Search Results (3,052)

Traffic state estimation (TSE) is a core task in intelligent transportation systems (ITSs) that seeks to infer key operational parameters—such as speed, flow, and density—from limited observational data. Existing methods often face challenges in practical deployment, including limited estimation accuracy, insufficient physical consistency, and weak generalization capability. To address these issues, this paper proposes a hybrid estimation framework that integrates multi-agent reinforcement learning (MARL) with the Lighthill–Whitham–Richards (LWR) traffic flow model. In this framework, each roadside detector is modeled as an agent that adaptively learns fundamental diagram (FD) parameters—the free-flow speed and jam density—by fusing local detector measurements with global CAV trajectory sequences via an interactive attention mechanism. The learned parameters are then passed to an LWR solver to perform sequential (rolling) prediction of traffic states across the entire road segment. We design a reward function that jointly penalizes estimation error and violations of physical constraints, enabling the agents to learn accurate and physically consistent dynamic traffic state estimates through interaction with the physics-based LWR environment. Experiments on simulated and real-world datasets demonstrate that the proposed method outperforms existing models in estimation accuracy, real-time performance, and cross-scenario generalization. It faithfully reproduces dynamic traffic phenomena, such as shockwaves and queue waves, demonstrating robustness and practical potential for deployment in complex traffic environments. Full article

We present an advanced statistical framework for estimating the relative intensity of astrophysical event distributions (e.g., Gamma-Ray Bursts, GRBs) on the sky tofacilitate population studies and large-scale structure analysis. In contrast to the traditional approach based on the ratio of Kernel Density Estimation (KDE), which is characterized by numerical instability and bandwidth sensitivity, this work applies a logistic regression embedded in a Bayesian framework to directly model selection effects. It reformulates the problem as a logistic regression task within a Generalized Additive Model (GAM) framework, utilizing isotropic Splines on the Sphere (SOS) to map the conditional probability of redshift measurement. The model complexity and smoothness are objectively optimized using Restricted Maximum Likelihood (REML) and the Akaike Information Criterion (AIC), ensuring a data-driven bias-variance trade-off. We benchmark this approach against an Adaptive Kernel Density Estimator (AKDE) using von Mises–Fisher kernels and Abramson’s square root law. The comparative analysis reveals strong statistical evidence in favor of this Preconditioned (Precon) Estimator, yielding a log-likelihood improvement of $\Delta \mathcal{L}\approx 74.3$ (Bayes factor $>{10}^{30}$) over the adaptive method. We show that this Precon Estimator acts as a spectral bandwidth extender, effectively decoupling the wideband exposure map from the narrowband selection efficiency. This provides a tool for cosmologists to recover high-frequency structural features—such as the sharp cutoffs—that are mathematically irresolvable by direct density estimators due to the bandwidth limitation inherent in sparse samples. The methodology ensures that reconstructions of the cosmic web are stable against Poisson noise and consistent with observational constraints. Full article

The advent of generative AI (GenAI) and its growing use in education has sparked a renewed wave of school digital transformation. School principals are pivotal in advancing and shaping school digital transformation, yet little is known about how they understand and lead digital transformation in the age of GenAI, particularly within China’s complex educational system. This study employed Q methodology to identify the perceptions and leadership styles of Chinese K–12 school principals toward school digital transformation in the age of GenAI. An analysis of a 30-item Q set with a P sample of 23 principals revealed four leadership types: Cautious Observation–Technological Gatekeeping Leadership, Moderate Ambition–Culturally Transformative Leadership, Moderate Ambition–Emotionally Empowering Leadership, and High Aspiration–Strategy-Driven Leadership. Overall, principals’ stances on GenAI formed a continuum, ranging from cautious observation and skeptical optimism to active embrace. These perceptions and leadership styles were shaped by Confucian cultural values, a flexible central–local governance arrangement, and parents’ high expectations for students’ academic achievement. Furthermore, structural constraints in resource provision further heightened principals’ reliance on maintaining guanxi-based relationships. This study enhances the understanding of the diversity of principals’ leadership practices worldwide and offers actionable insights for governments and principals to more effectively advance AI-enabled school digital transformation. Full article

As critical platforms for long-endurance ocean exploration, unmanned underwater vehicles (AUVs) play an increasingly vital role in marine surveying and resident observation. However, in extreme deep-sea environments, their energy systems face severe constraints imposed by hydrostatic pressure and thermodynamic conflicts within confined spaces. Therefore, developing energy technologies with high energy density, intrinsic safety, and high-pressure adaptability is of paramount importance. This paper provides a comprehensive review of the multi-physics coupling issues in deep-sea energy systems and the research progress of current mainstream deep-sea energy technologies. Based on energy sources and conversion principles, existing technological paths are categorized into four classes, with a detailed assessment of their performance and bottlenecks in deep-sea environments. Finally, the paper outlines key future development directions for deep-sea energy systems to provide reference for subsequent research. Full article

To investigate the mechanical properties, energy evolution, and failure behavior of coal–rock composite structures under mining disturbances, a mining-induced stress path was designed based on the actual stress evolution ahead of a mining face. Triaxial tests were carried out under these stress conditions on coal–rock composite samples at various confining pressures, supplemented by conventional triaxial compression tests for comparison. The results show that the coal–rock composite samples exhibited marked brittle failure under mining-induced stress, with no sign of the brittle–ductile transition observed in conventional triaxial tests as the confining pressure increased. Using dual circumferential extensometers, it was found that the circumferential deformation of the coal and rock was initially governed by their intrinsic mechanical properties and later controlled by crack propagation. At higher confining pressures, the growth rate of circumferential strain at failure increased significantly, indicating that deeper excavations result in more severe unloading-induced failure. Comparative analysis revealed that the coal component had a higher elastic energy density and faster energy accumulation and release rates than the rock, identifying coal as the dominant medium for elastic energy storage and release within the composite samples. Furthermore, at peak stress in mining-induced stress tests, the coal showed less circumferential deformation than in conventional tests, while the rock exhibited the opposite trend, confirming the presence of a bonding constraint effect at the coal–rock interface. These findings enhance our understanding of the mechanical behaviors and failure mechanisms of coal–rock composites under mining disturbances, thus providing practical guidance for ensuring safety and efficiency in deep coal mining. Full article

Monitoring soil CO₂ is essential for accurately quantifying the sources and sinks of atmospheric greenhouse gases and for providing carbon emission reduction strategies. However, the limited portability and high cost of conventional soil CO₂ monitoring equipment have severely restricted large-scale and long-term field observations. To address these constraints, this study has successfully designed and fabricated a portable and low-cost soil respiration system (SRS) based on non-dispersive infrared (NDIR) sensor technology and Long-range radio (LoRa) wireless communication. The SRS enables multi-point synchronous measurements and remote data transmission. Its reliability was rigorously evaluated through both simulated and field comparative experiments against the LI-8100A. The results demonstrated a high level of agreement between the measurements of the SRS and the LI-8100A, with the coefficients of determination (R²) of 0.996 and 0.997, respectively, for the simulation and field experiments, with the corresponding root mean square error (RMSE) of 0.090 and 0.089 μmol·m⁻²·s⁻¹. The Bland–Altman analysis further confirmed the consistency between the two systems, with over 95% of the data points falling within the acceptable limits of agreement. These findings indicate that the self-developed SRS substantially reduces costs while maintaining reliable measurement accuracy. With its wireless transmission and multi-point deployment capabilities, the SRS offered an efficient and practical solution for addressing the challenges of monitoring spatial heterogeneity of soil respiration, demonstrating considerable potential for broader application in CO₂ flux monitoring research. Full article

Tax incentives play a crucial role in enhancing firm dynamism and aiding a nation in becoming a significant trade power. Drawing on data from the Annual Survey of Industrial Firms Database and the Chinese Customs Database for the period 2010 to 2013, this study employs a difference-in-differences approach to assess the impact of China’s transition from a business tax to a value-added tax (RBTVAT) on the export diversification of manufacturing firms. The findings indicate that the tax reform significantly decreases the number of export categories, increases export value, and elevates the export unit price for manufacturing firms. Specifically, by promoting specialized production and encouraging the manufacture of products with higher export tax rebate rates, the reforms have led firms to narrow their range of export categories. This effect is particularly pronounced among firms experiencing higher financing constraints, lower profitability, weaker innovation capabilities, and larger size. Furthermore, a consistent negative impact is observed for both state-owned and non-state-owned enterprises. These results provide novel insights and empirical evidence for understanding the relationship between tax reform and export diversification. Full article

In precision oncology research, achieving joint modeling of tumor grading and treatment response, together with interpretable mechanism analysis, based on multimodal medical imaging and clinical data remains a challenging and critical problem. From a sensing perspective, these imaging and clinical data can be regarded as heterogeneous sensor-derived signals acquired by medical imaging sensors and clinical monitoring systems, providing continuous and structured observations of tumor characteristics and patient states. Existing approaches typically rely on invasive pathological grading, while grading prediction and treatment response modeling are often conducted independently. Moreover, multimodal fusion procedures generally lack explicit structural constraints, which limits their practical utility in clinical decision-making. To address these issues, a grade-guided multimodal collaborative modeling framework was proposed. Built upon mature deep learning models, including 3D ResNet-18, MLP, and CNN–Transformer, tumor grading was incorporated as a weakly supervised prior into the processes of multimodal feature fusion and treatment response modeling, thereby enabling an integrated solution for non-invasive grading prediction, treatment response subtype discovery, and intrinsic mechanism interpretation. Through a grade-guided feature fusion mechanism, discriminative information that is highly correlated with tumor malignancy and treatment sensitivity is emphasized in the multimodal joint representation, while irrelevant features are suppressed to prevent interference with model learning. Within a unified framework, grading prediction and grade-conditioned treatment response modeling are jointly realized. Experimental results on real-world clinical datasets demonstrate that the proposed method achieved an accuracy of $84.6\%$ and a kappa coefficient of $0.81$ in the tumor-grading prediction task, indicating a high level of consistency with pathological grading. In the treatment response prediction task, the proposed model attained an AUC of $0.85$, a precision of $0.81$, and a recall of $0.79$, significantly outperforming single-modality models, conventional early-fusion models, and multimodal CNN–Transformer models without grading constraints. In addition, treatment-sensitive and treatment-resistant subtypes identified under grading conditions exhibited stable and significant stratification differences in clustering consistency and survival analysis, validating the potential value of the proposed approach for clinical risk assessment and individualized treatment decision-making. Full article

The lack of clinical data for chronic kidney disease (CKD) prediction frequently results in model overfitting and inadequate generalization to novel samples. This research mitigates this constraint by utilizing a Conditional Tabular Generative Adversarial Network (CTGAN) to enhance a constrained CKD dataset sourced from the University of California, Irvine (UCI) Machine Learning Repository. The CTGAN model was trained to produce realistic synthetic samples that preserve the statistical and feature distributions of the original dataset. Multiple machine learning models, such as AdaBoost, Random Forest, Gradient Boosting, and K-Nearest Neighbors (KNN), were assessed on both the original and enhanced datasets with incrementally increasing degrees of synthetic data dilution. AdaBoost attained 100% accuracy on the original dataset, signifying considerable overfitting; however, the model exhibited enhanced generalization and stability with the CTGAN-augmented data. The occurrence of 100% test accuracy in several models should not be interpreted as realistic clinical performance. Instead, it reflects the limited size, clean structure, and highly separable feature distributions of the UCI CKD dataset. Similar behavior has been reported in multiple previous studies using this dataset. Such perfect accuracy is a strong indication of overfitting and limited generalizability, rather than feature or label leakage. This observation directly motivates the need for controlled data augmentation to introduce variability and improve model robustness. The dataset with the greatest dilution, comprising 2000 synthetic cases, attained a test accuracy of 95.27% utilizing a stochastic gradient boosting approach. Ensemble learning techniques, particularly gradient boosting and random forest, regularly surpassed conventional models like KNN in terms of predicted accuracy and resilience. The results demonstrate that CTGAN-based data augmentation introduces critical variability, diminishes model bias, and serves as an effective regularization technique. This method provides a viable alternative for reducing overfitting and improving predictive modeling accuracy in data-deficient medical fields, such as chronic kidney disease diagnosis. Full article

With the development of the Internet of Things (IoT) and smart industry, the demand for high-precision indoor positioning is becoming increasingly urgent. Ultra-ideband (UWB) technology has become a research hotspot due to its centimeter-level ranging accuracy, good penetration, and high multipath resolution. However, in complex environments, it still faces problems such as high cost of anchor node layout, gross errors in observation data, and difficulty in eliminating systematic errors such as electronic time delay. To address the aforementioned problems, this paper proposes a comprehensive UWB indoor positioning scheme. By constructing virtual reference stations to enhance the observation network, the geometric structure is optimized and the dependence on physical anchors is reduced. Combined with a gross error elimination method under short-baseline constraints and a double-difference positioning model including virtual observations, it systematically suppresses systematic errors such as electronic delay. Additionally, a quality control strategy with velocity constraints is introduced to improve trajectory smoothness and reliability. Static experimental results show that the proposed double-difference model can effectively eliminate systematic errors. For example, the positioning deviation in the Xdirection is reduced from approximately 2.88 cm to 0.84 cm, while the positioning accuracy in the Ydirection slightly decreases. Overall, the positioning accuracy is improved. The gross error elimination method achieves an identification efficiency of over 85% and an accuracy of higher than 99%, providing high-quality observation data for subsequent calculations. Dynamic experimental results show that the positioning trajectory after geometric enhancement of virtual reference stations and velocity-constrained quality control is highly consistent with the reference trajectory, with significantly improved trajectory smoothness and reliability. In summary, this study constructs a complete technical chain from data preprocessing to result quality control, effectively improving the accuracy and robustness of UWB positioning in complex indoor environments, and exhibits promising engineering application potential. Full article

The problem was recently reported that the far-zone electromagnetic momentum of light produced by scattering on a spatially anisotropic random medium can be the same at every azimuthal angle of scattering. Here, we extend the analysis to focus on the possibility of producing a rotationally symmetric spectral degree of coherence (SDOC) generated by scattering by an anisotropic process. The necessary and sufficient conditions for producing such a SDOC in the far zone are derived when a polychromatic electromagnetic plane wave is scattered by an anisotropic Gaussian Schell-model medium. We find that, unlike the generation of a rotationally symmetric momentum flow, it is not enough to simply restrict the structural characteristics of the medium and the incident light source to achieve a SDOC with rotational symmetry. An additional and essential requirement is that the azimuthal angles of scattering corresponding to the two observation points of the SDOC must be constrained to be equal. Only when all these constraints are satisfied simultaneously can a rotationally symmetric electromagnetic SDOC generated by scattering by an anisotropic process be realized. In addition, we find that although the medium parameter conditions for generating a rotationally symmetric SDOC and a rotationally symmetric momentum flow are completely different, it remains possible that the SDOC and the momentum flow produced by a spatially anisotropic medium can still simultaneously exhibit rotational symmetry, provided that the distribution of the correlation function of the scattering potential of the medium is isotropic in the plane perpendicular to the incident direction. Our results not only contribute to a deeper understanding of the far-field distribution of light scattering on an anisotropic scatterer, but also have potential applications in light-field manipulation and in the inverse scattering problem. Full article

To address the challenges of insufficient efficiency and accuracy in traditional detection models caused by minute target sizes, low signal-to-noise ratios (SNRs), and feature volatility in Satellite Laser Ranging (SLR) images, this paper proposes an efficient, lightweight, and high-precision detection model. The core motivation of this study is to fundamentally enhance the model’s capabilities in feature extraction, fusion, and localization for minute and blurred targets through a specifically designed network architecture and loss function, without significantly increasing the computational burden. To achieve this goal, we first design a DMS-Conv module. By employing dense sampling and channel function separation strategies, this module effectively expands the receptive field while avoiding the high computational overhead and sampling artifacts associated with traditional multi-scale methods, thereby significantly improving feature representation for faint targets. Secondly, to optimize information flow within the feature pyramid, we propose a Lightweight Upsampling Module (LUM). Integrating depthwise separable convolutions with a channel reshuffling mechanism, this module replaces traditional transposed convolutions at a minimal computational cost, facilitating more efficient multi-scale feature fusion. Finally, addressing the stringent requirements for small target localization accuracy, we introduce the MPD-IoU Loss. By incorporating the diagonal distance of bounding boxes as a geometric penalty term, this loss function provides finer and more direct spatial alignment constraints for model training, effectively boosting localization precision. Experimental results on a self-constructed real-world SLR observation dataset demonstrate that the proposed model achieves an ${\mathrm{mAP}}_{50:95}$ of 47.13% and an F1-score of 88.24%, with only 2.57 M parameters and 6.7 GFLOPs. Outperforming various mainstream lightweight detectors in the comprehensive performance of precision and recall, these results validate that our method effectively resolves the small target detection challenges in SLR scenarios while maintaining a lightweight design, exhibiting superior performance and practical value. Full article

Accurately reconstructing high-resolution (HR) images remains challenging in scenarios where HR observations cannot be captured due to optical, hardware, or cost constraints. To address this limitation, we introduce an image super-resolution (SR) framework that reconstructs HR content solely from multiple low-resolution (LR) measurements, without relying on any HR reference images. The proposed method formulates a unified degradation model that describes how HR pixels contribute to LR observations under subpixel shifts and anisotropic downsampling. Based on this model, we develop an adaptive search algorithm capable of identifying the minimal and most informative combination of LR images required to equivalently represent the latent HR image. The selected LR images are then used to construct a solvable linear system whose solution directly yields the HR pixel values. Experiments conducted on the USAF 1951 resolution target demonstrate that the proposed approach improves Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) by 27.33% and 44.64%, respectively, achieving a resolvable spatial frequency of 228 line pairs per millimeter. In semiconductor chip inspection, PSNR and SSIM increase by 22.36% and 40.38%. These results verify that the proposed LR-combination-based strategy provides a physically interpretable and highly practical alternative for applications in which HR reference images cannot be obtained. Full article

High-speed data acquisition systems based on field-programmable gate arrays (FPGAs) often face synchronization challenges when interfacing with commercial analog-to-digital converters (ADCs), particularly under constrained hardware routing conditions and vendor-specific clocking assumptions. This work presents a vendor-independent FPGA–ADC synchronization architecture that enables reliable and repeatable high-speed data acquisition without relying on clock-capable input resources. Clock and frame signals are internally reconstructed and phase-aligned within the FPGA using mixed-mode clock management (MMCM) and input serializer/deserializer (ISERDES) resources, enabling time-sequential phase observation without the need for parallel snapshot or delay-line structures. Rather than targeting absolute metrological limits, the proposed approach emphasizes a reproducible and transparent data acquisition methodology applicable across heterogeneous FPGA–ADC platforms, in which clock synchronization is treated as a system-level design parameter affecting digital interface timing integrity and data reproducibility. Experimental validation using a custom Kintex-7 (XC7K325T) FPGA and an AFE7225 ADC demonstrates stable synchronization at sampling rates of up to 125 MS/s, with frequency-offset tolerance determined by the phase-tracking capability of the internal MMCM-based alignment loop. Consistent signal acquisition is achieved over the 100 kHz–20 MHz frequency range. The measured interface level timing uncertainty remains below 10 ps RMS, confirming robust clock and frame alignment. Meanwhile, the observed signal-to-noise ratio (SNR) performance, exceeding 80 dB, reflects the phase–noise-limited measurement quality of the system. The proposed architecture provides a cost-effective, scalable, and reproducible solution for experimental and research-oriented FPGA-based data acquisition systems operating under practical hardware constraints. Full article

Independent vector analysis (IVA) has emerged as a powerful tool for fusing and analyzing functional magnetic resonance imaging (fMRI) data. Applying IVA to multi-task fMRI data enhances analytical power by capturing the relationships across different tasks in order to discover their underlying multivariate relationship to one another. Incorporation of prior information into IVA enhances the separability and interpretability of estimated components. In this paper, we demonstrate successful fusion of multi-task fMRI feature data under two settings: constrained IVA and constrained transposed IVA (tIVA). We show that using these methods for fusing multi-task fMRI feature data offers novel ways to improve the quality and interpretability of the analysis. While constrained IVA extracts components linked to distinct brain networks, tIVA reverses the roles of spatial components and subject profiles, enabling flexible analysis of behavioral effects. We apply both methods to a multi-task fMRI dataset of 247 subjects. We demonstrate that for task-based fMRI, structural MRI (sMRI) references provide a better match for task data than resting-state fMRI (rs-fMRI) references, and using sMRI priors improves identification of group differences in task-related networks, such as the sensory-motor network during the Auditory Oddball (AOD) task. Additionally, constrained tIVA allows for targeted investigation of the effects of behavioral variables by applying them individually during the analysis. For instance, by using the letter number sequence subtest, a measure of working memory, as a behavioral constraint in tIVA, we observed significant group differences in the auditory and sensory-motor networks during the AOD task. Results show that the use of two constrained approaches, guided by well-aligned structural and behavioral references, enables a more comprehensive analysis of underlying brain function as modulated by task. Full article