Distance-learning platforms increasingly seek objective EEG-based markers of lecture comprehension, yet robust, imbalance-aware models on real datasets remain scarce. Using the Kaggle “EEG data / Distance learning environment” corpus (8 students, 14 Emotiv Epoc X channels, 84 tabular features/segment, ≈21% “not understood”), we propose IASTE, an Imbalance-Aware Stacked Tree Ensemble combining gradient boosting, RUSBoost, and bagged trees via leakage-free out-of-fold stacking and a logistic meta-learner. Under a stratified 70/15/15 train–validation–test split, with hyperparameters selected solely by minority-class F1 on validation, IASTE attains 97.3% test accuracy, macro-F1 = 0.959, and F1₀ (“not understood”) = 0.936. This improves macro-F1 by ≈1–2 percentage points and F1₀ by up to ≈1.5 points over strong baselines including Random Forest, RUSBoost, SVM, LSTM, BiLSTM, and BiLSTM+attention. Subject-wise analysis shows per-student accuracies in the range 0.968–0.980, versus ≈0.956–0.964 for tree and deep baselines, indicating genuine cross-subject generalization. Ablations confirm that removing stacking, imbalance handling, or F1₀-based selection systematically degrades minority-class F1 and macro-F1, while performance remains stable across the explored tree-depth and NumLearningCycles grid. By enabling objective, data-driven monitoring of lecture comprehension in distance-learning environments, our approach supports more inclusive and effective digital education.
Light Fidelity (LiFi) technology enters the stage, optical wireless communication (OWC) is fast becoming a potential answer for fast indoor internet access. When employing visible light for data transfer, LiFi presents a safe, high-bandwidth substitute for conventional wireless communication technologies. Nevertheless, current approaches have limited coverage, interference, and scalability problems in challenging settings. By evaluating and maximizing the LiFi technology capabilities using a SWOT (strengths, weaknesses, opportunities, threats) analysis, the suggested technique addresses these difficulties. Methodically assessing the internal and external elements affecting LiFi's performance helps one to spot areas requiring system design, spectrum management, and interference reduction. The LiFi system combines power regulation, dynamic frequency allocation, and adaptive beamforming among other technologies to boost coverage and reduce likely interference. By means of LiFi's installation and strengthening of its practical relevance in contemporary interior spaces, the framework seeks to give a more consistent, fast indoor internet experience. In interior environments, LiFi is a practical and effective replacement for high-speed internet as the suggested approach shows significant improvements in signal quality, scalability, and general system performance. Furthermore, the SWOT analysis helps to identify particular potential and drawbacks, therefore assuring that future LiFi installations are more ecologically friendly and efficient in many different working environments.
Sustainable meeting of growing energy requirements relies on the efficiency of newly developed wind turbine engine design. This approach resolves electromagnetic wrapping differences impacting turbine efficiency in order to increase performance. Many times neglecting non-uniform electromagnetic dispersion, existing procedures lead to energy losses, mechanical stress, and limited operational lifespan. Overcoming these challenges, the proposed system models electromagnetic field behavior using finite element analysis (FEA) and finds wrapping process irregularities. This simulation-based approach helps to optimize electromagnetic layout and reduce inefficiency by means of detailed modeling of field interactions within the turbine engine. This method allows engineers to manage design variables and materials to provide uniform field distribution and lower power loss. Reduced electromagnetic losses and enhanced mechanical stability enable the proposed framework to exhibit a significant gain in turbine efficiency. These findings confirm the feasibility of incorporating advanced modeling tools into wind turbine engine design for highest reliability and performance. With mechanical stability of 98.3%, turbine efficiency of 97.6%, dependability of 96.5%,
In aviation uses, particularly under high-temperature running circumstances, the performance and dependability of wire insulation are very crucial. Maintaining system safety and operational effectiveness in harsh flying conditions depends on long-term insulation integrity being maintained. But conventional assessment techniques may lack the accuracy to replicate continuous thermal stress and fail to identify early-stage insulation breakdown, thereby maybe resulting in system failures. This paper presents the Accelerated Aging and Dielectric Strength Testing (AA-DST) paradigm to go beyond these constraints. This approach more precisely and effectively evaluates insulation durability by combining high-temperature accelerated aging with thorough dielectric strength research. The AA-DST framework simulates real-world aeronautical thermal conditions, enabling a proactive evaluation of insulation breakdown thresholds. The proposed method is applied to a variety of wire insulation materials commonly used in aircraft electrical systems. By subjecting them to elevated temperatures and controlled stress conditions, the AA-DST framework identifies degradation trends and quantifies dielectric strength over time. Findings reveal that AA-DST provides enhanced predictive insights into insulation failure mechanisms, enabling improved material selection and design strategies. This contributes to safer, more reliable electrical systems in aerospace engineering. The proposed method achieves the predictive insights by 97.4%, reliability by 95.8%, material selection by 98.7%.
Accurate detection and tracking of low-altitude targets, such as drones, is a cornerstone of security in smart cities but is profoundly challenged by multipath propagation and atmospheric distortions in low-angle scenarios. This paper introduces a novel radar waveform design to overcome these limitations. Our method focuses on optimizing an Orthogonal Frequency Division Multiplexing (OFDM) waveform by shaping its Wideband Ambiguity Function (WAF) to approach an ideal, high-resolution profile. The optimization is driven by a Genetic Algorithm (GA) that efficiently navigates the complex parameter space to minimize a defined cost function. Simulation results demonstrate that our designed waveform achieves a significant performance leap over existing techniques. Key improvements include a 22 dB suppression of the first ambiguity sidelobe and a dramatic reduction in tracking error. Quantitative analysis shows our method achieves a 41.66% greater improvement in RMSE versus SNR and a 66.66% greater improvement in RMSE versus Range compared to a leading benchmark. This work establishes that GA-optimized OFDM waveforms are a powerful tool for enhancing radar resolution and tracking precision, directly addressing critical safety needs in modern urban surveillance systems.
In cognitive radio networks (CRNs), the rendezvous problem—which entails creating a shared control channel among secondary users (SUs) in a distributed manner— continues to be a major challenge. In highly dynamic spectrum situations, traditional channel-hopping and blind rendezvous algorithms frequently have inefficiency, scalability problems, and excessive latency (Time-to-Rendezvous, or TTR). This paper proposes a novel, joint sensing and rendezvous for efficient sharing of common channel within the practical constraints. In contrast to traditional sensing and rendezvous schemes, the joint sensing and rendezvous technique suggested in this research uses a partial number of channels for both sensing and rendezvous attempts. The challenge of finding vacant channels among the risk of occupancy is addressed by using the statistical distribution of unoccupied channels and rendezvous in terms of probability mass function and cumulative mass function. With the right statistical distribution, the proposed method shows a lower rendezvous time for a guaranteed rendezvous. Our simulation results show that the joint sensing and rendezvous model outperforms traditional distributed rendezvous strategies in terms of rendezvous time in the large-scale and dynamic CRNs.
This study investigates the impact of atmospheric clouds on surface solar radiation and the performance of photovoltaic (PV) panels in Asian countries. Clouds significantly reduce solar radiation and lower solar energy production efficiency. Data from MODIS satellite remote sensing and the Copernicus Atmosphere Monitoring Service (CAMS) were used to analyze changes in total cloud cover (TCC), low cloud cover (LCC), mid-level cloud cover (MCC), high cloud cover (HCC), global horizontal irradiance (GHI), and direct normal irradiance (DNI) from 2016 to 2024 in Asian regions. Results show a significant increase in total cloud cover in Pakistan (from 0.14 to 0.88), Aksai Chin (from 0.24 to 0.88), and Jammu and Kashmir (from 0.36 to 0.94), which reduced GHI and DNI. Low cloud cover increased in Aksai Chin (0.36) and Taiwan (0.33). Mid-level cloud cover rose in Aksai Chin (0.40) but decreased in Macau and Hong Kong. High cloud cover increased in Jammu and Kashmir (0.72) and Pakistan (0.67) but sharply declined in the Kuril Islands. These changes, driven by increased atmospheric humidity and climate patterns, blocked direct radiation and reduced solar panel efficiency by 30–50%.
The production of energy from renewable wind sources is significantly affected by dynamic changes in wind speed and force, environmental parameters, and turbine operating conditions. These factors play a key role in reliability studies and wind energy production forecasting. In this context, the integration of remote sensing (such as high-frequency radars and satellite data) with artificial neural networks (ANN) provides an effective tool for accurate wind force prediction, and numerous studies with various but related objectives have been conducted to estimate real-time reliability and energy production. This article offers a comprehensive review of the literature on the application of remote sensing and ANN in predicting wind behavior for renewable energy production. Special focus is placed on describing the scope of case studies (such as wind forecasting), ranging from simple ANN models to hybrid deep learning approaches, and key variables like wind speed, direction, and climatic data. This study highlights research that utilizes these technologies to predict reliability issues and develop preventive maintenance policies.
Solar power plant systems face complex nonlinear dynamics, which make accurate prediction challenging with traditional methods due to atmospheric fluctuations, solar radiation variations, and environmental uncertainties. Remote sensing and deep neural networks (such as CNN, RNN, and LSTM) enable the analysis of spatial-temporal data, which provides superior performance in predicting solar radiation for renewable energy production. These methods are crucial in advanced sensor systems for the design, prediction, maintenance, and control of solar power plants, and they offer greater safety, reliability, and efficiency compared to classical approaches. This article aims to review remote sensing and neural network technologies, their advantages (high accuracy, generalizability), and their limitations compared to traditional methods for solar radiation prediction. Unlike other reviews, this study summarizes adaptive intelligent models, proposes simple yet effective methods based on remote sensing and neural network sensor systems, maps the digital transformation to smart solar power plants with integrated technologies, and evaluates the impact of these technologies on the renewable energy value chain.
This study aims to identify and separate solar panels in South Asia using semantic segmentation and scale optimization with the MRS algorithm. The main goal is to process satellite images with a mosaic approach to detect photovoltaic panels. The study focuses on selecting the best scales in the MRS algorithm to improve separation accuracy. Scales of 2, 4, 6, 8, 10, 12, and 14 meters are tested. Key metrics such as Precision Segmentation Error (PSE), Noise-to-Signal Ratio (NSR), and Edge Discrepancy (ED) are used for evaluation. Results show that medium scales, like 6 meters, perform best for identifying individual panels, with a minimum ED of about 0.3. Larger scales, such as 200 meters or more, are better for analyzing groups of panels and estimating energy production. Variations in metrics across scales highlight the method’s sensitivity to scale selection. A weight of 0.7 is prioritized for spectral data and 0.3 for panel shape. Given the rapid growth of solar energy in South Asia, this approach enables fast mapping, energy production estimation, and efficient resource management. The findings emphasize that scale choice depends on the analysis goal (individual or group panels) and regional features. This method can support sustainable energy policies.
Photovoltaic (PV) solar farms serve as a renewable energy source and help reduce carbon emissions. They can cause changes in land use. These changes may have positive and negative effects on vegetation diversity and local ecosystems. Solar panels create shade and alter the microclimate, such as increasing soil moisture and lowering surface temperature. These changes can improve growth conditions for some plant species. Data from 2016 to 2024 in studied Asian countries show an average increase in 2-meter air temperature by 2 to 2.5 Kelvin. This trend correlates with a moderate improvement in the NDVI index (r=0.3–0.45), indicating positive effects of solar farms on microclimate regulation and vegetation cover. Overall, the data suggest increased solar potential alongside opportunities and challenges for ecological sustainability in arid and semi-arid regions of Asia.
This study evaluates the impact of photovoltaic (PV) power plants on land surface temperature (LST) using remote sensing techniques in Asian regions. The main goal is to assess the environmental effects of these power plants. Satellite infrared thermal images from the MODIS sensor for 2003 and 2023 were used to analyze temperature changes in large PV plants. Results show that PV plants significantly reduced the average daily LST. The cooling effect was 0.81°C during the day and 0.24°C at night. The cooling rate depends on the plant's capacity: -0.32°C for the daily average, -0.48°C for daytime, and -0.14°C for nighttime per terawatt-hour. Nighttime cooling correlates with geographic factors such as latitude, elevation, annual average temperature, precipitation, and solar radiation. This indicates that temperature effects depend on regional conditions, climate, and vegetation. Statistical analysis from 2016 to 2024 in Asian countries like Tajikistan, Ukraine, Kyrgyzstan, Kazakhstan, and Pakistan shows an increase in average 2-meter temperature (up to 6.27°C in Tajikistan). However, decreases were observed in Kazakhstan and Pakistan (up to 3.92°C). Global horizontal irradiation (GHI) increased in some areas but decreased in others, such as Aksai, China. These findings highlight the importance of remote sensing for sustainable energy management.
