This studies paper introduces mind field, a clever jar ready with AI and IoT technology, designed to automate grocery monitoring. mind field is available in two versions: kind-1, which specializes in weight-based totally tracking and notifications, and type-2, which includes superior sensors to screen freshness, temperature, and humidity. The machine integrates with a cellular software to offer real-time updates, personalized indicators, and buying reminders. by leveraging AI-pushed facts analysis and cloud storage, mind field objectives to enhance grocery control performance for households and small agencies. This paper discusses the trouble, proposed solution, machine structure, advantages, and future scope of the brain container machine. The brain container device leverages cutting edge AI and IoT technology to provide an sensible and automatic answer for grocery management. not like conventional stock monitoring methods, it no longer only detects low stock tiers however additionally analyzes person intake patterns to optimize grocery purchases. The seamless integration with a cellular software guarantees actual-time tracking and proactive indicators, making it a noticeably handy and efficient device. This research explores the gadget's architecture, functionalities, and its impact on lowering meals wastage, optimizing family and enterprise inventory, and promoting sustainable intake practices.
Authors - Aishwarya Reehl Abstract - Predicting Crime is an integral part of keeping the community safe and harmonious. It provides valuable information to the respective authorities to anticipate concerns, prevent victims from being potential targets, and allocate their resources in the best possible way. This paper determines the use of a Machine learning algorithm to predict Crime in Montgomery County. We propose a new model designed to enhance the accuracy of crime data. We cover how descriptive models help understand and demonstrate the next potential move for various crimes. This research also shows how we can pre-process information as needed for prediction algorithms.
Authors - Mikhail Ola Adisa, Sonny Rosenthal, Ifeoma Adaji, Shola Oyedeji, Jari Porras Abstract - Civic campaigns and technology interventions are increasingly recognized as powerful drivers of sustainable waste management behaviors. This study investigates how civic campaigns and persuasive technologies interact to promote sustainable waste practices in Finland. Drawing on a mixed-methods approach, the study combines survey data from 255 residents with interviews from civic organizations to explore the role of engagement levels, campaign effectiveness, and ICT interventions in shaping pro-environmental behavior. Findings reveal that while residents generally exhibit strong recycling habits, sustained participation depends on continued civic outreach, awareness of environmental concerns, and value-based motivation. Digital persuasive tools, such as mobile/web apps, social media, and innovative technologies, were found to be effective for raising waste awareness and supporting sustainable efforts but do not independently drive behavioral change. The study highlights the importance of integrating civic strategies with persuasive technologies to bridge the intention-action gap, scale engagement, and reinforce sustainability norms. The paper contributes to the literature on ICT for sustainability by demonstrating how tailored civic-tech can complement grassroots efforts to foster long-term sustainable behavior change.
Authors - Md.Raza Sheikh, Md.Farid Hossain, Tapu Rayhan, Ehashan Ahmed, Md Zahidul Islam Abstract - This study explores the use of machine learning models, particularly Deep Neural Networks (DNN), for crop prediction in Bangladesh’s diverse agricultural context. A comparison of five models—Gaussian Naive Bayes, Logistic Regression, Decision Trees, Random Forests, and DNN—was conducted using a comprehensive agricultural dataset. The results revealed that while all models had strengths, the DNN outperformed the others, achieving an accuracy of 97.98% in training and 97.95% in validation, with near-perfect precision, recall, and F1 scores. The DNN’s performance, despite its interpretability challenges, underscores its potential in accurately predicting crops from complex, high-dimensional data, crucial for Bangladesh’s varied cropping patterns. This research emphasizes the need for robust agricultural data and suggests that DNNs could significantly improve crop planning, management, and food security, contributing to a sustainable future for precision agriculture in Bangladesh.
Authors - Saravana Kumari Shanmuga Sundaram, Shyam A V Abstract - e-Sanjeevani is India's national telemedicine system. It had its modest origin during the COVID lockdown and is now the biggest recorded platform worldwide for primary healthcare, especially for underprivileged communities. For the design, implementation, and assessment, this article investigates the two service models of the platform: e-Sanjeevani OPD (patient-to-doctor) and e-Sanjeevani HWC (doctor-to-doctor). This study assesses e-Sanjeevani's scalability, usage patterns, and integration with the more extensive health system, such as the Ayushman Bharat Health Accounts (ABHA), and National Digital Health Mission (NDHM) frameworks, based on research literature, secondary data from official sources, and policy documents. For those who do not have access to primary healthcare, the platform has provided several advantages. However, there have been difficulties reaching the system's maximum efficiency. By early 2025, the platform had helped over 342 million individuals throughout India, supporting several aspects of healthcare delivery like diabetic foot, caries in elderly persons, and so on. The research suggests a conceptual framework for incorporating Large Language Models (LLMs) into e-Sanjeevani, addressing the existing challenges and extending the solution’s possibilities. The framework includes LLM-driven features, including clinical decision support, real-time translation, automated documentation, and individualized patient education. It comprises a layered architecture effortlessly incorporated into telehealth, supporting artificial intelligence augmentation at pre-consultation, consultation, and post-consultation phases. This integration can significantly increase provider efficiency, lower workload, and raise the general quality of treatment. The results underline present achievements and the transforming opportunities of LLM-enabled fair telemedicine for India and other low- and middle-income nations.
Authors - Fredy Gavilanes-Sagnay, Edison Loza-Aguirre, Henry N. Roa, Narcisa de Jesus Salazar Alvarez Abstract - This study investigates the performance of various channel estimation and signal detection techniques, including Kalman Filtering, Convolutional Neural Net-works (CNNs), and Recurrent Neural Networks (RNNs); with a focus on their application in 5G/6G networks. We evaluate these methods based on key metrics, including Bit Error Rate (BER), Mean Squared Error (MSE), and computational complexity, under different Signal-to-Noise Ratio conditions. Our results demonstrate that Deep Learning models (CNNs and RNN) significantly outperform traditional methods in terms of accuracy, achieving lower BER and MSE values. However, these improvements come at the cost of increased computational complexity, making them less feasible for real-time applications in resource-constrained environments. Reinforcement Learning models also show promise, offering real-time adaptability for dynamic spectrum management and beam tracking but they also face challenges regarding computational efficiency. Despite some limitations, Kalman Filtering remains valuable for applications where low latency and computational efficiency are critical. Our findings highlight the importance of optimizing these models to balance accuracy and computational load for large-scale 5G/6G networks.
Authors - Diego Pucuji, Angel Cornejo, Paul Velasteguí Abstract - The intensive use of fossil fuels has led to increasing environmental degradation, driving the search for sustainable solutions such as photovoltaic systems applied to electric mobility. In this context, electric motorcycles stand out for their energy efficiency, low maintenance, and ease of recharging. The parish of Calderón, in Quito, provides an ideal setting to evaluate these technologies due to its varied topography and climatic variability, which directly impact the efficiency of solar systems. This study, conducted at the Instituto Superior Tecnológico Vida Nueva, analyzed the performance of an off-grid photovoltaic system used to charge electric motorcycles under real operating conditions between March and April 2025. The installed system consisted of four 550W solar panels, a 2,560Wh lithium-ion battery, and a 3,000W inverter. A controlled load was applied using heat guns, and key energy parameters, including autonomy, consumption, and efficiency, were monitored. Despite the high cloud cover and rainfall during the study period, the system achieved an average daily generation of 5.46 kWh. The tested motorcycles achieved ranges of 22 km and 25 km with energy consumption of 1111 Wh and 1259 Wh, respectively. The results demonstrate the feasibility of using solar energy in urban areas for light electric mobility.
