Falls are a major public health concern for the aging population, leading to significant injuries, loss of independence, and increased healthcare costs. While wearable devices present promising solutions, existing alg...
详细信息
Falls are a major public health concern for the aging population, leading to significant injuries, loss of independence, and increased healthcare costs. While wearable devices present promising solutions, existing algorithms are often hindered by the limitations of microcontroller units (MCU) in terms of computational power, memory, and energy consumption. To overcome these challenges, we introduce MicroFallNet, a lightweight convolutional neural network designed for accurate and efficient fall detection. MicroFallNet features a novel FireModel architecture, incorporating Squeeze and Expand layers to optimize computational efficiency and enhance feature extraction. The proposed algorithm demonstrates superior performance on the UMAFALL and FallAllD datasets, achieving geometric mean accuracies of 97.91 % and 97.86 %, respectively, significantly surpassing traditional methods. Additionally, MicroFallNet showcases excellent deployment efficiency across various microcontrollers, particularly excelling on the ESP32 smart wristband platform, where it achieves an inference time of just 30.3 milliseconds. This capability makes MicroFallNet ideally suited for real-time fall detection applications, advancing the development of wearable devices for the elderly and contributing substantially to the field of smart health monitoring. Our code will be publicly available at https://***/qwer12330/ MicroFallNet-A-Lightweight-Model-for-Real-Time-Fall-Detection-on-Smart-Wristbands-Using-Sm.
The regularly ordered core-shell nanorods provide an effective transport channel for the adsorption and desorption of the target molecules. However, it remains a great challenge to precisely control the shell thicknes...
详细信息
The regularly ordered core-shell nanorods provide an effective transport channel for the adsorption and desorption of the target molecules. However, it remains a great challenge to precisely control the shell thickness of the ordered core-shell nanorods and improve the catalytic activity of their surfaces due to the complexity of the reaction system, especially by enhancing their selectivity to short-chain alcohols. Herein, we propose ZnO/TiO2 core-shell nanorod arrays (NRs) with different shell thicknesses prepared by electron beam evaporation and found that they possess advanced selectivity for short-chain alcohols. The response time of a ZnO/TiO2 core-shell NR with a TiO2 shell of 30 nm to 100 ppm isoamyl alcohol is 17 s. Its response to 100 ppm of isoamyl alcohol is 807, which is approximately 158 times that of the pure ZnO NR. The ZnO/TiO2 NR (30 nm) sensor exhibited excellent selectivity for methanol, ethanol, isopropanol, n-butanol, and isoamyl alcohol. The response to alcohols increased with the increase of the alcohols' chain length. Its sensing mechanism for short-chain alcohols is explained in terms of adsorption energy, acidity, and electronegativity of H in organic groups by density functional theory in detail. Our results open an alternative route for the design of sensitive materials for the selective detection of short-chain alcohols.
Test-time adaptation (TTA) aims to provide neural networks capable of adapting to the target domain distribution using only unlabeled test data. Most existing TTA methods have achieved success under mild conditions, s...
详细信息
Test-time adaptation (TTA) aims to provide neural networks capable of adapting to the target domain distribution using only unlabeled test data. Most existing TTA methods have achieved success under mild conditions, such as independently sampled data from a single or multiple static domains. However, these attempts may fail in dynamic scenarios, where the test data distribution undergoes continuous changes over time. By digging into the failure cases, we find that high-entropy or noisy samples during long-term adaptation may lead to inevitable catastrophic failure. Thus, we propose a Robust Gradient Aware and Reliable entropy minimization approach, called RGAR, to further stabilize TTA from three aspects: (1) Boosting model robustness to distribution shift, we propose a dual-stream perturbation technique that enables two weak-to-strong perturbation views of the student model guided by a common strong view of the mean teacher model;(2) mitigating the impact of high-entropy samples from different scenarios, we present to minimize the reliable samples that take into account both the distribution shift and sample adaptation degree;(3) enabling the model to be insensitive to small perturbations by encouraging model weights to reach flatter minima while focusing on the maximal gradient norm. Extensive experimental results demonstrate the effectiveness of our proposed method, RGAR. We achieve state-of-the-art performance on widely used benchmark datasets, such as CIFAR10C, CIFAR100C, and ImageNet-C. Our source code is available at https://***/r/D152/.
In this paper, a second-order algorithm based on the spectral deferred correction method is constructed for the time-dependent natural convection problem, which allows one to automatically increase the accuracy of a f...
详细信息
In this paper, a second-order algorithm based on the spectral deferred correction method is constructed for the time-dependent natural convection problem, which allows one to automatically increase the accuracy of a first-order backward-Euler time-stepping method through using spectral integration on Gaussian quadrature nodes and constructing the corrections. A complete theoretical analysis is presented to prove that this algorithm is unconditionally stable and possesses second-order accuracy in time. Numerical examples are given to confirm the theoretical analysis and the effectiveness of our algorithm.
Lead-free double perovskite Cs2AgBiBr6 has garnered increasing attention in photoelectric applications owing to its good stability and excellent photoelectric properties. However, the poor carrier transport in Cs2AgBi...
详细信息
Lead-free double perovskite Cs2AgBiBr6 has garnered increasing attention in photoelectric applications owing to its good stability and excellent photoelectric properties. However, the poor carrier transport in Cs2AgBiBr6 thin films constraints their further application in photodetection. To overcome this issue, we have developed an innovative low-dimensional Cs2AgBiBr6/CdS heterojunction photodetector with substantially improved performance. The device achieved a high responsivity of 6.66 x 10(3) A/W, an outstanding specific detectivity of 2.10 x 10(14) Jones, and an impressive external quantum efficiency of 1.88 x 10(6) %. Additionally, the on/off current ratio of the heterojunction device reached an impressive 6.18 x 10(7). These key parameters are significantly better than those of most previously reported Cs2AgBiBr6-based photodetectors. Furthermore, scanning photocurrent mapping and band arrangement analysis were performed to elucidate the mechanism of photocurrent generation and transport in the low-dimensional Cs2AgBiBr6/CdS heterojunction photodetectors. This study highlights the outstanding performance of Cs2AgBiBr6/CdS heterojunction and provides a simple and effective strategy for developing high-performance Cs2AgBiBr6-based photodetectors.
Vacuum freeze-drying (VFD) technology has gained extensive application across various sectors, particularly in environmental applications, where it is primarily utilized for the fabrication of environmental functional...
详细信息
Vacuum freeze-drying (VFD) technology has gained extensive application across various sectors, particularly in environmental applications, where it is primarily utilized for the fabrication of environmental functional materials and the conservation of environmental organisms. This technology is applicable to soil enhancement, the remediation of aquatic pollutants, energy storage in thermoelectric materials, and the preservation of bacterial cultures. This review synthesizes the most recent advancements in VFD technology within the environmental domain, elaborating on its technical fundamentals, operational procedures, practical applications, and distinctive benefits. Furthermore, the article explores the prospective development trajectory and potential challenges for this technology in the environmental sector, offering scientific guidance for its continued application and insights into its innovative progression.
Biomass utilization in energy production is often regarded as almost carbon neutral, and biomass gasification has significant potential for obtaining products with more value and potential applications for utilizing a...
详细信息
Biomass utilization in energy production is often regarded as almost carbon neutral, and biomass gasification has significant potential for obtaining products with more value and potential applications for utilizing agricultural and forestry waste. Meanwhile, Phosphogypsum (PG) is a hazardous solid waste generated as a by-product of the wet process of phosphoric acid production, resulting in significant environmental issues. Based on the above, this work proposed that solid waste be utilized effectively by producing syngas through biomass gasification using PG as a gasification agent. The thermal characteristics of biomass with PG and the distribution of gaseous products were discussed, and the effects of the ratio of PG to biomass (O/B values) and temperature were investigated. The mechanism function and activation energy (Ea) were determined by the model-free (Kissinger-Akahira-Sunose method), model-fitting (Coats and Redfern method), and Malek methods. The results indicate that the solid-phase product of the process is calcium sulfide. The experiment obtained the highest peak mass loss R p2 and comprehensive devolatilization parameter D when the O/B value of 0.5, leading to better devolatilization behavior during the devolatilization stage. The kinetics observed during the devolatilization stage show a decreased activation energy as the reaction progresses. Conversely, it gradually increases during the biochar gasification stage. The primary gaseous product emissions exhibited observable peaks in the devolatilization and biochar gasification stages. A significant amount of methane is released during the biochar gasification phase. Alkali metals and their oxides in PG are believed to serve as catalysts or facilitate the process of tar cracking during gasification.
In this paper, we propose and analyze a two-grid algorithm based on Newton iteration for solving the stationary inductionless magnetohydrodynamic system. The method involves first solving a small nonlinear system on a...
详细信息
In this paper, we propose and analyze a two-grid algorithm based on Newton iteration for solving the stationary inductionless magnetohydrodynamic system. The method involves first solving a small nonlinear system on a coarse grid with grid size H, followed by solving two linear problems on a fine grid with grid size h. These linear problems share the same stiffness matrix but differ only in their right-hand sides. The scaling between the coarse and fine grids is improved by our new method, while the approximate solution retains the same order of convergence as that observed in conventional methods. Furthermore, H0(div,Q) x L20(Q)-conforming finite element pairs are utilized to discretize the current density and electric potential, ensuring that the discrete current density is exactly divergence-free. Stability and convergence analyses are rigorously derived, and L2-error estimates for the velocity are provided. Numerical experiments are presented to verify the theoretical predictions and demonstrate the efficiency of the proposed method.
The Greater Mekong Subregion (GMS) is experiencing significant changes in forest area, prompting an urgent investigation into whether the alterations in carbon stock from forest loss and restoration can meet the regio...
详细信息
The Greater Mekong Subregion (GMS) is experiencing significant changes in forest area, prompting an urgent investigation into whether the alterations in carbon stock from forest loss and restoration can meet the region's need for increased carbon sequestration. Therefore, utilizing remote sensing data such as Landsat and machine learning methods, we established distribution maps of primary and secondary forests and forest carbon density maps for the GMS from 2000 to 2020. By analyzing the gradient effect of forest carbon density across four altitude zones, we investigated the altitude asymmetry of the compensatory effect of secondary forests in the GMS, and predicted the carbon potential of the regional forests. The results indicate that, influenced by human activities, forests in the GMS have transitioned from the loss of primary forests in the 2000s to the recovery of secondary forests in the 2010s. While the rates of area change for loss (-2.22 x 105 ha yr- 1) and recovery (1.97 x 105 ha yr- 1) were similar, an altitude asymmetry caused a regional forest carbon imbalance. The low and mid- altitude regions, with higher carbon density and significant forest loss, can only compensate for 31.50 % of the carbon loss in low-altitude (122.28 TgC) and 47.57 % in mid-altitude (76.25 TgC) through secondary forest recovery. In contrast, the high-altitude region, with lower carbon loss (12.93 TgC) and larger recovery area, results in a forest net carbon sink of 10.66 TgC. Over the next decade, if primary forest loss continues at the current pace, existing secondary forest growth will absorb only 50.41 % of carbon emissions. Therefore, collaboration among GMS countries is essential to protect primary forests and promote secondary forest planting in low to mid-altitude areas for sustainable regional forest carbon development.
With the widespread application of intelligent optimization algorithms in engineering and scientific fields, the need for their improvement and optimization is increasingly evident. This paper proposes an Improved Gra...
详细信息
With the widespread application of intelligent optimization algorithms in engineering and scientific fields, the need for their improvement and optimization is increasingly evident. This paper proposes an Improved Gray Wolf Optimization (IGWO) algorithm based on the Gray Wolf Optimization (GWO) method, aiming to enhance global search capability and convergence speed. The proposed IGWO algorithm integrates goodset initialization and the golden sine strategy to enhance global search efficiency and improve local search accuracy. The algorithm is specifically applied to the temporal modulation optimization of antenna arrays, where the switching durations of array elements are modulated to optimize the synthesis of the radiation pattern. The core goal of temporal modulation is to adjust the switching times of individual elements to alter the array's radiation pattern, achieving effects such as sidelobe suppression, main beam optimization, and reduction of harmonic distortion. Experimental validation using four benchmark functions demonstrates that the IGWO algorithm significantly outperforms traditional algorithms, including the seagull optimization algorithm, hunger games search optimization algorithm, seahorse optimizer, and GWO algorithm, across all test cases. In addition, IGWO is applied to optimize three time-modulation models, particularly those that optimize both connection durations and inter-element spacing simultaneously. Despite higher second harmonic distortion than the differential evolution with wavelet mutation algorithm in some cases, IGWO consistently excels in minimizing fundamental sidelobes, first harmonics, and second harmonics. Experimental results validate the algorithm's fast convergence and high optimization accuracy in optimizing the radiation pattern of antenna arrays, showcasing its effectiveness and practicality in addressing electromagnetic problems. (c) 2025 Author(s). All article content, except where otherwise noted, is licensed under a
暂无评论