Search Results (5,696)

Vehicular communication networks demand highly efficient and accurate channel estimation to ensure reliable data exchange in high mobility scenarios. The IEEE 802.11p standard is widely regarded as the foundation of the Vehicle-to-Vehicle (V2V) communication channel; however, it is constrained by limited pilot resources and a fixed pilot structure, which degrade the performance and effectiveness of traditional estimation techniques, particularly in dynamic environments. Recent advances in deep learning offer significant potential for addressing these issues by improving estimation accuracy and modelling complex channel dynamics. Though deep learning-based methods introduce trade-offs in computational complexity and accuracy, these are crucial constraints in latency-sensitive V2V scenarios. This article presents a comprehensive review of deep learning-based channel estimation techniques, analysing methods for the IEEE 802.11p standard and critically examining their limitations in both classical and deep learning-based approaches. Additionally, the article highlights improvements introduced by IEEE 802.11bd, which features an enhanced pilot structure and advanced modulation schemes, providing a more robust framework for adaptive, efficient channel estimation. By identifying future research pathways that balance delay, complexity, and accuracy, an intelligent and effective transportation system can be established. Full article

In the context of rapid urban tourism expansion and the growing emphasis on equitable and sustainable transport development, understanding how transport systems support different types of attractions has become increasingly important. This study investigates how attraction hierarchy and functional type interact with public transport accessibility to shape urban tourism patterns and equity. Whereas prior work emphasizes objective metrics, the alignment between perceived accessibility and actual transport conditions remains understudied. Using Wuhan’s A-rated and popular unrated attractions as a case, we have developed an innovative “ objective–perceived coupling framework that integrates GIS network analysis, travel cost matrix, non-parametric testing, and online comment text mining methods to examine how scenic spot levels (A-level and unrated popular scenic spots) and functional types interact with the public transportation system from both objective and perceptual dimensions. Results show: (1) A-rated attractions cluster in suburbs with low accessibility, while unrated sites concentrate centrally with high rail-bus connectivity, revealing a “high-grade–low-accessibility” mismatch. (2) Accessibility varies by type: natural sites are lowest, cultural/leisure venues intermediate, and comprehensive sites highest due to multimodal hub proximity. (3) Sentiment and topic analyses based on transport-related review content suggest that some A-rated attractions receive less favorable evaluations of access conditions (e.g., transfers, waiting, last-mile walking, wayfinding, and parking), whereas many popular unrated sites are evaluated more positively in these transport-specific aspects. (4) Quadrant analysis shows many highly rated attractions fall into a “low objective–low perceived” disadvantage, while most unrated ones exhibit strong objective–perceived coupling. These findings underscore structural imbalances among administrative grading, attraction function, and transit provision, offering evidence for optimizing public transport service to tourist attractions. They help optimize the spatial structure of urban tourism, improve resource allocation efficiency, guide differentiated scenic spot development strategies, and promote sustainable and experience-oriented urban tourism governance. Full article

Growing global energy demand and concerns over climate change and fossil fuel depletion have increased interest in sustainable bioproducts such as ethanol. Unlike first-generation (1G) ethanol derived from food crops (e.g., corn), second-generation (2G) ethanol is produced from lignocellulosic biomass, an abundant non-food resource that addresses key sustainability concerns. Consolidated bioprocessing (CBP) integrates enzyme production, hydrolysis, and fermentation into a single step, using either microbial consortia or engineered microorganisms, thereby simplifying the process and potentially reducing costs compared with separate hydrolysis and fermentation (SHF) and simultaneous saccharification and fermentation (SSF). However, CBP systems are complex due to dynamic interactions among microbial communities, metabolic pathways, and process conditions. Addressing this complexity requires modeling approaches that capture nonlinear relationships and support robust process optimization. Machine learning (ML)-based models offer data-driven tools to represent complex bioprocess dynamics, improve predictive accuracy, and optimize bioproduct formation, thereby supporting progress toward commercial viability. Although CBP can be applied to a range of bioproducts, this review primarily focuses on lignocellulosic ethanol and closely related biofuels. The review provides a comprehensive overview of key CBP processes, the current state of CBP modeling, major limitations, and the emerging role of ML in addressing modeling challenges. It summarizes recent modeling techniques for CBP, including polynomial models and response surface methodologies, and discusses regression and neural network approaches in detail. Both first-principles and data-driven modeling strategies are considered, highlighting advances that can improve the scalability and efficiency of CBP for bioproduction. Overall, this review offers perspectives on modeling-enabled pathways for utilizing low-cost lignocellulosic biomass in sustainable bioprocessing. Full article

Crowd mass disasters occur over a relatively short time, and rescue operations in disasters, such as earthquakes, are challenging because of people’s behavior, type, or location. Therefore, it is essential to devise means and methods to manage such problems to minimize the consequences as much as possible. During disasters, rescue operations should be conducted in a timely conducted to save people’s lives. Otherwise, losses and consequences are severe, and if there are no proper rescuing operation models, the situation worsens, and the consequences are devastating. In particular, the allocation and coordination of limited rescue resources have a critical impact on response times and the number of lives saved. This paper aims to develop an Agent-Based Simulation (ABS) model for rescuing operations in crowd-mass disasters with six main intelligent agents. The proposed model explicitly represents the interactions among victims, rescuers, command-and-control entities, transportation assets, road networks, and affected infrastructure within a GIS-based urban environment. The developed model is based on an enhanced approach to improve rescue agents’ tasks allocation operations that enable modeling and simulation to make critical decisions for people to be rescued in a crowded mass disaster. Our task-allocation mechanism incorporates dynamic accessibility of roads, time-dependent rescue capacity, and context-aware prioritization of victims. Three related task-allocation strategies from the literature are used as baselines under identical scenarios, and performance is compared in terms of average rescue time and number of rescued victims. Results show that the proposed model achieves more efficient and robust rescue operations in most simulated experiments. Full article

This study examines how strongly demand-load prediction and adaptive load control in photovoltaic heating systems rely on computationally intensive artificial neural network (ANN) models. To streamline the computational workflow and reduce runtime resource requirements, we propose an ANN load-prediction-and-validation algorithm coupled with a corresponding offline control strategy. By optimizing the algorithmic structure and shifting heavy computations away from online execution, the proposed method substantially lowers the operational computational burden while preserving predictive accuracy, enabling efficient real-time load prediction and adaptive control. Based on a modelling study of a monocrystalline PV string comprising two 330 W modules connected in series, the proposed simplified prediction method produced annual cumulative energy outputs of 139.9, 391.2, 320.2, 251.4, and 154.1 kW·h across the five irradiance intervals [200, 400), [400, 600), [600, 800), [800, 1000), and [1000, ∞), respectively. Compared with a conventional artificial neural network (ANN)-based prediction approach, the corresponding deviations were 1.1%, −0.1%, 0.0%, 0.1%, and −0.4%, the total annual cumulative energy outputs across all intervals was 1256.7 kW·h with a mean deviation of −0.07%. Moreover, the simplified load-control strategy required only 3.57% of the computational resources consumed by the conventional ANN method. In addition, the method rapidly reallocates computational resources in response to changes in real-time input data, thereby minimizing redundant computation. Overall, the results demonstrate that the proposed framework markedly reduces computational complexity without sacrificing accuracy, providing an effective alternative to traditional ANN-based solutions and facilitating the practical deployment of photovoltaic heating systems. Full article

Ideological and political education (IPE) is a cornerstone of higher education in China. As IPE-related book collections expand rapidly, university libraries face a growing challenge of information overload, which hinders the accurate characterization of student reading preferences and the efficient matching of resources to demand. To address these issues, this study proposes GDNN, a practical hybrid recommendation system designed for both warm-start and cold-start scenarios. For warm-start users with historical borrowing records, we develop the PPSM-GCN framework. This framework enhances the classical graph convolutional collaborative filtering model LightGCN by integrating a novel potential positive sample mining (PPSM) strategy, which effectively mitigates data sparsity and improves the modeling of latent interests. For cold-start users without interaction history, we introduce an embedding and MLP architecture. This deep neural network learns implicit reader–book associations from reader attributes and book metadata, enabling personalized recommendations even in the absence of historical data. Experimental results demonstrate that PPSM-GCN and the embedding and MLP method achieve significant performance gains in their respective scenarios. This research provides both technical support and practical insights for the precise delivery of IPE resources and the overall enhancement of educational effectiveness in higher education. Full article

Unmanned surface vehicles (USVs) increasingly rely on vision-based perception for safe navigation and maritime surveillance, while onboard computing is constrained by strict size, weight, and power (SWaP) budgets. Although deep convolutional neural networks (CNNs) offer strong recognition performance, their computational and memory requirements pose significant challenges for deployment on low-cost embedded platforms. This paper presents a hardware–software co-design architecture and deployment study for CNN acceleration on a heterogeneous ARM–FPGA system, targeting energy-efficient near-sensor processing for embedded maritime applications. The proposed approach exploits a fully streaming hardware architecture in the FPGA fabric, based on line-buffered convolutions and AXI-Stream dataflow, while the ARM processing system is responsible for lightweight configuration, scheduling, and data movement. The architecture was evaluated using representative CNN models trained on a maritime ship dataset. Our experimental results on a Zynq-7020 system-on-chip demonstrate that the proposed co-design strategy achieves a balanced trade-off between throughput, resource utilisation, and power consumption under tight embedded constraints, highlighting its suitability as a practical building block for onboard perception in USVs. Full article

The enormously growing demand for seafood has resulted in the over-exploitation of marine resources, pushing certain species to the brink of extinction. Overfishing is one of the main issues in sustainable marine development. To support marine resource protection and sustainable fishing, this study proposes advanced fish classification techniques using state-of-the-art machine learning (ML). Specifically, the proposed method enables the precise identification of protected fish species, among other features. In this paper, we present a system-level optimization of the MobileNet architecture, termed M-MobileNet, designed to operate efficiently on resource-limited hardware environments. Our classifier is constructed by a refined modification of the well-known MobileNet neural network, resulting in a reduction of parameters. Furthermore, we have collected, organized, and compiled an original and comprehensive labeled dataset of 37,462 images of fish native to the Indonesian archipelago. The proposed model is trained on this dataset to classify images of captured fish and accurately identify their respective species. Furthermore, the system provides recommendations regarding the consumability of the catch. Compared to the MobileNet deep neural network structure, our model utilizes only 50% of the top-layer parameters, with approximately 42% GTX 860M utility. This configuration results in achieving up to 97% accuracy of classification. Considering the constrained computing capacity prevalent on many fishing vessels, our proposed model offers a practical solution for on-site fish classification. Moreover, synchronized implementation of the proposed model across multiple vessels can provide valuable insights into the movement and location of various fish species. Full article

As a critical step in industrial quality control, surface defect detection in aluminum materials remains challenging for minor defects despite advances in deep learning. To address this, this paper proposes an enhanced YOLOv8-based model, BFI-YOLO, that incorporates a Bidirectional Multi-scale Residual Network. Specifically, we design a Bidirectional Multi-scale Feature Pyramid Network (BM-FPN) based on BiFPN to strengthen cross-scale feature fusion. The parameter-free SimAM attention module is embedded to enhance subtle defect responses while suppressing background texture interference, without introducing additional computational overhead.Furthermore, we develop a Multi-scale Residual Convolution (MSRConv) module to capture defects of varying sizes on aluminum surfaces comprehensively. MSRConv utilizes multi-scale convolutional kernels to adapt to cross-scale defect features and retains shallow details via residual connections, thereby strengthening the model’s representation of fine defects. Extensive experiments on the public TAPSDD dataset show that BFI-YOLO achieves a precision of 91.3%, a recall of 89.8%, and mAP@0.5 of 92.1%, with only 1.8 M parameters. Compared to the baseline, BFI-YOLO reduces parameters by 40% while increasing mAP@0.5 by 4.2%, effectively balancing detection accuracy and lightweight performance. Optimized for resource-constrained industrial platforms such as embedded systems and mobile robots, BFI-YOLO meets real-time monitoring requirements while achieving competitive detection accuracy, providing an efficient and practical solution for metal surface defect detection. Full article

The sustainability of experience-intensive wellness tourism services increasingly depends on managers’ ability to understand heterogeneous and implicit tourist preferences that are rarely captured through traditional survey-based approaches. In the context of Korean Templestay tourism, this study develops a data-driven decision-support framework that leverages large-scale unstructured review data to address managerial challenges such as choice overload, inefficient resource allocation, and cold-start conditions. Using 74,015 user-generated reviews collected between 2020 and 2024, the framework integrates Optical Character Recognition (OCR) to extract image-embedded text, achieving a validated character-level accuracy of 96.8%. In addition, a weak supervision strategy is applied to identify latent tourist preferences in a cost-efficient and scalable manner. Preference classification is conducted using Random Forest models combined with SMOTE, followed by clustering and user-based collaborative filtering to support personalized recommendations. The findings indicate that the Templestay market is better understood as an interconnected preference network rather than a set of mutually exclusive segments. Across user groups, “rest” emerges as a shared foundational value, while differentiated sub-preferences coexist within the network. The proposed framework successfully generates recommendations for all users in the dataset, demonstrating strong applicability for mitigating cold-start risks and supporting adaptive and sustainable program design. Full article

Cataloging task resource scheduling is a key technology for the efficient utilization of ground-based radar networks and for supporting space situational awareness. This problem is highly challenging due to the large scale of tasks, strict time window constraints, and complex resource-task mapping relationships. It requires algorithms to effectively balance multiple conflicting optimization objectives within a huge and sparse solution space, placing extremely high demands on the convergence, diversity maintenance, and computational efficiency of the algorithms. This paper presents a systematic review of the latest research progress in cataloging resource scheduling methods. First, commonly used optimization objectives and constraint conditions in this field are outlined, and two key subproblems—priority modeling and conflict resolution—are analyzed in depth. Subsequently, following the trajectory of technological evolution, the application paradigms, performance characteristics, and limitations of mainstream algorithms are reviewed. Given the inherent multi-objective optimization nature of the problem, the advantages and challenges of multi-objective optimization algorithms are discussed. Finally, based on a unified problem context, the performance and operational boundaries of existing algorithms are compared and analyzed, and future research directions and core challenges in the field are presented. Full article

In the power grid scheduling process, load forecasting serves as the foundation for ensuring stability and economic dispatch. It not only optimizes resource allocation but also strengthens the system’s productivity and stability, helps prevent potential risks, and ensures the reliability and safety of power supply. Therefore, a predictive soft actor–critic network (PreSAC-Net) algorithm is proposed, which aims to reduce grid operating costs and enhance system stability through an enhanced load forecasting model and an optimized scheduling strategy. First, the load forecasting is performed using a sequential feature fusion model with gated recurrent attention and diffusion (SeqFusion-GRAD), which integrates gated recurrent units (GRU), attention mechanisms, and generative diffusion models to strengthen time-series modeling and accurately predict household electricity loads. Second, a multidimensional data fusion technique incorporates meteorological and other relevant factors into household load data, improving the forecast accuracy and robustness. Furthermore, the scheduling optimization is conducted with the soft actor–critic (SAC) algorithm, which explores scheduling schemes to minimize cost under multiple constraints. The integrated approach not only balances the electricity supply and demand effectively but also supports the sustainable development of intelligent grids. Based on the experimental results, the proposed method significantly enhances power system operational efficiency and stability. Full article

Accurate and efficient modeling of AlGaN/GaN HEMTs is essential for the design of next-generation power electronics. This study introduces a hybrid Auxiliary Classifier Generative Adversarial Network (ACGAN)–mixup data augmentation framework to enhance deep neural network application in AlGaN/GaN high-electron-mobility transistor modeling with limited data. Based on only 20 distinctive devices, ACGAN uses technology computer-aided design (TCAD)-calibrated data to generate high-quality synthetic drain current ($I_{ds}$) under various electronic bias conditions. The quality of the generated data is validated via Jensen–Shannon divergence with an average of 0.0341. A one-dimensional convolutional neural network (1D-CNN) predictive model is trained on augmented data and achieves stable convergence, with a mean absolute error of 0.002 A/mm for the off-state $I_{ds}$ and 0.052 A/mm for the linear region. It also shows improved robustness over the model trained on original non-augmented data. The proposed approach offers a low-cost alternative to resource-intensive TCAD simulations, enabling accurate device modeling with limited data. Full article

Accurate interpretation of chest radiographs is central to the early diagnosis and management of pulmonary disorders. This study introduces an explainable deep learning framework that integrates biomedical signal fidelity analysis with transfer learning to enhance diagnostic reliability and transparency. Using the publicly available COVID-19 Radiography Dataset (21,165 chest X-ray images across four classes: COVID-19, Viral Pneumonia, Lung Opacity, and Normal), three architectures, namely baseline Convolutional Neural Network (CNN), ResNet-50, and EfficientNetB3, were trained and evaluated under varied class-balancing and hyperparameter configurations. Signal preservation was quantitatively verified using the Structural Similarity Index Measure (SSIM = 0.93 ± 0.02), ensuring that preprocessing retained key diagnostic features. Among all models, ResNet-50 achieved the highest classification accuracy (93.7%) and macro-AUC = 0.97 (class-balanced), whereas EfficientNetB3 demonstrated superior generalization with reduced parameter overhead. Gradient-weighted Class Activation Mapping (Grad-CAM) visualizations confirmed anatomically coherent activations aligned with pathological lung regions, substantiating clinical interpretability. The integration of signal fidelity metrics with explainable deep learning presents a reproducible and computationally efficient framework for medical image analysis. These findings highlight the potential of signal-aware transfer learning to support reliable, transparent, and resource-efficient diagnostic decision-making in radiology and other imaging-based medical domains. Full article

Background/Objectives: Virtual histological staining offers a rapid, cost-effective alternative to physical reprocessing but faces challenges related to spatial misalignment and staining heterogeneity between Hematoxylin and Eosin (H&E) and Masson’s Trichrome (MT) domains. This study develops a robust framework for H&E-to-MT virtual staining to enable accurate fibrosis assessment without additional tissue consumption. Methods: We propose a transformer-based generative adversarial network (TbGAN) supported by a multi-stage alignment pipeline (SIFT (scale-invariant feature transform) coarse alignment, ORB/homography patch registration, and B-spline free-form deformation) and a weighted fusion mechanism combining four configuration outputs (O/10/3, O/3/10, R/10/3, and R/3/10). The framework was validated on 27 whole-slide images (>100,000 aligned patches) through 24 independent experiments. Results: The fused approach achieved state-of-the-art performance: MI = 0.9815 ± 0.0934, SSIM = 0.7474 ± 0.0597, NCC = 0.9320 ± 0.0220, and CS = 0.9946 ± 0.0014. Statistical analysis confirmed enhanced stability through narrower interquartile ranges, fewer outliers, and tighter 95% confidence intervals compared to individual configurations. Qualitative assessment demonstrated preserved collagen morphology critical for fibrosis staging. Conclusions: Our framework provides a reliable, IRB-compliant solution for virtual MT staining that maintains high structural fidelity suitable for diagnostic support. It enables resource-efficient fibrosis quantification and supports integration into clinical digital pathology workflows without patient-specific recalibration. Full article