This study proposed an intervention model for improving the teaching ability of teachers in general practice standardized training bases through intelligent computing and decision support. Focusing on the scenes of teaching rounds, case discussion, outpatient guidance and skill teaching, a data set containing 126 teachers from 8 general practice standardized training bases, 18,640 behavior records, 4,320 trainees 'feedback and 1,260 structured evaluation forms was constructed. The teaching frequency, interaction intensity, supervision stability and feedback consistency are encoded by the timing feature representation module, and the intervention suggestions for different ability shortboards are generated by the adaptive scheduling algorithm. A system consisting of data access layer, capability analysis layer, decision service layer and interactive application layer was developed to support data maintenance, identification, intervention allocation and result tracking. Experimental results show that the accuracy of this method is 93.1%, the F1 value is 91.8%, and the average response time is 1.4 s. It has good application and deployment value in complex clinical teaching scenarios.

In the internet of medical things, data primarily exhibits time-series and streaming characteristics, featuring typical attributes such as large-scale volume, high transmission rates, and significant heterogeneity. Given these data properties and the application requirements of medical scenarios, the development of specialized data platforms tailored to these needs holds considerable research significance and practical value. This study innovatively proposes the internet of medical things data platform solution based on a cloud-edge-end architecture, and elaborates on its architecture, functions, and implementation effects. The edge side is responsible for streaming data access, storage, and computation; the cloud side encompasses three layers of services: resources, data, and applications, constructing a data lake to provide data analysis services. This study has been implemented in PLA General Hospital for verification. From 2021 to 2024, 263 medical devices have been connected accumulatively, with a total data volume of 24.07 TB and stable operation within 4 years. In the performance stress test, the platform achieved the data access throughput of 23.91 MB/s and the data storage efficiency of 30.98 MB/s. These results demonstrate the feasibility of the architecture platform. This study has engineered and successfully applied the cloud-edge-end architecture in complex internet of medical things scenarios, addressing challenges such as heterogeneous protocol compatibility of medical devices, real-time response to clinical operations, and large-scale storage and application of the internet of things data. The established data platform provides a solid data foundation for smart medical applications and holds significant value for the research of medical artificial intelligence and the construction of future smart hospitals.

The development of e-commerce in Indonesia has shown an upward trend, with gross merchandise value (GMV) reaching USD 59 billion in 2022. This study aims to develop a mobile-based e-commerce application, with a case study of the Berlian Tech Online Store. This study employs a Research and Development (R&D) method using a software engineering approach based on the Software Development Life Cycle (SDLC) with the waterfall model. The waterfall model provides a sequential lifecycle approach, beginning with the analysis phase and proceeding to design, coding, testing, and maintenance. The requirements analysis is divided into two categories: functional requirements and non-functional requirements. The system architecture comprises four entities: authentication, products, and the shop. The software architecture adopts the Model-View-ViewModel (MVVM) pattern due to its superior CPU efficiency, with an average CPU usage of 8.92% and a memory consumption of 121.48 MB. The MVVM architecture is combined with a service layer and an API gateway. The Online Store application was successfully designed using the Model-View-ViewModel (MVVM) system architecture, combined with a service layer that facilitates communication between the API gateway and the view. The NoSQL database implemented with Firebase provides high flexibility for storing data in a non-relational structure. The results of black-box testing indicate that all core system features, including user authentication, product management, shopping cart, checkout, and transaction processing, function properly and align with user requirements.

ABSTRACT To address the issues of lengthy encryption time, low clustering accuracy, and poor performance in existing privacy‐preserving clustering methods for grid data, this paper proposes a fault‐tolerant data clustering method for smart grids based on the Boneh‐Goh‐Nissim (BGN) homomorphic encryption algorithm. A system architecture is constructed comprising a cloud server layer, a fog node layer, a smart meter layer, and a trusted third party. Private data collected by smart meters are first denoised using robust locally weighted regression. The preprocessed data are then encrypted with the BGN algorithm. K‐means clustering is applied to mine valuable data, with decryption performed at the cloud server layer. A two‐layer fault‐tolerant mechanism—utilizing secure channels to trusted organizations—is implemented across both the cloud servers and smart meters to ensure robust privacy‐preserving clustering. Experimental results in a scenario with 100 smart meters show the proposed method requires only 10 s for encryption, significantly less than conventional methods. Clustering performance is excellent: valid and invalid data are clearly distinguished, the deviation between actual and computed cluster centers is small, and clustering accuracy reaches 89.54%. Furthermore, by integrating a Shamir (3,5) threshold secret sharing scheme and a redundancy strategy for fog node data storage, the method maintains continuous operation and data integrity despite server failures or meter data loss. The data recovery rate exceeds 98%, with less than 4% loss in clustering accuracy. These results demonstrate the method achieves efficient encryption, high clustering accuracy, and strong fault tolerance, effectively enhancing private information security in smart grids.

Advanced microarchitecture courses demand rich, hands-on exploration of pipelines, caches, and memory systems. Though traditional hardware labs are costly, location-bound, and difficult to scale for remote or hybrid delivery. This paper presents a cloud-hosted, multi-user virtual reality (VR) laboratory designed specifically for advanced microarchitecture education and remote teaching. The proposed platform delivers an experiment-rich environment where students collaboratively inspect, instrument, and modify microarchitectural components such as pipeline stages, cache hierarchies, and branch predictors in real time. Architecturally, the system combines a web-based cloud front-end for authentication and session management, a scalable VR services layer providing multi-user scenes and collaboration tools, and a backend to adapt microarchitecture simulators whose internal state is visualized in 3D. Pedagogically, define learning objectives around instruction-level parallelism, hazard analysis, and memory hierarchy behavior, and instantiate these through structured labs on pipeline hazards, cache performance, and branch prediction. A mixed-method evaluation in an advanced microarchitecture course contrasts a control group using traditional 2D tools with an experimental group using the VR lab over 4–6 weeks. According to quantitative findings, the VR group outperformed the control group by 24.5% in post-test scores, decisive the idea that immersive collaborative contact is a key factor in conceptual understanding. The cloud-based architecture may provide responsive multi-user experiences under practical bandwidth limits, as shown by system-level metrics like latency and frame rate. Future integration of hardware-in-the-loop and adaptive learning analytics will be informed by qualitative feedback from instructors and students that demonstrates how immersive visualization and collaborative interaction may demystify complicated microarchitectural behavior and ease remote teaching.

The tension between open source and proprietary software constitutes one of the defining structural conflicts of the contemporary digital economy — a conflict that is simultaneously technical, economic, political, and philosophical. This article presents a comprehensive comparative and analytical mapping of both paradigms across their technical, economic, organizational, governance, social, and geopolitical dimensions, drawing on peer-reviewed scholarship from information systems research, political economy, organizational theory, science and technology studies, and legal analysis. The analysis traces the historical emergence of proprietary software as a commercial category following IBM's 1969 unbundling decision; the countervailing institutionalization of the free software and open source movements from 1983 onward; and the subsequent decades of competitive and collaborative interaction that have produced the hybrid landscape of the present. Across the technical domains examined — operating systems, databases, programming languages, cloud infrastructure, development tooling, artificial intelligence, and cryptographic security — the article documents a structural asymmetry: open source has achieved categorical dominance at the infrastructure layer of the digital economy, while proprietary models maintain dominance at the interface and service layers where economic and political power is concentrated. Key tensions analyzed include the free rider problem and chronic underinvestment in open source commons; the strategic commoditization logic by which large technology corporations selectively open-source components to erode competitors' proprietary advantages while retaining proprietary control at value-capturing layers; the governance risks of maintainer burnout, corporate capture, and succession failure; the emerging conflicts between regulatory frameworks — including the EU AI Act, the Cyber Resilience Act, and data protection legislation — and the open source ecosystem; and the geopolitical dimensions of digital sovereignty in which the open/proprietary choice has become a matter of national strategic policy. Six paradigmatic case studies — Linux versus Windows Server, Android's open core architecture, HashiCorp's relicensing of Terraform, the Redis and Elasticsearch licensing conflicts with Amazon Web Services, Meta's LLaMA strategy, and Kubernetes' multi-stakeholder governance — are analyzed to ground the theoretical framework in empirically documented outcomes. The article concludes that the productive question is not which model is superior in the abstract, but rather who benefits from each model, under what institutional conditions, and at which layer of the technology stack — a distributional and contextual analysis that reveals the open source versus proprietary debate to be, at its deepest level, a contest over who holds the power to define how the digital world functions.

This paper presents a taxonomy-based survey of AI-driven network optimization mechanisms relevant to the transition from fifth generation (5G) to sixth generation (6G) mobile communication systems. In contrast to earlier generational shifts that are often described as technology replacement cycles, the 5G-to-6G evolution is increasingly characterized in the literature as a prolonged period of coexistence, hybrid operation, and progressive integration of new capabilities across radio, edge, core, and service layers. To structure this transition, the paper organizes prior work into a transition-oriented taxonomy covering migration strategies, AI-enabled closed-loop control, RAN disaggregation and edge intelligence, core virtualization and slice orchestration, spectrum-aware coexistence, service-driven requirements, and security-aware governance. Rather than introducing a new optimization algorithm or an experimentally validated architecture, the contribution of this survey is analytical and integrative. Specifically, it consolidates fragmented research directions into a reference view of how AI-driven control mechanisms are distributed across spectrum, RAN, edge, and core domains during hybrid 5G–6G operation. In addition, the paper includes a structured evidence synthesis of performance trends, deployment maturity signals, and recurring methodological limitations reported across the literature. The review indicates that meeting anticipated 6G objectives, including ultra-low latency, high reliability, scalability, and improved energy efficiency, depends less on isolated enhancements at individual protocol layers and more on coordinated cross-layer optimization supported by AI-native control loops. At the same time, the surveyed literature reveals persistent gaps in service-to-control mapping, security-aware orchestration, interoperability across heterogeneous domains, and reproducible evaluation methodologies for hybrid 5G–6G environments. The survey is intended to provide researchers, network operators, and standardization stakeholders with a structured analytical basis for assessing how AI-driven optimization can support the staged evolution from 5G systems toward 6G-ready infrastructures.

Unmanned underwater vehicles (UUVs) play a pivotal role in marine applications such as resource exploration, maritime search and rescue. However, communication signal loss remains a critical bottleneck, constraining UUV autonomous operation and mission reliability across four dimensions: navigation, coordination, monitoring, and planning. To address these challenges in communication-denied environments, this paper proposes a UUV digital twin system utilizing motion prediction technology, such as virtual mapping, prediction, and autonomous decision support. Based on a four-layer architecture—comprising the Physical Entity Layer, Virtual Entity Layer, Twin Data & Connectivity Layer, and Services Layer, the system achieves full-state mapping and real-time visualization. Specifically, a hybrid prediction model integrating Transformer and Convolutional Neural Networks (CNN) architectures is developed to extract multi-scale features for resistance prediction, which serves as the critical basis for UUV motion state forecasting. Experimental validation confirms the system’s capability for real-time resistance tracking and high-precision prediction, providing a robust foundation for autonomous navigation control and energy management. These results advance the development of specialized UUV digital twin systems and establish a robust foundation for their engineering applications.

As sixth-generation (6G) mobile communication accelerates its evolution towards the vision of a satellite-terrestrial integrated network, the deep integration of terrestrial networks (TNs) and non-terrestrial networks (NTNs) has become an inevitable trend for achieving seamless, ubiquitous, and ultra-reliable global communication services. From the perspective of artificial intelligence (AI) technology development, this paper systematically elucidates the application potential of key technologies such as machine learning (ML), deep learning (DL), and large language models (LLMs) in satellite communication networks. This paper first outlines the evolutionary trend of AI technology and analyzes its unique advantages in handling challenges such as highly dynamic topologies, resource constraints, and massive access in low Earth orbit (LEO) satellite networks. Then, from the three dimensions of the physical layer, the network layer, and the service layer, it delves into the application of AI in core scenarios such as channel estimation, anti-interference, transmission optimization, network topology design, intelligent routing, beam resource management, and satellite edge computing. Finally, addressing key challenges such as large-scale node deployment, highly dynamic topology changes, and wide-area seamless coverage, it proposes a systematic solution based on game theory, deep learning, network slicing, and network simulation, and looks forward to future research directions for AI-driven satellite-terrestrial integrated networks, providing important references for building a fully covered, low-latency, highly reliable, and intelligent next-generation mobile communication architecture.

Monitoring the degree of cutting tool wear plays a key role in ensuring the quality of machined parts. Intelligent systems for monitoring technological equipment using machine learning can improve machining quality and reduce unexpected equipment downtime by enabling timely operator response to degrading cutting tool quality. Predictive maintenance, unlike classical approaches, allows for assessing the current state of the cutting tool and making replacement decisions based on real-time data. This article presents the development of a scalable architecture for monitoring metal-cutting equipment for predictive maintenance. The proposed architecture consists of three layers: a hardware layer for data collection and preprocessing using a Raspberry Pi microcomputer, a server layer with microservice architecture, and a visualization layer for displaying the current equipment condition. The architecture provides for collecting data from sensors, extracting features from raw signals, sending them to the server for input to a machine learning model, and visualizing classification results in a web interface. The research also shows a prototype web interface to demonstrate the system concept. The key advantage of the proposed architecture is modularity, scalability, and low implementation cost due to the use of open-source technologies, making it accessible for small and medium-sized manufacturing enterprises. Keywords: system design, machine learning, predictive maintenance, condition monitoring, smart manufacturing, maintenance optimization.

Against the backdrop of the popularization and digital transformation of higher education, the economic pressure and psychological development needs faced by the impoverished student group in colleges and universities are intertwined, becoming an important issue affecting educational equity and the quality of talent cultivation. Reading therapy, as an economical, safe and profoundly influential psychological aid, is increasingly highlighting its value. However, a single university library has limitations in terms of resource reserves and professional service capabilities, making it difficult to implement large-scale, personalized and continuous reading intervention for impoverished students. This research takes "sharing, precision and empowerment" as its core concepts, aiming to explore how to rely on the inter-library mutual assistance mechanism of digital and intelligent libraries in universities within the province to build a systematic, collaborative and intelligent reading therapy support system for impoverished students in universities. The article first analyzes the current situation and predicament of reading therapy services carried out by university libraries at present, especially those for students from poor families. Secondly, it expounds the necessity and feasibility of combining digital intelligence technology with inter-library mutual assistance. Furthermore, from four dimensions: the resource layer, the platform layer, the service layer, and the mechanism layer, the theoretical model and implementation framework of this support system were constructed in detail, and a closed-loop service process including precise demand identification, intelligent resource matching, dynamic process intervention, and scientific effect evaluation was designed. Finally, the guarantee strategies and future prospects for the implementation of the system were proposed, with the aim of providing theoretical references and practical paths for promoting educational equity, enhancing the accuracy and effectiveness of psychological education in colleges and universities, and driving the intelligent transformation and service innovation of libraries. This study contains an explanatory table on the "core components and functions of the support system".

Modern remote sensing platforms use cloud-native architectures with standards-based separation of control and data planes to achieve cost and performance targets at the petabyte scale. Cloud-native stacks are designed around Cloud Optimized GeoTIFF (COG), a byte-range streamable format, SpatioTemporal Asset Catalogs for discoverable metadata management, and Open Geospatial Consortium (OGC) standards for interoperable service delivery. Performance is illustrated in distributed compute engines, including large-scale spatial joins and analyses over multi-temporal data. Service layer design considerations include tile-first APIs and tiled data pipelines, alongside aggressive tile caching at the content delivery network (CDN) and edge layers. Performance considerations model tile latency, time to first pixel, and egress efficiency in distributed systems. These standards enable elastic scalability of platforms and interactive visualization workflows to meet the variability in analytics consumption patterns. Demonstrations have established that performance can be improved with internal tiling, multi-resolution overviews, and HTTP range requests that reduce bandwidth and latency in a wide variety of client ecosystems.

Scientific data management is at a critical juncture, driven by exponential data growth, increasing cross-domain dependencies, and a severe reproducibility crisis in modern research. Traditional centralized data management approaches are not only struggling with data volume but also fail to address the fragmentation of research results across domains. This hinders scientific reproducibility and cross-domain collaboration and increases concerns about data sovereignty and governance. This article proposes FAIR and federated Data Ecosystems as an improved architectural pattern for future research data ecosystems. It tries to incorporate the latest advancements in decentralized, distributed systems into existing research infrastructure to promote cross-domain collaboration. Based on established patterns from Data Commons, Data Meshes, and Data Spaces, our approach focuses on a layered architecture that consists of governance, data, service, and application layers. With this, it could be possible to preserve domain-specific expertise and control while facilitating data integration through standardized interfaces and semantic enrichment. Key requirements include adaptive metadata management, simplified user interaction, robust security, and transparent data transactions. Our architecture supports compute-to-data as well as data-to-compute paradigms, implementing a decentralized peer-to-peer network that scales horizontally. This article aims to provide both an impulse for the technical architecture as well as concepts for a governance framework so that FAIR and federated Data Ecosystems could enable researchers to build on existing work while maintaining control over their data and computing resources. This could provide a practical path towards an integrated research infrastructure that respects domain autonomy as well as interoperability requirements.

The increasing decentralization of industrial processes in Industry 4.0 necessitates the distribution and coordination of resources such as machines, materials, expertise, and knowledge across organizations in a value chain. To facilitate effective operations in such distributed environments, it is essential to digitize processes and resources, establish interconnectedness, and implement a scalable management approach. The present paper addresses these challenges through the knowledge-based production planning (KPP) system, which was originally developed as a monolithic prototype. It is argued that the KPP-System must evolve towards a service-oriented architecture (SOA) in order to align with distributed and interoperable Industry 4.0 requirements. The paper provides a comprehensive overview of the motivation and background of KPP, identifies the key research questions that are to be addressed, and presents a conceptual design for transitioning KPP into an SOA. The approach under discussion is notable for its consideration of compatibility with the Arrowhead Framework (AF), a consideration that is intended to ensure interoperability with smart production environments. The contribution of this work is the first architectural concept that demonstrates how KPP components can be encapsulated as services and integrated into local cloud environments, thus laying the foundation for adaptive, ontology-based process planning in distributed manufacturing. In addition to the conceptual architecture, the first implementation phase has been conducted to validate the proposed approach. This includes the realization and evaluation of the mediator-based service layer, which operationalizes the transformation of planning data into semantic function blocks (FBs) and enables the interaction of distributed services within the envisioned SO-KPP architecture. The implementation demonstrates the feasibility of the service-oriented transformation and provides a functional proof of concept for ontology-based integration in future adaptive production planning systems.

Abstract The deployment of artificial intelligence (AI) technology in various emerging network applications has spawned a large number of computing tasks, which require dynamic collaboration of multi-dimensional resources from the perspective of communication and computing to meet service requirements such as ultra-low latency, ultra-high reliability, and ultra-fast response. In this paper, we introduce a Computing Integration Networking (CIN) architecture, which built to integrate ubiquitous heterogeneous resources in the Internet into a distributed pool, enabling CIN paradigm to fully utilize all available heterogeneous resources and provide efficient customized services evolving both communication and computation. Owing to network function virtualization (NFV) technology, computing services can be implemented as service chains, which are composed of ordered virtual network functions. We then design a service chain-driven CIN architecture with “three layers and three domains” characteristics, which consists of generalized service layer, mapping adaption layer and converged network layer. The designed architecture enables the adaptability for users, flexibility for CIN, and profitability for providers. Furthermore, we outline key technologies such as measurement and modeling of multi-dimensional heterogeneous resources, fine-grained multi-dimensional awareness, multi-dimensional identification networking and heterogeneous resources allocation. In addition, we formulate the heterogeneous resource joint optimization problem and verify the effectiveness of the designed scheme. Furthermore, we explore open issues based on our review and indicate potential research trends of the new computing paradigm.

Modern enterprises increasingly rely on distributed groups, subscription-based communities, and collaborative teams, which introduce significant challenges in managing payments, memberships, and operational analytics across disparate systems. Traditional approaches using spreadsheets, manual reminders, and fragmented payment links result in data silos, reconciliation delays, and lack of actionable insights. This paper presents the design and implementation of the Payment Portal Management System (PPMS)—a mobile-first, enterprise-grade platform that unifies group portal creation, payment tracking, activity logging, and analytics within a single, secure ecosystem. The system employs modular service layers for activity logging and analytics, real-time dashboards, and intuitive mobile interfaces developed using Flutter and Dart. Our implementation demonstrates a 40% reduction in administrative effort compared to manual methods while providing real-time balance tracking, structured event logging, and analytics-driven dashboards that significantly improve transparency and operational efficiency.

Recent advances integrating Artificial Intelligence (AI) and Virtual Reality (VR) are accelerating the expansion of intelligent services in smart media environments. In fashion retail, there is a growing need to enhance VR stores and Visual Merchandising (VM) to compensate for the spatial experience that online shopping cannot fully provide. This study aims to design an AI-based VR store VM system from a smart media service perspective and to propose a system architecture for implementing intelligent VM. Based on a review of prior research, the study conceptualizes VM in VR environments and derives key design principles, and proposes a three-layer architecture consisting of a user data collection and processing layer, an AI agent–based VM decision-making layer, and a VR store VM service layer. The findings suggest that the proposed system can serve as a design framework for intelligent VM services grounded in user behavioral data, and provide a foundation for future empirical studies using KPI-based evaluation such as gaze, navigation paths, and purchase intention.

Service-Oriented Architecture (SOA) structures applications into collections of modular, independent, and reusable services. We propose an SOA-based intelligent service agent framework for building AI applications that decomposes complex tasks into independent functional units. In the framework, the agent operates as an intelligent executor that dynamically orchestrates and invokes diverse services and tools to achieve its goals. The agent is exposed as a self-contained service with a well-defined API, allowing external applications to invoke it directly. By instrumenting requests and responses at both the service and agent layers, the framework enables tracing of the agent’s capabilities, performance, and decision-making. We present the design of an operational scheme for the agent with DID handling, verifiable credentials (VC), and verifiable presentations (VP). Each of the agents collaborates on a shared workspace based on blackboard to handle tasks to reach a goal. Finally, we demonstrate its feasibility through a proof-of-concept (PoC) for Agentic AI service architecture. This proof-of-concept, structured across Phase 1 (discovery, verification, and scoped authorization) and Phase 2 (problem posting and blackboard-mediated collaboration), demonstrates that DID-backed credentialing can securely support multi-agent execution under a least-privilege operational model.

The article examines global submarine data cable projects from 1989 to 2028 (N=804) to explore the evolving political economy of digital media and communication infrastructures. Focusing on the growing number of new cable projects and their ownership, we identify several phases of network expansion, which correspond with different types of market dominance and ownership constellations. In the early phases of submarine cable buildout, the main market actors were national and regional telecommunications corporations, often collaborating in large consortia, whereas the later phases see a growing concentration of ownership around a set of global market actors, often establishing their own networks to support their extensive platform and data businesses. In interpreting this development, we seek inspiration from previous historical analyses of large-scale technological systems and argue that the internet has reached a momentum where its built-in affordances lead to a fundamental shift in institutional power. We end by discussing the implications of this historical transformation, emphasizing that the new generations of cable systems amplify an already walled-off ecosystem, where access to the superior infrastructure is conditioned by the usage of proprietary services, leading to a reintegration of the network and service layer that the internet originally separated.

With the advancement of digitization in the era of Industry 4.0 (I4.0), highly automated, semi-autonomous, and fully autonomous systems are emerging. Within this context, multi-agent systems (MAS) offer a promising approach for automating tasks and processes based on autonomous agents that work together in an overall system to increase the degree of system autonomy stepwise in a modular and flexible way. A critical research challenge is determining how these agents can collaboratively engage with both other agents and human operators to facilitate the gradual transition from automated to fully autonomous industrial systems. To close transparency and connectivity gaps, this study contributes with a framework for the collaboration of agents and humans in increasingly autonomous MAS based on a Digital Twin (DT). The framework specifies a standards-based data model for MAS representation and proposes to introduce a DT infrastructure as a service layer for system coordination, supervision, and interaction. To demonstrate the feasibility and assess the quality of the framework, it is implemented and evaluated in a case study in a real-world industrial scenario. Although additional long-term evaluations across different contexts are needed, the assessment of functional completeness and selected quality attributes show that the proposed framework provides a solid technical foundation that facilitates a transparent and seamless collaboration between agents and humans within increasingly autonomous industrial MAS.