Overview Of Architecture Research Articles

SIGN-LANGUAGE TRANSLATOR USING MACHINE LEARNING

This research introduces a cutting-edge system designed to facilitate communication between individuals who are deaf and those who are hearing. By employing sophisticated machine learning algorithms and specific convolutional neural networks (CNNs), the proposed system can effectively sign language gestures into text or speech. A meticulously curated collection of sign language gestures serves as the foundation for training the model, ensuring its proficiency in accurately classifying several hand shapes and positions. To optimize classification performance, the system incorporates data preprocessing techniques that highlight the most distinctive features of the hands, thereby streamlining the computational process. This paper provides a comprehensive overview of the system's architecture, training methodology, and evaluation results, emphasizing the critical role of machine learning in developing inclusive communication tools that empower the deaf community. Future research initiatives will focus's on expanding the gesture dataset and refining real-time processing capabilities to further enhance the system's effectiveness and accessibility. Key Words: CNN, TensorFlow, Keras, NLTK

Tradeoffs in Key Rotation Strategies for Industrial Internet of Things Devices and Firmware

This paper provides an overview of several secure boot architectures with a focus on key rotation. It expands on a practitioner note that the authors submitted to the 2023 IEEE Secure Development Conference. Key rotation is important due to the frequency of lost signing keys and the difficulty of managing secret keys for the long lifetimes of Industrial Internet of Things (IIOT) devices. Key rotation is not simple for IIOT due to limited resources during a secure boot process and the constraints of the firmware utilities that come from the chip vendors. This paper reviews and compares five common architectures for a secure boot that are seen across the IIOT community. For each architecture, it provides some key strengths and weaknesses associated with that architecture. The paper then provides a detailed comparison and analysis of the architectures to convince the IIOT community to move towards a strong use of certificates (instead of the traditional use of raw public keys). The intent of this paper is to provide a practitioner’s perspective on these challenges and the tradeoffs in hopes of inviting comments from chip vendors and the broader community.

DIANA: An underwater analog space mission

The DIANA mission represents an underwater analog space mission designed to simulate and study the impact of long-duration spaceflight and extraterrestrial habitation on crew performance, psychosocial dynamics, and technological systems. The mission utilized an underwater habitat, the Hydronaut H03 DeepLab, to mimic the isolated, confined, and extreme (ICE) environment of space. Over eight days, a six-member crew lived and worked in underwater (3) and water surface (3) habitats, performing scientific experiments and operational tasks. The mission schedule encompassed a variety of activities such as drone exploration, extravehicular activities (EVAs), soil sampling, and media interactions, culminating in a simulated departure from the lunar surface. Data collection methods included continuous biomedical monitoring, cognitive task assessments, and sociomapping to analyze team communication and cooperation. This paper provides an overview of the mission architecture and outcomes, offering valuable insights into the challenges of future human space exploration and informing improvements in crew selection, training, and support systems.

The Community Solid Server: Supporting research & development in an evolving ecosystem

The Solid project aims to empower people with control over their own data through the separation of data, identity, and applications. The goal is an environment with clear interoperability between all servers and clients that adhere to the specification. Solid is a standards-driven way to extend the Linked Data vision from public to private data, and everything in between. Multiple implementations of the Solid Protocol exist, but due to the evolving nature of the ecosystem, there is a strong need for an implementation that enables qualitative and quantitative research into new features and allows developers to quickly set up varying development environments. To meet these demands, we created the Community Solid Server, a modular server that can be configured to suit the needs of researchers and developers. In this article, we provide an overview of the server architecture and how it is positioned within the Solid ecosystem. The server supports many orthogonal feature combinations on axes such as authorization, authentication, and data storage. The Community Solid Server comes with several predefined configurations that allow researchers and developers to quickly set up servers with different content and backends, and can easily be modified to change many of its features. The server will help evolve the specification, and support further research into Solid and its possibilities.

Tabular deep learning: a comparative study applied to multi-task genome-wide prediction

PurposeMore accurate prediction of phenotype traits can increase the success of genomic selection in both plant and animal breeding studies and provide more reliable disease risk prediction in humans. Traditional approaches typically use regression models based on linear assumptions between the genetic markers and the traits of interest. Non-linear models have been considered as an alternative tool for modeling genomic interactions (i.e. non-additive effects) and other subtle non-linear patterns between markers and phenotype. Deep learning has become a state-of-the-art non-linear prediction method for sound, image and language data. However, genomic data is better represented in a tabular format. The existing literature on deep learning for tabular data proposes a wide range of novel architectures and reports successful results on various datasets. Tabular deep learning applications in genome-wide prediction (GWP) are still rare. In this work, we perform an overview of the main families of recent deep learning architectures for tabular data and apply them to multi-trait regression and multi-class classification for GWP on real gene datasets.MethodsThe study involves an extensive overview of recent deep learning architectures for tabular data learning: NODE, TabNet, TabR, TabTransformer, FT-Transformer, AutoInt, GANDALF, SAINT and LassoNet. These architectures are applied to multi-trait GWP. Comprehensive benchmarks of various tabular deep learning methods are conducted to identify best practices and determine their effectiveness compared to traditional methods.ResultsExtensive experimental results on several genomic datasets (three for multi-trait regression and two for multi-class classification) highlight LassoNet as a standout performer, surpassing both other tabular deep learning models and the highly efficient tree based LightGBM method in terms of both best prediction accuracy and computing efficiency.ConclusionThrough series of evaluations on real-world genomic datasets, the study identifies LassoNet as a standout performer, surpassing decision tree methods like LightGBM and other tabular deep learning architectures in terms of both predictive accuracy and computing efficiency. Moreover, the inherent variable selection property of LassoNet provides a systematic way to find important genetic markers that contribute to phenotype expression.

Devika AI: An Open-Source Revolution In AI-Powered Code Generation

Devika AI is an innovative open-source platform designed to revolutionize software development through AIpowered code generation. Leveraging advanced natural language processing (NLP), machine learning algorithms, and a modular architecture, Devika AI translates high-level human instructions into executable code across multiple programming languages. This paper presents a comprehensive overview of Devika AI's architecture, methodologies, and functionalities. We discuss its ability to improve developer productivity, reduce coding errors, and facilitate collaborative development. Experimental results demonstrate Devika AI's effectiveness in various coding tasks, highlighting its potential as a free and accessible alternative to proprietary AI coding assistants. The implications of Devika AI's adoption on the future of software development are also explored

Modeling Brain Functional Networks Using Graph Neural Networks: A Review and Clinical Application

The integration of graph neural networks (GNNs) with brain functional network analysis is an emerging field that combines neuroscience and machine learning to enhance our understanding of complex brain dynamics. We first briefly introduce the fundamentals of brain functional networks, followed by an overview of Graph Neural Network principles and architectures. The review then focuses on the applications of these networks and address current challenges in the field, such as the need for interpretable models and effective integration of multi-modal neuroimaging data. We also highlight the potential of GNNs in clinical perimenopausal areas such as perimenopausal depression research, demonstrating the broad applicability of this approach. The review concludes by outlining future research directions, including the development of more sophisticated architectures for large-scale, heterogeneous brain graphs, and the exploration of causal inference in brain networks. By synthesizing recent advances and identifying key research directions, this review aims to summarize the focal points of brain functional network analysis and GNNs, explore the potential of their integration, and provide a reference for advancing this interdisciplinary field.

Drug Tracing System Using Blockchain Technology

Healthcare supply chains are complicated networks that span various departmental and geographical limits and provide the foundation for many benefits that are necessary for everyday life .The difficulties of this systems can result in contaminations such as erroneous data and lack of clarity and poor data origin. one issue of present supply chain restrictions is the production of counterfeit drugs, which not only has negative impact on people health but also has negative impact on financial status of pharmaceutical business. A recent research emphasised the importance of a well founded end-to-end track and trace system for health care supply chains .The smart contract ensures data provenance and gives all participants access to a secure unalterable transaction history .Moreover, it eliminates the need for intermediaries we give a high-level overview of thesystem architecture and algorithms that form the basis of the key elements of our proposed solution. we test and analyse the system to find how effectively it can enhance traceabilityinside health care supply chains before providing a cost and security analysis Keywords— Drug Counterfeiting , Supply Chain, Blockchain, Traceability, Healthcare, Trust, Securit

Deep Learning Models for PV Power Forecasting: Review

Accurate forecasting of photovoltaic (PV) power is essential for grid scheduling and energy management. In recent years, deep learning technology has made significant progress in time-series forecasting, offering new solutions for PV power forecasting. This study provides a systematic review of deep learning models for PV power forecasting, concentrating on comparisons of the features, advantages, and limitations of different model architectures. First, we analyze the commonly used datasets for PV power forecasting. Additionally, we provide an overview of mainstream deep learning model architectures, including multilayer perceptron (MLP), recurrent neural networks (RNN), convolutional neural networks (CNN), and graph neural networks (GNN), and explain their fundamental principles and technical features. Moreover, we systematically organize the research progress of deep learning models based on different architectures for PV power forecasting. This study indicates that different deep learning model architectures have their own advantages in PV power forecasting. MLP models have strong nonlinear fitting capabilities, RNN models can capture long-term dependencies, CNN models can automatically extract local features, and GNN models have unique advantages for modeling spatiotemporal characteristics. This manuscript provides a comprehensive research survey for PV power forecasting using deep learning models, helping researchers and practitioners to gain a deeper understanding of the current applications, challenges, and opportunities of deep learning technology in this area.

RD53 pixel chips for the ATLAS and CMS Phase-2 upgrades at HL-LHC

The Phase-2 upgrades at the High-Luminosity LHC of ATLAS and CMS experiments at CERN will require a new tracker with readout electronics operating in extremely harsh radiation environment (1 Grad), high hit rate (3.5 GHz/cm2) and high data rate readout (5 Gb/s). The RD53 collaboration is a joint effort between the ATLAS and CMS to qualify the chosen 65 nm CMOS technology in high radiation environment and develop the pixel readout chips of both experiments. After a half-scale demonstrator (RD53A) and full scale prototypes of the two ASICs (RD53B-ATLAS and RD53B-CMS), largely used by the two communities to characterize 3D and planar sensors, RD53 developed and submitted to foundry in 2023 the production chips. A general overview of the chip architecture will be described.

High-Performance Spatial Data Analytics: Systematic R&D for Scale-Out and Scale-Up Solutions from the Past to Now

We released open-source software Hadoop-GIS in 2011, and presented and published the work in VLDB 2013. This work initiated the development of a new spatial data analytical ecosystem characterized by its large-scale capacity in both computing and data storage, high scalability, compatibility with low-cost commodity processors in clusters and open-source software. After more than a decade of research and development, this ecosystem has matured and is now serving many applications across various fields. In this paper, we provide the background on why we started this project and give an overview of the original Hadoop-GIS software architecture, along with its unique technical contributions and legacy. We present the evolution of the ecosystem and its current state-of-the-art, which has been influenced by the Hadoop-GIS project. We also describe the ongoing efforts to further enhance this ecosystem with hardware accelerations to meet the increasing demands for low latency and high throughput in various spatial data analysis tasks. Finally, we will summarize the insights gained and lessons learned over more than a decade in pursuing high-performance spatial data analytics.

Development of an Anti-rollover Warning and Active Intervention Controller for Wheeled Tractor

Background: Tractor rollover accidents are the most severe problem in agricultural production. According to the statistical analysis of the rollover incidents, some rollover accidents can be avoided by timely anti-rollover warning and active control. There are a few researches on the anti-rollover control for agricultural tractors, although great progress in the research on tractor security has been achieved, such as rollover protective structure design. Objective: The purpose of this patent study was to develop a controller with the function of warning as well as active anti-rollover control employing the power motor installed on the steering column to adjust the front wheel angle and control the rollover index within the safety range. Methodology: The structure block for the hardware of the controller was devised that included data obtaining, rollover risk estimation, rollover warning, and front-wheel steering control sections. The main circuits have been designed in detail by adopting module methods, such as motor control circuit, positioning and transmission circuit, etc. The control scheme has been demonstrated by designing an overview of the overall architecture, including rollover indicator calculation, load transfer ratio (LTR) controller design, and front-wheel steering angle tracking controller design. C language has been used to accomplish the software design. Results: The active anti-rollover controller can calculate and keep the LTR within the stable range by adjusting the front wheel angle in real-time. The fast response and smoothness of the results demonstrate that the controller can fulfill the designed function. Conclusion: The developed controller can meet the control requirement due to several advantageous aspects, such as reliability, effectiveness, and practicability. It is prospective for the developed controller to provide a valuable reference to reduce tractor rollover accidents.

Blockchain technology for advanced therapy medicinal products: Applications in tracking, data sharing, and supply chain automation

Advanced therapy medicinal products (ATMPs) like cell and gene therapies offer transformative treatment options for many diseases. However, coordinating the decentralized, patient-specific manufacturing of autologous ATMPs across multiple hospitals poses major supply chain challenges. This paper provides a comprehensive analysis of how blockchain technology can enhance decentralized ATMP manufacturing networks. First, background on ATMPs and complexities of decentralized production is reviewed. An overview of blockchain architecture, key attributes, and existing use cases then follows. The major opportunities for blockchain integration in ATMP manufacturing are discussed in depth, including tracking autologous products across locations, enabling data sharing between hospitals to power AI-based optimization, automating supply chain processes, and maintaining provenance records. Critical limitations around scalability, privacy, regulation, and adoption barriers are examined. Design considerations for developing blockchain ecosystems tailored to the unique ATMP environment are also explored. Blockchain shows immense promise for transforming visibility, coordination, automation, and data unification in decentralized ATMP manufacturing networks. Despite current challenges, blockchain is prepared to profoundly impact the advancement of personalized cell and gene therapies through enhanced supply chain instrumentation. This paper provides a comprehensive analysis of this emerging technological innovation and its applications to address critical needs in ATMP translation and manufacturing.

IoT-Based Framework for Implementation of Leaf Color Chart for Nitrogen Monitoring in Crop

Effective nitrogen management is essential in precision agriculture to optimize crop yields and promote environmental sustainability. Conventional techniques for evaluating nitrogen levels, although commonly used, suffer from a lack of accuracy and scalability. This research presents a novel design framework that combines the Leaf Colour Chart (LCC), a well-established agricultural tool for nitrogen evaluation, with modern Internet of Things (IoT) technologies. The proposed system utilizes specialized sensors to gather precise leaf color data, which is subsequently analyzed using cloud-based computer vision techniques to correctly and instantaneously predict nitrogen levels. The framework streamlines data gathering and processing, allowing for accurate nitrogen management. This enables the application of fertilizer to specific areas and minimizes wastage. This article provides an overview of the system architecture, examines the difficulties and resolutions in integrating Internet of Things (IoT) technology in agriculture, and showcases case studies that demonstrate the effectiveness of the framework in practical environments. The incorporation of IoT technology with the Leaf Colour Chart signifies a notable progression in agricultural technology, offering the potential to enhance crop management techniques and support sustainable agriculture endeavors.

StateOS : A Memory Sufficient Hybrid Operating System for IOT Devices

The proliferation of Internet of Things (IoT) technology has spurred the need for efficient operating systems tailored to the constraints of IoT devices. This paper presents StateOS, a hybrid operating system designed to meet the demands of sensor-based IoT devices and wireless sensors. StateOS combines the memory efficiency of event-driven systems with the clarity of control flow provided by multithreaded approaches. By implementing a microkernel architecture, cross-layer network protocol design, and a hybrid task scheduler, StateOS offers real-time capabilities while minimizing memory footprint. Macro-based task APIs and state machine-based visual programming facilitate cooperative threaded programming, making StateOS suitable for both experienced developers and novices. The paper provides an in-depth overview of StateOS's architecture, kernel functionalities, code examples, implementation on various platforms, performance evaluation, and concludes by highlighting its potential as a valuable tool for modern IoT application development. Keywords — Hybrid operating system (OS), Internet of Things (IoT) and Memory efficiency.

Human Gait Recognition using Machine Learning Teachnolgies for Inclusive Innovation

Gait recognition, or the ability to identify people based on how they walk, is used in a variety of contexts, including human computer interaction, security checks, and health monitoring. Due to Out of touch and uncooperative individuals, gait-based human recognition is an emerging behavioural biometric feature for intelligent surveillance monitoring. In video surveillance, gait recognition can be used to detect objects at a distance and assist in low-resolution object identification. Recent years have seen an explosion in the study of gait analysis for a wide range of uses, such as animation, video surveillance, health monitoring, and authentication. A sophisticated new technology called gait recognition can identify persons Shoot from a distance and perform well on movies with no resolution. This document provides various walkthroughs written in the form of examples and free samples. Four main processes make up the survey of gait detection algorithms covered in this study: feature extraction, classification, preprocessing, and data collection. The descriptions of the pathology database, Vision base, and wearable sensor are compiled. In recent years, deep architectures have made great progress in improving human recognition performance. This article provides an up-to-date overview of deep architectures for gait recognition, highlighting the use of convolutional neural networks along with other architectures. Furthermore, the overall problems of gait recognition are examined together with potential directions for future research.

Integrated Extended Kalman Filter and Deep Learning Platform for Electric Vehicle Battery Health Prediction

As the demand for electric vehicles (EVs) rises globally, ensuring the safety and reliability of EV battery systems becomes paramount. Accurately predicting the state of health (SoH) and state of charge (SoC) of EV batteries is crucial for maintaining their safe and consistent operation. This paper introduces a novel approach leveraging deep learning methodologies to predict battery SoH, focusing on implementing a system prototype for real-world applications. The proposed system integrates an extended Kalman filter (EKF) with a deep learning framework, forming a system prototype known as FELL, aimed at EV battery diagnosis and prediction. We devise an algorithm utilizing the EKF to estimate the SoH of the battery. We present a detailed overview of the system architecture and implementation, showcasing its predictive capabilities. Experimental results demonstrate the effectiveness of the system in accurately estimating battery SoH with notable improvements in prediction accuracy. Additionally, the FELL system provides users with real-time predictions and comparative analysis across multiple prediction models, offering valuable insights for EV battery management.

Scalable Cloud Architectures for Deploying AI Applications

The integration of Artificial Intelligence (AI) into cloud computing represents a pivotal shift in the technological landscape, offering unprecedented opportunities for innovation across industries. Deploying AI applications at scale on the cloud poses unique challenges, necessitating the development of scalable cloud architectures tailored to meet the computational demands and data processing needs inherent to AI. This article explores the foundational aspects of cloud computing architectures, emphasizing the principles of scalability essential for AI applications. It delves into the core challenges faced when deploying AI on the cloud, such as computational requirements, data management, network constraints, and security concerns. Further, it presents various architectural models that facilitate scalability, including containerization, serverless computing, and cloud-native AI services, drawing from realworld case studies to illustrate effective strategies and best practices. Additionally, the article examines performance optimization techniques, security considerations, and the future directions of cloud-based AI deployments, highlighting the role of emerging technologies such as quantum computing and edge AI. By providing a comprehensive overview of scalable cloud architectures for AI applications, this article aims to guide researchers, practitioners, and organizations in leveraging cloud computing to its full potential, thereby enabling more efficient, secure, and scalable AI solutions.

Context-Aware IoT System Development Approach Based on Meta-Modeling and Reinforcement Learning

Integrating context awareness into the Internet of Things systems is essential for enhancing their adaptability to their context, particularly their user preferences and behaviors. This paper proposes an approach to model and develop context-aware self-adaptive IoT systems, capable of adapting their actions according to their users’ preferences. The approach consists of three main axes. The first axis involves establishing an overview of the system architecture that provides a high-level understanding of the various components of a context-aware IoT system. The second axis concerns the creation of a context-aware IoT systems meta-model, encapsulating the essential elements, relationships, and dependencies governing context awareness within the IoT system in a domain-independent manner. The third axis proposes a reinforcement learning reasoning process to enable intelligent decision-making within context- aware IoT systems. To validate the feasibility of the proposed approach, a simulation was conducted using the OpenAI Gym framework to emulate a context-aware smart home system. The results highlight the feasibility of the approach, and its potential to enhance real-life IoT systems’ awareness of their users’ context.

A real-time digital twin for active safety in an aircraft hangar

The aerospace industry prioritises safety protocols to prevent accidents that can result in injuries, fatalities, or aircraft damage. One of the potential hazards that can occur while manoeuvring aircraft in and out of a hangar is collisions with other aircraft or buildings, which can lead to operational disruption and costly repairs. To tackle this issue, we have developed the Smart Hangar project, which aims to alert personnel of increased risks and prevent incidents from happening. The Smart Hangar project uses computer vision, LiDAR, and ultra-wideband sensors to track all objects and individuals within the hangar space. These data inputs are combined to form a real-time 3D Digital Twin (DT) of the hangar environment. The Active Safety system then uses the DT to perform real-time path planning, collision prediction, and safety alerts for tow truck drivers and hangar personnel. This paper provides a detailed overview of the system architecture, including the technologies used, and highlights the system’s performance. By implementing this system, we aim to reduce the risk of accidents in the aerospace industry and increase safety for all personnel involved. Additionally, we identify future research directions for the Smart Hangar project.