Search Results (11,206)

Accurate workload prediction is essential for proactive resource allocation in large-scale Content Delivery Networks (CDNs), where traffic patterns are highly dynamic and geographically distributed. This paper introduces a CDN-tailored prediction and autoscaling framework that integrates statistical and deep learning models within an adaptive feedback loop. The framework is evaluated using 18 months of real traffic traces from a production multi-tier CDN, capturing realistic workload seasonality, cache–tier interactions, and propagation delays. Unlike generic cloud-edge predictors, our design incorporates CDN-specific features and model-switching mechanisms to balance prediction accuracy with computational cost. Seasonal ARIMA (S-ARIMA), Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM), and Online Sequential Extreme Learning Machine (OS-ELM) are combined to support both short-horizon scaling and longer-term capacity planning. The predictions drive a queue-based resource-estimation model, enabling proactive cache–server scaling with low rejection rates. Experimental results demonstrate that the framework maintains high accuracy while reducing computational overhead through adaptive model selection. The proposed approach offers a practical, production-tested solution for predictive autoscaling in CDNs and can be extended to other latency-sensitive edge-cloud services with hierarchical architectures. Full article

Background/Objectives: With rapid population aging, concerns about cognitive health—especially mild cognitive impairment (MCI), a prodromal stage of dementia—are growing. Although MCI prevalence is rising, limited empirical evidence exists on changes in healthcare expenditures associated with its diagnosis. This study aimed to assess shifts in medical spending before and after MCI diagnosis and to identify factors influencing healthcare costs among Korean adults. Methods: We used data from the National Health Insurance Service–National Health Information Database (NHIS–NHID) from 2020 to 2022. This study analyzed 4162 Korean adults aged ≤84 who were newly diagnosed with MCI in 2021. Annual healthcare expenditures were tracked from 2020 to 2022. Generalized estimating equations (GEEs) were employed to examine changes over time, adjusting for sociodemographic characteristics, comorbidities, healthcare utilization, and long-term care insurance (LTCI) enrollment. Results: The average annual healthcare expenditure increased from 74,767 KRW before diagnosis to 87,902 KRW after diagnosis, reflecting a 12.51% rise. Regression analysis showed a significant decrease in costs in the year prior to diagnosis (β = −0.117, p < 0.01) and an increase in the year following diagnosis (β = 0.061, p < 0.01). Higher expenditures were associated with greater outpatient visits (β = 0.385, p < 0.01), longer hospital stays (β = 0.039, p < 0.01), LTCI enrollment (non-graded: β = 0.035, p = 0.02; graded: β = 0.027, p = 0.04) and higher comorbidity levels (CCI = 2: β = 0.088, p < 0.01, CCI ≥ 3: β = 0.192, p < 0.01). Conversely, older age (β = −0.003, p = 0.02) and female sex (β = −0.093, p < 0.01) were associated with lower costs. Sex-stratified analyses revealed consistent cost trends but different predictors for male and female patients. Conclusions: Healthcare expenditures rise significantly after MCI diagnosis. Early identification and interventions tailored to patient characteristics—such as age, sex, and comorbidity status—may help manage future costs and support equitable care for older adults. Full article

Received Signal Strength Indicator (RSSI) prediction is valuable for network planning and optimization as it helps determine the optimal placements of wireless access points and enables better coverage planning. It is also crucial for efficient handover management between cells or access points, reducing dropped connections and improving service quality. Additionally, RSSI prediction supports indoor positioning systems, power management optimization, and cost-efficient network deployment. Path loss models have historically served as the foundation for RSSI prediction, providing a theoretical framework for estimating signal strength degradation. However, modern machine learning approaches have emerged as a revolutionary solution for network optimization, providing more versatile and data-driven methods to enhance wireless network performance. In this paper, an adaptive machine learning framework integrating environmental sensing parameters such as temperature, relative humidity, barometric pressure, and particulate matter for RSSI prediction is proposed. Performance analysis reveals that RSSI values are influenced by environmental factors through complex, non-linear interactions, thereby challenging the conventional linear assumptions of traditional path loss models. The proposed model demonstrates improved predictive accuracy over the baseline, with relative increases in variance explained of 6.02% and 2.04% compared to the baseline model excluding and including environmental parameters, respectively. Additionally, the root mean squared error is reduced to 1.40 dB. These results demonstrate that cognitive methods incorporating environmental data can substantially enhance RSSI prediction accuracy in wireless communications. Full article

The estimation of soil hydraulic properties—such as water retention and hydraulic conductivity—is essential for irrigation management and agro-hydrological modeling. This study presents the development and application of SOILHP, a low-cost, IoT-integrated device designed to monitor laboratory evaporation experiments for the estimation of soil hydraulic properties using inverse modeling tools. SOILHP incorporates mini-tensiometers, a precision balance, microcontrollers, and cloud-based data logging via Google Sheets. SOILHP enables the remote, real-time acquisition of soil pressure head and mass variation data without the need for commercial dataloggers. Evaporation experiments were conducted using undisturbed soil samples, and inverse modeling with Hydrus-1D was used to estimate van Genuchten–Mualem parameters. The optimized parameters showed low standard errors and narrow 95% confidence intervals, demonstrating the robustness of the inverse solution, confirming the device’s sensors accuracy. Forward simulations of internal drainage were performed to estimate the field capacity under different drainage flux criteria. The field capacity results aligned with values reported in the literature for tropical soils. Overall, SOILHP proved to be a reliable and economically accessible alternative for monitoring evaporation experiments aimed at fitting parameters of analytical functions that describe water retention and hydraulic conductivity properties within the soil pressure head range relevant to agriculture. Full article

Resource maps for open-loop ground source heat pump (GSHP) systems were developed based on heating and cooling experiments to identify areas with potential for reduced operational costs. Experiments conducted at a public hall, where groundwater temperatures fluctuate seasonally, clarified the relationships between the coefficient of performance (COP) of a heat pump and three key parameters: groundwater temperature, flow rate, and energy consumption. Multiple regression analysis produced equations for estimating the energy consumption of both the heat pump and the water pump. Results indicate that groundwater temperature influences the COP, increasing it by 0.07969 per °C during heating and decreasing it by 0.1721 per °C during cooling. These equations enable the estimation of energy consumption in open-loop systems from groundwater temperature, groundwater depth, and building type. Resource maps developed for the Nobi Plain in central Japan reveal that annual energy consumption is lower in the northwestern region, where groundwater temperatures are generally lower, except in marginal zones for hospitals and offices. Full article

Green hydrogen is gaining global attention as a zero-carbon energy carrier with the potential to drive sustainable energy transitions, particularly in regions facing rising fossil fuel costs and resource depletion. In sub-Saharan Africa, financing mechanisms and structured off-take agreements are critical to attracting investment across the green hydrogen value chain, from advisory and pilot stages to full-scale deployment. While substantial funding is required to support a green economic transition, success will depend on the effective mobilization of capital through smart public policies and innovative financial instruments. This review evaluates financing mechanisms relevant to sub-Saharan Africa, including green bonds, public–private partnerships, foreign direct investment, venture capital, grants and loans, multilateral and bilateral funding, and government subsidies. Despite their potential, current capital flows remain insufficient and must be significantly scaled up to meet green energy transition targets. This study employs a mixed-methods approach, drawing on primary data from utility firms under the H₂Atlas-Africa project and secondary data from international organizations and the peer-reviewed literature. The analysis identifies that transitioning toward Net-Zero emissions economies through hydrogen development in sub-Saharan Africa presents both significant opportunities and measurable risks. Specifically, the results indicate an estimated investment risk factor of 35%, reflecting potential challenges such as financing, infrastructure, and policy readiness. Nevertheless, the findings underscore that green hydrogen is a viable alternative to fossil fuels in sub-Saharan Africa, particularly if supported by targeted financing strategies and robust policy frameworks. This study offers practical insights for policymakers, financial institutions, and development partners seeking to structure bankable projects and accelerate green hydrogen adoption across the region. Full article

In response to the global push for sustainable development, green building certification systems have become a key policy instrument for reducing carbon emissions in the construction sector. In Taiwan, the EEWH (Ecology, Energy Saving, Waste Reduction, and Health) system serves as the primary framework for evaluating building sustainability. However, while government incentives such as floor area ratio (FAR) bonuses aim to encourage adoption, private sector participation remains limited, especially in the residential sector. This study investigates the cost implications and incentive benefits of upgrading green building certification from the Silver level to the Gold level under the EEWH system, using eight collective housing projects in the Taipei metropolitan area as case studies. Through a detailed analysis of certification components, upgrade strategies, and construction cost estimates, this research quantifies the additional costs required for each sustainability indicator and evaluates the alignment between upgrade investments and incentive rewards. The findings reveal that the average cost increase associated with the Silver-to-Gold upgrade ranges between 1% and 3% of total construction costs, with certain design strategies offering high cost-effectiveness. Moreover, the study examines whether the current FAR bonus policy provides adequate motivation for developers to pursue higher certification levels. The results provide valuable insights for policymakers seeking to optimize incentive structures and for developers considering sustainable building investments. Full article

In wheat pest and disease control methods, pesticide application occupies a dominant position, and the use of UAVs for precise pesticide application is a key technology in precision agriculture. However, it is difficult for existing UAV spraying systems to accurately achieve variable spraying according to crop growth conditions, resulting in pesticide waste and environmental pollution. To address this issue, this paper proposes a LiDAR-assisted UAV variable-speed spraying system. Firstly, a biomass estimation model based on LiDAR data and RGB data is constructed, LiDAR point cloud data and RGB data are extracted from the target farmland, and, after preprocessing, key parameters including LiDAR feature variables, canopy cover, and visible-light vegetation indices are extracted from the two types of data. Using these key parameters as model inputs, multiple machine learning methods are employed to build a wheat biomass estimation model, and a variable spraying prescription map is generated based on the spatial distribution of biomass. Secondly, the variable-speed spraying system is constructed, which integrates a prescription map interpretation module and a PWM control module. Under the guidance of the variable spraying prescription map, the spraying rate is adjusted to achieve real-time variable spraying. Finally, a comparative experiment is designed, and the results show that the LiDAR-assisted UAV variable spraying system designed in this study performs better than the traditional constant-rate spraying system; while maintaining equivalent spraying effects, the usage of chemical agents is significantly reduced by 30.1%, providing a new technical path for reducing pesticide pollution and lowering grain production costs. Full article

We have developed a generalisable machine learning framework for reservoir quality prediction in deeply buried clastic systems. Applied to the Lower Jurassic deltaic sandstones of the Tilje Formation (Halten Terrace, North Sea), the approach integrates sedimentological facies modelling with mineralogical and petrophysical prediction in a single workflow. Using supervised Extreme Gradient Boosting (XGBoost) models, we classify reservoir facies, predict permeability directly from standard wireline log parameters and estimate the abundance of porosity-preserving grain coating chlorite (gamma ray, neutron porosity, caliper, photoelectric effect, bulk density, compressional and shear sonic, and deep resistivity). Model development and evaluation employed stratified K-fold cross-validation to preserve facies proportions and mineralogical variability across folds, supporting robust performance assessment and testing generalisability across a geologically heterogeneous dataset. Core description, point count petrography, and core plug analyses were used for ground truthing. The models distinguish chlorite-associated facies with up to 80% accuracy and estimate permeability with a mean absolute error of 0.782 log(mD), improving substantially on conventional regression-based approaches. The models also enable prediction, for the first time using wireline logs, grain-coating chlorite abundance with a mean absolute error of 1.79% (range 0–16%). The framework takes advantage of diagnostic petrophysical responses associated with chlorite and high porosity, yielding geologically consistent and interpretable results. It addresses persistent challenges in characterising thinly bedded, heterogeneous intervals beyond the resolution of traditional methods and is transferable to other clastic reservoirs, including those considered for carbon storage and geothermal applications. The workflow supports cost-effective, high-confidence subsurface characterisation and contributes a flexible methodology for future work at the interface of geoscience and machine learning. Full article

Faster-than-Nyquist (FTN) tight filtering introduces serious inter-symbol interference (ISI) impairment, leading to an insufficient compensation range for conventional IQ imbalance compensation algorithms. Furthermore, receiver (Rx) IQ imbalance and ISI impairments significantly increase the convergence cost required by the squared Gardner phase detector (SGPD) timing recovery algorithm to establish a timing synchronization loop. This paper proposes a joint Rx IQ compensation and timing recovery scheme. By embedding a two-stage IQ imbalance compensation algorithm into the timing recovery feedback loop, the proposed scheme could effectively estimate and compensate for Rx IQ imbalance. Meanwhile, thanks to the innovative scheme, which equalizes Rx IQ imbalance and ISI during the timing feedback loop, the convergence cost of timing recovery could be reduced compared with the conventional blind frequency domain (BFD) scheme. The simulation results of 128 GBaud polarization multiplexing (PM) 16-quadrature amplitude modulation (QAM) FTN wavelength division multiplexing (WDM) transmission systems demonstrate that the proposed scheme could bring about 14%, 12.5%, and 16.6% improvements in the compensation range for Rx IQ amplitude imbalance, phase imbalance, and skew, respectively, compared with the conventional one. Meanwhile, the convergence cost is reduced by at least 31% with a 0.9 acceleration factor. In addition, 40 GBaud PM-16QAM FTN experiment results show that the proposed scheme could bring about a 0.8 dB improvement in the optical signal noise ratio (OSNR) compared with the conventional BFD scheme. Full article

The compressive strength (Fc) of cement mortar (CM) is a key parameter in ensuring the mechanical reliability and durability of cement-based materials. Traditional testing methods are labor-intensive, time-consuming, and often lack predictive flexibility. With the increasing adoption of machine learning (ML) in civil engineering, data-driven approaches offer a rapid, cost-effective alternative for forecasting material properties. This study investigates a wide range of supervised linear and nonlinear ML regression models to predict the Fc of CM. The evaluated models include linear regression, ridge regression, lasso regression, decision trees, random forests, gradient boosting, k-nearest neighbors (KNN), and twelve neural network (NN) architectures, developed by combining different optimizers (L-BFGS, Adam, and SGD) with activation functions (tanh, relu, logistic, and identity). Model performance was assessed using the root mean squared error (RMSE), coefficient of determination (R²), and mean absolute error (MAE). Among all models, NN_tanh_lbfgs achieved the best results, with an almost perfect fit in training (R² = 0.9999, RMSE = 0.0083, MAE = 0.0063) and excellent generalization in testing (R² = 0.9946, RMSE = 1.5032, MAE = 1.2545). NN_logistic_lbfgs, gradient boosting, and NN_relu_lbfgs also exhibited high predictive accuracy and robustness. The SHAP analysis revealed that curing age and nano silica/cement ratio (NS/C) positively influence Fc, while porosity has the strongest negative impact. The main novelty of this study lies in the systematic tuning of neural networks via distinct optimizer–activation combinations, and the integration of SHAP for interpretability—bridging the gap between predictive performance and explainability in cementitious materials research. These results confirm the NN_tanh_lbfgs as a highly reliable model for estimating Fc in CM, offering a robust, interpretable, and scalable solution for data-driven strength prediction. Full article

This paper presents a literature review of low-power hybrid wind–photovoltaic (PV) systems and introduces a 3 m diameter prototype rotor featuring twelve PV-coated pivoting blades stiffened by a peripheral rim. Existing solutions—foldable umbrella concepts, Darrieus rotors with PV-integrated blades, and morphing blades—are surveyed, and current gaps in simultaneous wind + PV co-generation on a single moving structure are highlighted. Key performance indicators such as power coefficient (C_p), DC ripple, cell temperature difference (ΔT), and levelised cost of energy (LCOE) are defined, and an integrated assessment methodology is proposed based on blade element momentum (BEM) and computational fluid dynamics (CFD) modelling, dynamic current–voltage (I–V) testing, and failure modes and effects analysis (FMEA) to evaluate system performance and reliability. Preliminary results point to moderate aerodynamic penalties (ΔC_p ≈ 5–8%), PV output during rotation equal to 15–25% of the nominal PV power (P_PV), and an estimated 70–75% reduction in blade–root bending moment when the peripheral ring converts each blade from a cantilever to a simply supported member, resulting in increased blade stiffness. Major challenges include the collective pitch mechanism, dynamic shading, and wear of rotating components (slip rings); however, the suggested technical measures—maximum power point tracking (MPPT), string segmentation, and redundant braking—keep performance within acceptable limits. This study concludes that the concept shows promise for distributed microgeneration, provided extensive experimental validation and IEC 61400-2-compliant standardisation are pursued. This paper has a dual scope: (i) a concise literature review relevant to low-Re flat-blade aerodynamics and ring-stiffened rotor structures and (ii) a multi-fidelity aero-structural study that culminates in a 3 m prototype proposal. We present the first evaluation of a hybrid wind–PV rotor employing untwisted flat-plate blades stiffened by a peripheral ring. Using low-Re BEM for preliminary loading, steady-state RANS-CFD (k-ω SST) for validation, and elastic FEM for sizing, we assemble a coherent load/performance dataset. After upsizing the hub pins (Ø 30 mm), ring (50 × 50 mm), and spokes (Ø 40 mm), von Mises stresses remain < 25% of the 6061-T6 yield limit and tip deflection ≤ 0.5%·R acrosscut-in (3 m s⁻¹), nominal (5 m s⁻¹), and extreme (25 m s⁻¹) cases. CFD confirms a broad efficiency plateau at λ = 2.4–2.8 for β ≈ 10° and near-zero shaft torque at β = 90°, supporting a three-step pitch schedule (20° start-up → 10° nominal → 90° storm). Cross-model deviations for C_p, torque, and pressure/force distributions remain within ± 10%. This study addresses only the rotor; off-the-shelf generator, brake, screw-pitch, and azimuth/tilt drives are intended for later integration. The results provide a low-cost manufacturable architecture and a validated baseline for full-scale testing and future transient CFD/FEM iterations. Full article

Background: Respiratory rate (RR) is a key vital sign and one of the most sensitive indicators of physiological conditions, playing a crucial role in the early identification of clinical deterioration. The monitoring of RR using electrocardiography (ECG) and photoplethysmography (PPG) aims to overcome limitations of traditional methods in clinical settings. Methods: The proposed approach extracts RR from ECG and PPG signals using different morphological and temporal features from publicly available datasets (iAMwell and Capnobase). The algorithm was used to develop and test with a selection of relevant ECG (e.g., R-peak, QRS area, and QRS slope) and PPG (amplitude and frequency modulation) characteristics. Results: The results show promising performance, with the ECG-derived signal using the R-peak–based method yielding the lowest error, with a mean absolute error of 0.99 breaths/min in the iAMwell dataset and 3.07 breaths/min in the Capnobase dataset. In comparison, the RR PPG-derived signal showed higher errors of 5.10 breaths/min in the iAMwell dataset and 10.66 breaths/min in the Capnobase dataset, for the FM and AM method, respectively. Bland–Altman analysis revealed a small negative bias, approximately −0.97 breaths/min for the iAMwell dataset (with limits of agreement from −2.62 to 0.95) and −1.16 breaths/min for the Capnobase dataset (limits of agreement from −3.37 to 1.10) in the intra-subject analysis. In the inter-subject analysis, the bias was −0.84 breaths/min (limits of agreement from −1.76 to 0.20) for iAMwell and −1.22 breaths/min (limits of agreement from −7.91 to 5.35) for Capnobase, indicating a slight underestimation. Conversely, the PPG-derived signal tended to overestimate RR, resulting in higher variability and reduced accuracy. These findings highlight the higher reliability of ECG-derived features for RR estimation in the analyzed datasets. Conclusion: This study suggests that the proposed approach could guide the design of cost-effective, non-invasive methods for continuous respiration monitoring, offering a reliable tool for detecting conditions like stress, anxiety, and sleep disorders. Full article

This article examines the production costs of cocoa farming in Côte d’Ivoire, West Africa, taking into account the opportunity cost approach. To this end, surveys were conducted among 228 farmers in three regions, Bonon, Soubré and Biankouma, following an east–west gradient. The estimated costs of using family labor and land were based on the opportunity cost approach. The financial costs associated with production were also taken into account. Comparative analyses between different localities and cropping systems highlighted specific workload characteristics. Finally, a principal component analysis (PCA) was used to profile producers according to their income levels and profits. The findings showed that family labor was the main component of cocoa production costs. Prices paid to farmers did not always cover all production costs, with 38% of farmers producing at a loss, and this was contingent on the agro-ecological zone. Furthermore, the agroforestry system proved to be more economical in terms of labor than the full-sun system. These results underline the relevance of the opportunity cost approach in assessing production costs and setting cocoa selling prices. This should lead to a review of public price-setting mechanisms to ensure fair remuneration for family labor. Full article

Visually impaired individuals are at a high risk of accidents due to sudden changes in walking surfaces and surrounding obstacles. Existing smart cane systems lack the capability to detect pathway surface transition points with accurate distance estimation and danger-level assessment. This study proposes a low-cost smart cane that integrates a novel Pathway Surface Transition Point Detection (PSTPD) method with enhanced obstacle detection. The system employs dual RGB cameras, an ultrasonic sensor, and YOLO-based models to deliver real-time alerts based on object type, surface class, distance, and severity. It comprises three modules: (1) obstacle detection and classification into mild, moderate, or severe levels; (2) pathway surface detection across eight surface types with distance estimation using weighted bounding boxes and depth mapping; and (3) auditory notifications. Experimental results show a mean Average Precision (mAP@50) of 0.70 for obstacle detection and 0.92 for surface classification. The average distance estimation error was 0.3 cm for obstacles and 4.22 cm for pathway surface transition points. Additionally, the PSTPD method also demonstrated efficient processing with an average runtime of 0.6 s per instance. Full article