Search Results (3,996)

Ambient backscatter systems enable passive sensing and information transfer by utilizing the reflection and modulation of incident radio-frequency (RF) signals. However, in real-world scenarios involving non-cooperative targets such as off-the-shelf printed circuit boards (PCBs), the backscattered signal is extremely weak and often obscured by strong direct-path self-interference (SI) at the receiver. This issue becomes even more severe when unintentional PCB structures act as radiating elements. In this work, we explore ambient backscatter leakage from a compromised PCB using a realistic measurement setup that includes separated transmit and receive antennas and a direct-conversion Universal Software Radio Peripheral (USRP)-based receiver. We demonstrate that residual carrier frequency offset (CFO), caused by oscillator mismatch and hardware imperfections, can spread the dominant SI in the baseband and completely mask the weak backscattered signal. To solve this problem, a software-based post-processing framework is applied. This method leverages the complex baseband representation enabled by the homodyne receiver to jointly manage the carrier and SI components without relying on intermediate-frequency processing or prior knowledge of the target signal parameters. Experimental results show that this approach significantly improves the detectability of weak backscattered baseband information that would otherwise be concealed within the raw I/Q data. This study emphasizes the importance of CFO-aware digital processing in ambient backscatter systems and offers new insights into unintended electromagnetic leakage mechanisms from commercial PCB platforms. Full article

Industrial-Internet security faces a core challenge: improving detection accuracy for critical minority-class network attacks. The existing intrusion detection methods based on Conditional Generative Adversarial Nets (CGANs) aim to achieve data balance by reconstructing minority-class attack samples. However, they encounter problems such as generating deceptive samples, poor sample quality, vanishing gradients and difficulties in training. This paper proposes an intrusion detection method based on the Multi-Discriminator Conditional Classification Generative Adversarial Network (MDCCGAN), an improved variant of CGAN, which integrates multiple discriminators and an independent classifier into the traditional CGAN framework. The multiple discriminators reduce the probability of generating deceptive samples, the independent classifier decouples the classification loss to clarify the direction of gradient updates, and the introduction of the Wasserstein distance fundamentally addresses the gradient-vanishing problem. Experiments conducted on the NSL-KDD and UNSW-NB15 datasets demonstrate that the proposed method significantly improves the recall, F1-score and accuracy for minority-class attacks. Specifically, on the NSL-KDD dataset, the overall accuracy increases from 74% to 94%, and the F1-score for the extremely rare U2R attack surges from 0% to 77%. Similarly, on the UNSW-NB15 dataset, the accuracy reaches 88%, a 10% improvement over the baseline DNN, and the F1-scores for extreme minority attacks such as Analysis, Backdoor, and Worms improved to 97%, 62%, and 84%, respectively. These results confirm that our method effectively outperforms traditional generation models and common class-balancing methods. It provides reliable technical support for industrial-Internet security. Full article

Accurate short-range precipitation prediction plays a crucial role in daily life and disaster mitigation. However, the existing methods often suffer from inefficient large-scale feature extraction, severe redundant information interference, and insufficient attention to the problem of imbalanced data distributions, leading to unsatisfactory performance. To address these issues, in this paper, we first propose a novel spatiotemporal module called Group Large Kernel Convolution (GLKC) and develop a short-range precipitation forecasting model based on it, GLKC-Net, using multiple meteorological variables. Specifically, we use decomposed large-kernel convolution to enhance the ability to understand large-scale atmospheric processes. Meanwhile, we introduce the group convolution and channel shuffle operator to control the fusion of channel-wise information, enabling efficient information exchange and reducing redundancy in the channel dimension with multiple variables. Furthermore, we treat the causes of poor model performance for extreme precipitation events with an imbalanced data distribution perspective and design a Multi-threshold Adaptive Loss function (MTA Loss). This function strengthens the model’s focus on high-threshold precipitation events that are inherently difficult to forecast, aiming to improve model performance for extreme events. Finally, forecasting experiments for validation were conducted over southwestern China using ERA5-Land and CMPAS datasets. The results demonstrate that our proposed method outperforms several existing approaches in terms of forecasting accuracy. Full article

Despite increasing efforts by the scientific community to raise awareness of breed-related health problems through educational campaigns, public information initiatives, and veterinary outreach programmes, brachycephalic dog breeds remain highly popular. As the number of brachycephalic dogs increases, the prevalence of associated health problems rises accordingly. Ethical and animal welfare considerations appear to play a limited role in breed selection. In German-speaking regions, extensive educational efforts have been undertaken in recent years to address the issue of so-called torture breeding, defined as intentional selection for extreme phenotypic traits that impair health, reduce welfare, and cause chronic suffering, particularly in brachycephalic breeds. The aim of this study was to determine the underlying reasons for the decision to buy and keep a brachycephalic dog. Although the veterinary profession is already improving education and communication, this qualitative study intended to find new starting points for targeted education against animal suffering and to explore the sociological background of the ownership of such dogs. For this purpose, semi-structured interviews with people with brachycephalic dogs were conducted throughout Switzerland (n = 16). The focus was on the animal–human relationship. The interviews were defined by systematically applied guidelines for the design of the interview process, while still allowing maximum openness (all possibilities for expression). The transcribed interviews were coded and analysed according to the Kuckartz methodology, which allows us to set certain focal points of analysis and to structure them according to codes. The results of this study indicate that, although awareness of torture breeding is present within the broader population, owners of brachycephalic dogs frequently rely on individualised arguments and rationalisations. These typically involve emphasising the perceived health, functionality, or exceptional characteristics of their own animal (e.g., claims that their dog is “healthy” or not affected by breed-related problems), thereby distancing their personal ownership experience from the general welfare concerns associated with the breed. This psychological pattern can be interpreted as cognitive dissonance, in which contradictory beliefs are harmonised through selective perception or re-evaluation. The results also show that brachycephalic dogs offer a very strong projection surface: their owners assign them a variety of social roles that go beyond the classic animal–human relationship—for example, as a substitute for children, a romantic partner, or a best friend. This qualitative study provides differentiated insights into the attitudes and motivations of owners of brachycephalic dogs and illustrates that traditional awareness campaigns have not been sufficient to effectively change problematic breeding practices and ownership patterns. In order to develop long-term effective solutions, interdisciplinary cooperation is therefore needed—for example, between veterinary medicine, animal welfare, communication science, psychology and law. In addition to individual education, new, target-group-specific communication strategies and consistent legal regulations are needed to protect animal welfare in the long term. This study is intended to serve as a catalyst for a broader ethical and social debate on the keeping of torture breed dogs. Full article

The power fluctuation of large-scale photovoltaic (PV) clusters is significantly affected by cloud movement. Aiming at the engineering reality that meteorological observation data are generally lacking for most power stations in wide-area PV clusters, as well as the problem that existing models overfit second-order high-frequency noise such as microscopic cloud deformation, this paper proposes a disturbance reconstruction and smoothing effect quantification method for PV clusters focusing on the first-order dominant meteorological component. First, a clear-sky model is introduced as a deterministic trend filter to extract the purely random disturbance sequence that induces grid-connection risks from the measured output power. Second, the dimensionality reduction modeling concept of “macro-advection dominance and microscopic deformation filtering” is established: the PV cluster is finely partitioned by fusing Dynamic Time Warping (DTW) and geographical distance, and a cross-space inversion of the macro-cloud velocity vector is realized, driven by pure power data using the Time-Lagged Cross-Correlation (TLCC) algorithm, thus constructing a disturbance power generation model that accounts for the phase misalignment of power output. Independent verification based on measured data in Jilin Province shows that the 95% confidence interval of the power reconstructed only by the first-order advection characteristics can cover 90.2% of the measured fluctuations, and the reconstruction error of the fluctuation standard deviation—an indicator that determines the system reserve demand—is merely 5.9%. This verifies that the macro-cloud displacement is the absolute dominant factor governing the extreme fluctuations of PV clusters. Finally, a normalized Smoothing Factor (SF) characterizing the “reserve capacity release ratio” is constructed, and combined with its statistical indicators, it is used to quantitatively evaluate the smoothing benefits provided by different spatial layout schemes. Under data-constrained conditions, the method proposed in this paper verifies the engineering rationality that microscopic meteorological noise can be safely neglected at the macro-PV cluster scale, providing a reliable quantitative basis for the safe grid expansion and peak-shaving energy storage capacity sizing of high-proportion PV bases. Full article

Detecting small location shifts in stochastic processes is a fundamental problem in sequential statistical monitoring. Classical procedures such as Shewhart-type schemes, exponentially weighted moving average (EWMA), and cumulative sum (CUSUM) methods are known to perform well under normality or near-symmetry assumptions; however, their effectiveness may deteriorate in the presence of right-skewed distributions. In such settings, mean-based monitoring statistics can be highly sensitive to tail behavior, which may lead to delayed detection of small shifts or unstable false alarm performance. This paper introduces a monitoring scheme referred to as the Golden Ratio (GR) control chart, designed for detecting small location shifts in right-skewed distributions. The proposed method is constructed using a median-centered statistic combined with a geometrically decaying weighting mechanism derived from the golden ratio. Unlike classical time-based weighting schemes, the GR chart assigns weights according to the rank-based distance from the sample median, thereby attenuating the influence of isolated extreme observations while preserving sensitivity to persistent distributional changes. The run-length performance of the proposed chart is investigated using Monte Carlo simulation experiments. All competing procedures are calibrated to achieve comparable in-control average run lengths. The GR chart is compared with classical EWMA and CUSUM charts under several skewed distributions, including Gamma, Lognormal, and Weibull models. Simulation results indicate that the proposed approach provides a robust and stable monitoring alternative for skewed processes. In particular, the GR chart demonstrates competitive performance for detecting small location shifts while reducing the influence of extreme observations commonly encountered in right-skewed environments. Full article

Short-term forecasting of solar and wind power generation is critical for smart grid management but challenging due to non-stationarity and extreme generation events. This study addresses a multi-task learning problem: regression-based forecasting of power output and binary detection of extreme events defined by a quantile-based threshold (q = 0.90). A hybrid spatio-temporal model, DP-STH++, is proposed, implementing parallel causal fusion of LSTM, GRU, a causal Conv1D stack, and a lightweight causal transformer. The architecture employs regression and classification heads, while an uncertainty-weighted mechanism stabilizes multitask optimization in the regression tasks; extreme event detection performance is evaluated using AUC. Training and evaluation follow a leakage-safe protocol with chronological data processing, calendar feature integration, time-aware splitting, and training-only estimation of scaling parameters and extreme thresholds. Experimental results obtained with a one-hour forecasting horizon and a 24 h context window demonstrate that DP-STH++ achieves the best regression performance on the hold-out set (RMSE = 257.18, MAE = 174.86–287.90, MASE = 0.2438, R² = 0.9440) and the highest extreme event detection accuracy (AUC = 0.9896), ranking 1st among all compared architectures. In time-series cross-validation, the model retains the leading position with a mean MASE = 0.3883 and AUC = 0.9709. The advantages are particularly pronounced for wind power forecasting, where DP-STH++ simultaneously minimizes regression errors and maximizes AUC = 0.9880–0.9908. Full article

As the global aging population increases, the early detection and prevention of Alzheimer’s disease (AD) have become important in public health. To solve the problems of subjectivity and low timeliness of traditional assessment methods, this paper proposes a multimodal dementia prevention system that combines physiological sensing, a gamification interface, and a classification model. The system includes an interactive joystick to measure pulse and blood pressure. A Chinese music game app increases the participation of the elderly and reduces their sense of rejection through gamification interaction. After the physiological data were standardized by Z-score, they were input into three small sample classifiers (Gaussian Naïve Bayes, Fisher Linear Discriminant Analysis, and Logistic Regression) for the binary classification of AD. The system performance was evaluated using the Leave-One-Out cross-validation method. Experimental results show that Logistic Regression performed best in situations with extremely small samples and class imbalance, with an F1-score of 0.700, which was higher than the other two. Dynamic features and model fusion technologies need to be integrated to further enhance the clinical application potential of the system in the early prediction of dementia. Full article

This paper addresses the problem of multi-class classification of bacterial microscopic images using a rigorous experimental protocol designed to prevent information leakage and improve performance. The dataset consists of 2034 images representing 33 taxa, organized by class. Data integrity checks confirmed the absence of corrupted or unreadable files. To formalize image characteristics and ensure quality control, indirect geometric and textural features were calculated, including minimum frame size, brightness statistics (mean and standard deviation), Shannon entropy, Laplace variance, and Sobel gradient energy. Quality checks revealed a small proportion of images with extreme brightness (2.5074%), while no samples with critically low sharpness according to the selected criteria were detected. Statistical analysis of interclass differences using the Kruskal–Wallis test with multiple comparison correction demonstrated the high discriminatory power of texture features, specifically gradient energy (ε² = 0.819987) and Laplace variance (ε² = 0.709904). Feature correlations were consistent with their physical interpretation, revealing a strong positive relationship between sharpness and gradient energy. Principal component analysis confirmed a strong structural pattern, with the first two components explaining 75.5766% of the total variance. For a unified comparison, classical machine learning, transfer learning, and modern deep architectures were evaluated within a single protocol. Full article

Microplastic (MP) pollution in freshwater is an important global issue affecting an increasing number of areas. MP is ingested by aquatic organisms and transferred through food chains, causing impacts on both aquatic life and human health. While studies on MP uptake in the gastrointestinal tract (GIT) of fish are numerous globally, in Romania, there are extremely few. As a result, we conducted research on this phenomenon in fish species from the Crișul Repede River (CR) in two river sectors with different levels of anthropogenic impact. We found out that 100% of the collected fish had MPs in their GIT, with most of the particles being small-sized fragments (0.025–0.1 mm). Upstream, benthopelagic species ingested more MPs than downstream, whereas for benthic species, the amount of MPs in the GIT was greater downstream. Larger individuals contained more MPs than smaller ones. The presence of MPs in fish bodies can pose a problem if these particles enter internal organs and trigger adverse physiological effects. Full article

With the continuous rise in the energy density of power batteries, battery heat generation has become an increasingly severe issue. Particularly under extreme conditions combining high summer temperatures and high discharge rates, battery thermal safety is facing tremendous challenges. To address this problem, this study proposes a honeycomb liquid cooling–PCM hybrid battery thermal management system (BTMS) based on fin–casing synergistic heat transfer enhancement. We analyzed the effects of the longitudinal fins and thermal conductive casing on the thermal characteristics of the system, further investigated the influence patterns of key factors including fin number, battery spacing and contact thermal resistance on the thermal performance of the honeycomb BTMS, and clarified the action mechanisms of each structure and parameter on battery temperature rise and temperature uniformity. The results show that the fin structure enhances longitudinal heat conduction, improves liquid cooling efficiency, and effectively reduces the maximum battery temperature, while the thermal conductive casing significantly improves battery temperature uniformity. The BTMS with fin–casing synergistic heat transfer enhancement can control the maximum battery temperature and temperature difference within 60 °C and 5 °C, respectively, under extreme operating conditions. Full article

Developing polymer composites with improved mechanical and thermal properties requires strategies to overcome the problem of agglomeration and weak interfacial interactions of carbon nanofillers. This paper presents an effective strategy for the covalent functionalization of graphene oxide (GO) with melamine to create high-performance epoxy nanocomposites. The functionalization results in the formation of nitrogen-containing heterocyclic structures on the GO surface, as confirmed by FTIR and Raman spectroscopy. The addition of the obtained modified filler (mel-GO) into the epoxy matrix provides a synergistic effect: the melamine amino groups catalytically accelerate curing, reducing the gelation time from 146 to 48 min and increasing the maximum self-heating temperature from 94 to 122 °C, thus indicating enhanced interfacial interaction. This interaction results in a remarkable overall improvement in mechanical properties: tensile and flexural strengths increase by more than 20%, and elastic moduli by 31% and 58%, respectively, compared to the composite containing unmodified GO. At the same time, impact strength (from 14 to 23 kJ/m²) and hardness (up to 87 Shore D) increase. A key achievement is a dramatic increase in thermal and thermal-oxidative stability: the onset temperature of decomposition (T₅%) increases by 27 °C, the half-decomposition temperature (T₅₀%) by 45 °C, and the thermal stability index (THRI) increases from 119.3 to 128.9 °C. A more than twofold increase in coke residue yield (to 9.29%) and an increase in the Vicat softening point to 175 °C confirm the formation of an effective thermally stabilizing barrier layer due to the combined action of nitrogen-containing groups and dispersed graphene flakes. The proposed approach to functionalizing graphene oxide with melamine opens the way for the creation of next-generation epoxy composites with a record-breaking combination of strength, impact toughness, and thermal stability for applications in aerospace, electronics, and composite structures operating under extreme conditions. Full article

The frequency of extreme weather events has become higher, and electricity consumption has also become more complex. These changes increase the risk of overload in distribution transformers (DTs), and this risk threatens the stability and reliability of the power grid. Existing methods have significant limitations. Traditional static threshold methods (based on DGA gas ratios and electrical signal thresholds) fail to consider temporal changes and complex links between factors, while modern machine learning models lack cause–effect relationships over time and clear ways to describe uncertainty. With such motivations, this paper proposes a causal-enhanced hybrid framework, which combines Long Short-Term Memory (LSTM) networks and Random Forest (RF) algorithms. The framework uses causal Seasonal Trend decomposition using Loess (STL) to reveal load patterns at different time scales. The mutual information index and spatiotemporal graph convolutional network (ST-GCN) are used to explore nonlinear relations and reveal how temperature affects load changes. The LSTM model captures time dependence in load series, and the Bayesian optimized Random Forest is used to solve the problem of data imbalance and quantify uncertainty. In addition, the framework constructs an early warning system that combines data from many sources in real time. Test results show that the proposed algorithm exhibits excellent performance in multi-source data environments. Full article

Climate-induced natural hazards are increasingly disrupting metro operations in megacities, necessitating robust and generalizable frameworks for system-wide resilience. While current studies often treat infrastructure degradation, operational adjustment, and passenger flow redistribution as separate problems, this study proposes a resilience-oriented decision framework that couples these universal processes together to address diverse disruptive events. Taking extreme rainfall as a critical representative scenario, a multi-objective recovery optimization model is developed to jointly optimize repair resource cost and average section saturation. Resilience is quantified through the demand satisfaction ratio over the disruption–recovery process, ensuring the framework’s applicability across different hazard types. A case study of the Shanghai metro system under a real extreme rainfall event demonstrates the model’s efficacy in capturing complex system dynamics. Results show a clear Pareto trade-off between repair resource cost and average section saturation, while increasing service capacity on adjacent lines improves the Pareto frontier. Prioritizing repairs on lines with the fewest damaged sections effectively reduces network saturation by restoring corridor throughput. The resilience curve proves that higher repair resources not only shorten recovery time but also raise the minimum demand satisfaction ratio. These findings provide a scalable methodology for designing resilient metro recovery strategies under various climate-related disruptions globally. Full article

Nowadays, direct-drive systems are widely used in the actuators of computer numerical control machine tools. Linear motors are widely used in high-end computer numerical control machine tools due to their high positioning accuracy, good dynamic response, and simple transmission structure. First, a high-thrust-density concentrated magnetic linear permanent magnet vernier motor is proposed in this paper, which is designed by machine learning and optimized through an artificial intelligence optimization algorithm, to improve the air-gap magnetic density of the motor and improve the thrust density of the motor in principle; compared with traditional linear permanent magnet synchronous motors, the thrust density is increased by 40%. Second, using finite element calculations, a regression machine learning algorithm is proposed, which involves introducing a regression machine learning algorithm (called extreme learning machine (ELM)) to solve the computational modeling problem; compared with traditional ELM networks, it has faster training speed and higher stability. By mapping the nonlinear complex relationship between input structural factors and output motor performance, the superiority of the intelligent optimization algorithm is confirmed by comparative verification. Full article