Bare-metal deployments of Kubernetes-based container platforms such as Red Hat OpenShift lack the native, cloud-provider-integrated load-balancing services that are taken for granted in public cloud environments. MetalLB fills this gap by exposing Kubernetes Services of type Load Balancer on bare-metal or on-premises infrastructure, using either Layer 2 (ARP/NDP) or Border Gateway Protocol (BGP) advertisement modes. This paper presents a design and evaluation study of the BGP advertisement mode of MetalLB as deployed on OpenShift, augmented with Bidirectional Forwarding Detection (BFD) for rapid link and peer failure detection. We describe the underlying protocol mechanics of BGP and BFD, the MetalLB controller/speaker architecture, and the integration points with OpenShift's OVN-Kubernetes network stack. A reference architecture is proposed in which each cluster node peers with top-of-rack switches over eBGP and advertises Virtual IP (VIP) service routes, while BFD sessions running beneath the BGP peering’s shrink failure-detection time from tens of seconds to the sub-second range. A testbed methodology and representative performance results are presented, comparing convergence and failover behavior with and without BFD. The study concludes that combining MetalLB, BGP, and BFD provides a practical, vendor-neutral, and highly available load-balancing fabric for on-premises Kubernetes and OpenShift clusters, at the cost of additional operational complexity in router and cluster network configuration.
In database management systems (DBMSs), query optimization is a frequent practice. Its primary goal is to maximize query execution efficiency among many possibilities. In recent years, the world of data-driven applications has grown rapidly, and effective query processing is an essential element of high-performance systems and efficient resource use in cloud computing environments. In this paper, the authors review key query optimization strategies used in state-of-the-art DBMSs. These include heuristic-based optimization, cost-based optimization, semantic-based optimization for both join and indexing strategies and sub-query rewriting methods. The study contextually explores new AI/ML-based optimization techniques that contribute towards enhancing adaptive query execution and automated tuning in the dynamic environment. Apart from this, the paper also discusses the role of query execution plans, cost models and optimization frameworks and how they can improve the performance of databases in enterprise systems, databases in a distributed environment of the cloud and big data analytics platforms. Using a systematic literature review of recent relevant studies, researchers identify current progress and challenges in query optimization alongside possible future research directions. The paper shows that in the context of many complex (possibly correlated) workloads, modern database systems need some degree of intelligent/elastic behaviour to operate with bounded latency and resource consumption.
Real-time video surveillance is an essential component of the modern security system, particularly in the context of monitoring public areas, particularly in densely populated areas such as airports, train stations, and urban junctions. Occlusion, quick movement, and high item density are some of the issues that traditional surveillance systems face when accurately detecting and effectively monitoring objects. YOLO, which stands for "You Only Look Once," is a framework that includes DeepSORT, which stands for "Simple Online and Realtime Tracking with a Deep Association Metric," for effective object tracking. This framework is proposed in this study as a robust multi-object identification and tracking system. By utilizing appearance traits and Kalman filtering, YOLO guarantees that objects are localized and classified quickly and accurately, whereas DeepSORT ensures that identification is maintained consistently between frames. Under various lighting and crowd density situations, the combined model was implemented and tested on video feeds captured from the real world. Given that the system can handle occlusions and re-identify lost targets with high accuracy, low latency, and resilience, as demonstrated by the results of the experiments, it is a feasible option for applications that include intelligent surveillance. When applied in dynamic and complex public contexts, this technique dramatically improves situational awareness and assists proactive security management.
With the increasing number of vehicles on the road, parking management has become a significant challenge in urban areas. The Smart Parking Lot Helper aims to optimize parking space utilization using IoT and AI technologies. This paper explores sensor-based detection, mobile application integration, automated fee deduction systems, and the overall benefits of a smart parking system. We analyse existing research in smart parking solutions, present a feasible implementation plan, and discuss the potential impacts on smart cities.
The enhancement of the energy efficiency of railway transport has turned into an urgent task due to the rising mobility demand and the rising concerns regarding the environment. Even though the railways are already more energy-efficient than most other modes of transport, additional energy consumption and emissions cuts are necessary to meet sustainability goals. The paper will introduce in-depth review of the energy-efficient techniques in the railroad system, including both technological, infrastructural, modelling and operational techniques. Major technologies, including regenerative braking, onboard and infrastructure-based energy storage systems and other alternative propulsion technologies, including fuel cell-battery hybrid systems, are addressed. The paper also discusses advanced modeling and simulation frameworks that are needed to study the dynamics of trains and traction power supply and the energy flows. In addition, operational strategies to decrease energy consumption such as speed profile optimization, coasting control, time-based energy optimization and traffic control in terms of headway control are discussed. This review offers a comprehensive picture of how coherent technological and operational strategies could deliver a great deal of energy efficiency to the railway transportation systems.
The Air Quality Index (AQI) measures how air pollution affects human health. As pollution levels rise in Indian cities, we need reliable prediction models for better environmental management. This paper analyses different machine learning techniques for predicting AQI. We use Support Vector Regression (SVR), Random Forest Regression (RFR), and CatBoost Regression (CR) on data from New Delhi, Bangalore, Kolkata, and Hyderabad. We evaluate model performance using Root Mean Square Error (RMSE) and accuracy. Experimental results show that RFR performs best in most cities, while CR is most effective in New Delhi. To tackle dataset imbalance, we use the Synthetic Minority Oversampling Technique (SMOTE), which improves prediction accuracy for all models. Additionally, we assess other models, including SARIMA, Support Vector Machine (SVM), and Long Short-Term Memory (LSTM) for Ahmedabad. Among these, SVM with a radial basis function (RBF) kernel shows the best results. The findings emphasize how combining data balancing methods with machine learning models can improve AQI prediction. This approach can help with pollution control strategies and better decision-making.
The main objective of this study was to propose suitable equipment for drying in humid tropical regions. Following the design and distribution of two online survey forms via Google Forms one for producers and the other for consumers a total of 547 completed forms were collected. The main findings showed that, in terms of growing areas, all ten regions of Cameroon are suitable for growing this plant, with the Littoral Region leading the way (15.70%), followed by the Central Region (14.40%), the West Region (13.10%), the North-West Region (11.50%) and the Adamaoua Region (10.20%). It also emerged that three main varieties are cultivated: white ginger (58.15%), black ginger (28.43%) and turmeric (11.82%). With regard to preservation methods, indirect solar drying was found to be the most widely used technique (67.40%) and the most recommended (96.50%), ahead of traditional drying. Furthermore, we have designed a PV/T solar dryer for fruit and vegetables. The design was carried out using the functional analysis method, which enabled us to identify the problem and define the functions to be fulfilled. The technological solutions were derived from the functional analysis of requirements and resulted in a hybrid PV/T solar dryer. The proposed dryer consists of two units: a drying chamber with a volume of 0.35 m³, in which air circulation is provided by two (02) continuous fans operating at 1.2 m/s, and a PV/T collector comprising a 150 Wp solar PV panel, coupled with climate data from Ngaoundéré.
Autonomous driving is dependent upon reliable perception of traffic elements including traffic lights and traffic signs in order to ensure both safety and efficiency in making decisions. However, latest computer vision-based approaches treat object detection and semantic state recognition as either separate tasks, or they utilize expensive, hardware-dependent sensor technology to continuously feed in video data that can greatly increase the processing cost and limit deployments in resource-limited environments. This problem has been addressed through the development of Re-FusionNet; a lightweight, attention-enhanced convolutional architecture designed to perform both joint traffic light and traffic sign detection along with semantic state recognition utilizing only RGB image data. The Re-FusionNet utilizes a dual-path fusion encoder (DPFE), differential attention module (DAM), along with a compact convolutional backbone to enable contextual feature representation to be enhanced and to focus on relevant traffic-related objects. In addition to its ability to process images sequentially by only sampling a few frames from a video stream rather than continuously feeding it into the network, it enables efficient capture of contextual information. Finally, all three tasks are performed simultaneously via a single multi-task prediction head that predicts both object location and state, allowing for joint training. Results demonstrate that the Re-FusionNet achieves an average precision (mAP) of 97.3% @ 0.5 IoU on the BDD100K dataset while achieving a state recognition accuracy of 90.0%, and an mAP of 94.2% @ 0.5 IoU on the LISA dataset. Moreover, due to the low computational requirements, it can achieve frame rates above 150 fps. Overall, results clearly show that the Re-FusionNet represents an efficient and deployable method for real-time traffic perception applications within autonomous vehicles and intelligent transportation systems.
Construction sites remain highly prone to accidents, with falls from heights and formwork collapses being main causes of fatalities worldwide. Traditional safety practices emphasize worker training and hazard mitigation; however, the Safety-by-Design (SbD) approach shifts the focus toward eliminating risks at the design stage itself. This study explores the role of SbD in addressing fall hazards and formwork failures through a comprehensive review of existing literature. Evidence indicates that a significant proportion of construction accidents are linked to design-related factors, highlighting the critical responsibility of designers in improving safety outcomes. SbD promotes the integration of safety considerations into planning, material selection, and structural design, including provisions such as guardrails, scaffolding systems, and stable formwork structures. The approach not only reduces accidents and associated costs but also enhances productivity and fosters a proactive safety culture. Despite its potential, challenges remain in implementation, particularly in developing countries, necessitating further research and strategic adoption frameworks.
Salesforce, as a leading cloud-based Customer Relationship Management (CRM) platform, plays a critical role in enabling digital transformation, scalable data management, and intelligent business operations. This paper presents a comprehensive analysis of Salesforce platform architecture, database administration, and data governance mechanisms. It explores the Salesforce ecosystem, including its multi-tenant architecture, data storage strategies, and metadata-driven framework that support secure, flexible, and efficient data environments. The study further examines key database administration aspects such as data modeling, lifecycle management, and performance optimization to enhance scalability and operational efficiency. Additionally, it highlights Salesforce security and governance features, including identity management, access control, encryption, and monitoring, which ensure data protection and regulatory compliance. A detailed literature review emphasizes recent advancements in Salesforce-based systems, particularly in multi-cloud integration, interoperability, and AI-driven governance. The findings indicate that Salesforce significantly improves data security, system performance, and decision-making capabilities. However, challenges such as integration complexity and lack of standardized governance frameworks still require further research and development.
Digital transformation and Artificial Intelligence (AI) have become important sources of innovation and competitive edge in the contemporary business. In a more dynamic and technology-intensive corporate world, companies are implementing the use of more sophisticated digital solutions to increase efficiency, better decision-making, and provide superior customer experiences. This paper discusses how AI-based digital transformation can be applied in large enterprises with respect to its conceptual framework, enabling technologies, and strategic approaches. It brings out major attributes of digital transformation, such as its technology-based, data-driven, customer-oriented, and agile nature, and new practices as digital strategy, organizational change management, and data governance. The paper also discusses the fundamental AI technologies, such as Machine Learning, Natural Language Processing, Robotic Process Automation, and Computer Vision, which facilitate intelligent automation and data-driven innovation. Also, it addresses strategic growth strategies that are embraced by large firms, including entering the new market, product development, market expansion, partnership, and mergers and acquisition. The research also identifies key procedures towards successful implementation such as market research, objective setting, performance checking and flexibility. In general, the article reinforces the idea that the successful AI-attracted transformation involves the balanced combination of technology, strategy, and organizational preparedness to attain sustainable growth and long-term business success.
With the rapid convergence of the Internet of Things (IoT) and embedded robotics, Wi-Fi-driven mobile platforms have emerged as practical solutions for remote sensing, navigation, and automation. This paper presents the conceptual framework, hardware architecture, software design, and performance evaluation of a Wi-Fi-controlled robotic system built around an ESP8266/ESP32 microcontroller. The platform communicates with a user device-smartphone, tablet, or laptop-via TCP/IP over a local wireless network, enabling low-latency directional control and optional live data streaming. Compared with earlier radio-frequency (RF) or infrared (IR) paradigms, the proposed system achieves superior operating range, bandwidth, and expandability. Experimental outcomes confirm consistent command execution and stable wireless connectivity under typical indoor conditions. Potential deployment scenarios include surveillance, industrial inspection, search-and-rescue operations, and smart-home automation. The work underscores how consumer-grade Wi-Fi chipsets and open-source microcontroller frameworks can lower the barrier to building capable, network-aware robotic systems.
Light Fidelity (Li-Fi) is an emerging wireless communication technology that uses visible light for high-speed data transmission instead of conventional radio frequency waves. With the rapid growth of wireless devices, radio spectrum congestion and security concerns have become major challenges. Li-Fi offers an effective solution by utilizing light emitted from Light Emitting Diodes (LEDs) to transmit data through rapid modulation that is imperceptible to the human eye. At the receiver end, a photodiode detects the light signals and converts them into electrical signals for data recovery. This project focuses on the design and implementation of a Li-Fi communication system that demonstrates secure, fast, and interference-free data transmission in indoor environments. Compared to traditional Wi-Fi systems, Li-Fi provides higher data rates, improved security, and immunity to electromagnetic interference, making it suitable for sensitive areas such as hospitals, aircraft cabins, and defense applications. The system is energy-efficient as it utilizes existing lighting infrastructure for communication. Li-Fi technology has the potential to revolutionize future wireless communication by enabling smart lighting, Internet of Things (IoT) integration, and high-density data networks. This project highlights the working principle, advantages, and practical significance of Li-Fi as a next- generation communication technology.
Petri net modeling is an effective approach for analyzing the performance of industrial manufacturing systems. Its major advantage over conventional simulation techniques lies in its capability to represent actual operating conditions with greater accuracy. Data collected from manufacturing environments often contain significant uncertainty, and Petri net models are well suited to incorporate such variability to realistically capture the dynamic behavior of individual sub-systems within a plant. Through Petri net–based simulation, long-term availability of system components can be evaluated by simulating plant operations in a virtual environment. The outcomes of this analysis assist in identifying critical sub-systems that have a substantial impact on overall system availability. Based on these insights, targeted maintenance strategies can be developed for individual sub-systems, thereby minimizing production losses caused by equipment downtime. The applicability of the proposed approach is illustrated in this study using a complex repairable manufacturing system.
Perovskite ABO₃ materials have garnered significant attention due to their potential for electronic and microelectronic applications. Lead-based compounds, such as PbTiO₃, Pb(Zr,Ti)O₃, and Pb(Mg₁/₃Nb₂/₃)O₃, are widely utilized but pose environmental risks, prompting a search for lead-free alternatives. To this end, morphotropic phase boundary composition of Ba-modified (Bi0.5Na0.5)TiO3 is considered to be a promising candidate Ba₀.₀₆(Bi,Na)₀.₉₄TiO₃, which shows promising dielectric and piezoelectric properties. In order to further enhance the dielectric characteristics of Ba₀.₀₆(Bi,Na)₀.₉₄TiO₃, the Ti⁴⁺-ions have been substituted with the pseudo-cation (〖Ni〗_(1/3)^(3+) 〖Nb〗_(2/3)^(5+) )^(4+). Accordingly, in this study, lead-free solid solutions of (1-ϕ)Ba₀.₀₆Na₀.₄₇Bi₀.₄₇TiO₃–ϕBa₀.₀₆Na₀.₄₇Bi₀.₄₇(Ni₁/₃Nb₂/₃)O₃ (0 ≤ ϕ ≤ 1.0) have been synthesized using the solid-state reaction process. Also, the real and imaginary components of permittivity at 1 kHz were analysed as functions of filler (Ba₀.₀₆Na₀.₄₇Bi₀.₄₇(Ni₁/₃Nb₂/₃)O₃) concentration. The suitability of dielectric mixture equations, including Bruggeman, Furukawa, Rother-Lichtenecker, modified Rother-Lichtenecker, and Knott models, was evaluated. Additionally, a first-order exponential growth model provided an excellent fit to the data (r² > 0.99).
The semiconductor industry faces increasing challenges in maintaining high yields and reducing costs asmanufacturing processes become more complex. A new and effective tool for optimising processes is big dataanalytics, enabling manufacturers to extract valuable insights from vast amounts of production data and make data-driven decisions. This study proposes a comprehensive machine learning (ML) pipeline tailored for analyzingtelemetry data using the SECOM dataset from the UCI repository. The methodology includes data cleaning, missingvalue imputation, feature scaling via Min-Max normalization, dimensionality reduction, and Synthetic MinorityOversampling Technique (SMOTE) to handle class imbalance. A Decision Tree Classifier (DTC) is utilized to classifygood and defective products, achieving an accuracy of 88% in addition to excellent results in terms of recall, F1-score,ROC-AUC, and accuracy. Based on a comparison, the offered DTC model performs much better than populartraditional and deep learning techniques and can be trusted for spotting and addressing faults in real life.
A common kind of cancer is breast cancer. Raising the survival rate of breast cancer patients is mainly dependent onbreast cancer recurrence prognosis. The accuracy of cancer detection and diagnosis has increased with the progressof technology and ML approaches. Machine learning (ML) provides a number of statistical and probabilisticapproaches. This study introduces a deep learning-based approach to automatically classify breast cancer imagesfrom the BreakHis dataset. Feature extraction was performed using a Convolutional Neural Network (CNN) toautomatically detect significant tissue structures. The MobileNetV2 architecture was employed for its efficiency inhandling large-scale data while maintaining high classification accuracy. The model achieved an impressive accuracyof 98.18%, with precision of 98.38%, sensitivity of 97.37%, and an F1score of 97.85%. When compared to otherarchitectures, MobileNetV2 outperformed Xception, ResNet101, and EfficientNet, which demonstrated lower accuracyand sensitivity. These results highlight the potential of MobileNetV2 for reliable, fast, and cost-effective breast cancerdetection, offering a promising tool for clinical applications.
The fast change of telecoms to multi-layer, cloud-native, and extremely distributed architectures, in effect, increased the attack surface to a level where the conventional security models that rely on perimeters are no longer effective. Advanced breaches that exploit the identity compromise, lateral movement, and API vulnerability would demand a full shift of the focus of identity-centric to security. how Zero Trust concepts and an Identity Fabric can secure multilayer telecom networks traversing Radio Access Network, transport, core, service, management, cloud, and edge layers with identity as the primary control plane, the proposed Zero Trust Identity Fabric would permit ongoing authentication, highly detailed authorization, and real-time risk evaluation of users, devices, workloads, and network functions. The study examines the identity requirements of telecom layers and the identity threats, and also represents a conceptual architecture consisting of points of policy decision and enforcement, identity providers, and device posture management. A stable and scalable digital platform, and so, on which the advantages consist of the following: reduced lateral flow, policy enforcement, heterogeneous environment interoperability, and endurance through ongoing monitoring and analytics. Zero Trust Identity Fabric integration is a necessity to ensure that the telecom ecosystems in the present or the future are flexible, scalable, and secure.
One of the most admirable uses of IoT is in healthcare. Doctors can use the Internet to help people because to IoT. The distance between the patient and the physician can be greatly decreased with portable Internet of Things-based health monitoring equipment. may approach each patient separately, assess their current state of health, and determine the best course of action for them thanks to IoT. Healthcare is fast changing thanks to the Internet of Things (IoT), which promises individualized treatment and eases systemic stress. Devices whispering updates to physicians, tablets reminding patients to take them, and remote monitoring for the elderly are all possible with the development of networked systems. IoT is growing in the developing world because to a number of issues, including a lack of access to healthcare, an increase in chronic illnesses, high healthcare costs, and a growing need for telemedicine. There are suggestions for future research directions along with a presentation of the security, privacy, wearability, and low-power operation challenges that the healthcare IoT faces and conduct a thorough literature assessment of current research on IoT-based healthcare monitoring systems. review. The effectiveness, efficiency, data protection, privacy, security, and monitoring of various systems are compared in the literature study. The report concludes with recommendations and ideas for IoT healthcare applications as well as future directions pertaining to a number of current technological advancements.
Financial organizations face growing threats to their security because of digital banking along with online financial transactions. The research demonstrates a method to boost financial security that implements machine learning anomaly detection algorithms on fraudulent payment systems. The research utilizes the Credit Card Fraud (CCF) dataset with substantial discrepancy between authentic and fraudulent records while executing comprehensive data preprocessing techniques that utilize outlier identification methods in addition to random under-sampling strategies. The important features are comprised of 31 attributes that include anonymized variables (V1–V28) and transaction parameters (time and amount) with their assigned class label. The data has been partitioned into training, which takes up 70%, and testing, which occupies 30%. The method known as Isolation Forest (iForest) turns out to be the most effective classifier when tested on anomalous transactions with 98.65% accuracy coupled with 98.20% precision along with 98.64% recall and 98.52% F1-score performance. Anomaly detection-based machine learning methods indicate their clear ability to detect fraudulent transactions through both precise and high-recall manner. The results prove that sophisticated machine learning systems function as effective security instruments to stop financial system fraud.
