Sliding Data Window Research Articles

Real-Time Ubiquitous Radar Target Classification with 1D ResNet-SE-Based Multi-Channel Network

Ubiquitous radar has significant advantages over traditional radar in detecting and identifying low, slow, and small (LSS) targets in a strong clutter environment. It effectively addresses challenges faced in low-altitude target monitoring within the low-altitude economy (LAE). The working mode of ubiquitous radar, which tracks first and identifies later, provides high-resolution Doppler data to the target identification module. Utilizing high-resolution Doppler data allows for the effective identification of LSS targets. To meet the needs of real-time classification, this paper first designs a real-time classification process based on sliding window Doppler data. This process requires the classifier to classify targets based on multiple rows of high-resolution Doppler spectra within the sliding window. Secondly, a multi-channel parallel perception network based on a 1D ResNet-SE network is designed. This network captures features within the rows of sliding window data and integrates inter-row features. Experiments show that the designed real-time classification process and multi-channel parallel perception network meet real-time classification requirements. Compared to the 1D CNN-MLP multi-channel network, the proposed 1D ResNet-MLP multi-channel network improves the classification accuracy from 98.71% to 99.34%. Integrating the 1D Squeeze-and-Excitation (SE) module to form the 1D ResNet-SE-MLP network further enhances accuracy to 99.58%, with drone target accuracy, the primary focus of the LAE, increasing from 97.19% to 99.44%.

Linear Discriminant Analysis for Predicting Net Blotch Severity in Spring Barley with Meteorological Data in Finland

Predictive information on plant diseases could help to reduce and optimize the usage of pesticides in agriculture. This research presents classification procedures with linear discriminant analysis to predict three possible severity levels of net blotch in spring barley in Finland. The weather data utilized for classification included mathematical transformations, namely features of outdoor temperature and air humidity with calculated dew point temperature values. Historical field observations of net blotch density were utilized as a target class for the identification of classifiers. The performance of classifiers was analyzed in sliding data windows of two weeks with selected, cumulative, summed feature values. According to classification results from 36 yearly data sets, the prediction of net blotch occurrence in spring barley in Finland can be considered as a linearly separable classification task. Furthermore, this can be achieved with linear discriminant analysis by combining the output probabilities of separate binary classifiers identified for each severity level of net blotch disease. In this case, perfect classification with a resolution of three different net blotch severity levels was achieved during the first 50 days from the beginning of the growing season. This strongly suggests that real-time classification based on a few weather variables measured on a daily basis can be applied to estimate the severity of net blotch in advance. This allows application of the principles of integrated pest management (IPM) and usage of pesticides only when there is a proven need.

PathEL: A novel collective entity linking method based on relationship paths in heterogeneous information networks

Collective entity linking always outperforms independent entity linking because it considers the interdependencies among entities. However, the existing collective entity linking methods often have high time complexity, do not fully utilize the relationship information in heterogeneous information networks (HIN) and most of them are largely dependent on the special features associated with Wikipedia. Based on the above problems, this paper proposes a novel collective entity linking method based on relationship path in heterogeneous information networks (PathEL). The PathEL classifies complex relationships in HIN into 1-hop paths and 3 types of 2-hop paths, and measures entity correlation by the path information among entities, ultimately combining textual semantic information to realize collective entity linking. In addition, facing the high complexity of collective entity linking, this paper proposes to solve the problem by combining the variable sliding window data processing method and the two-step pruning strategy. The variable sliding window data processing method limits the number of entity mentions in each window and the pruning strategy reduces the number of candidate entities. Finally, the experimental results of three benchmark datasets verify that the model proposed in this paper performs better in entity linking than the baseline models. On the AIDA CoNLL dataset, compared to the second-ranked model, our model has improved P, R, and F1 scores by 1.61%, 1.54%, and 1.57%, respectively.

Fault diagnosis method for redundancy sensors of electric actuator

In order to solve the singular fault modes of redundant position sensors in electric actuators, a fault diagnosis method based on Hall sensor analytical model is proposed on the basis of traditional redundancy comparison algorithm. Through the complete self-monitoring design of Hall sensor and the analytical model design, non-similar redundancy signals of sensors are formed. At the same time, a hypothesis testing method based on sliding data window is adopted, and by adjusting the width of the sliding window to meet the indicators of false alarm rate and missed alarm rate, a fault diagnosis algorithm for position sensors is formed as a whole. Finally, an algorithm validation was conducted for the 2∶2 singular fault mode of the position sensor. The results showed that the fault diagnosis algorithm based on the Hall sensor model can effectively solve the problems caused by 2∶2 singular faults, greatly improve the availability of sensor signals, and ensure the control accuracy and fault tolerance of the servo closed-loop.

Real-Time Intelligent Diagnosis of Co-frequency Vibration Faults in Rotating Machinery Based on Lightweight-Convolutional Neural Networks

The co-frequency vibration fault is one of the common faults in the operation of rotating equipment, and realizing the real-time diagnosis of the co-frequency vibration fault is of great significance for monitoring the health state and carrying out vibration suppression of the equipment. In engineering scenarios, co-frequency vibration faults are highlighted by rotational frequency and are difficult to identify, and existing intelligent methods require more hardware conditions and are exclusively time-consuming. Therefore, Lightweight-convolutional neural networks (LW-CNN) algorithm is proposed in this paper to achieve real-time fault diagnosis. The critical parameters are discussed and verified by simulated and experimental signals for the sliding window data augmentation method. Based on LW-CNN and data augmentation, the real-time intelligent diagnosis of co-frequency is realized. Moreover, a real-time detection method of fault diagnosis algorithm is proposed for data acquisition to fault diagnosis. It is verified by experiments that the LW-CNN and sliding window methods are used with high accuracy and real-time performance.

Optimized CNN‐Bi‐LSTM–Based BCI System for Imagined Speech Recognition Using FOA‐DWT

Speech imagery is emerging as a significant neuro‐paradigm for designing an electroencephalography (EEG)‐based brain–computer interface (BCI) system for the purpose of rehabilitation, medical neurology, and to aid people with disabilities in interacting with their surroundings. Neural correlates of speech imagery EEG signals are variable and weak as compared to the vocal state; hence, it is challenging to interpret them using machine learning (ML)–based classifiers. The applicability of modern deep learning methods such as convolutional neural networks (CNNs) and bidirectional long short‐term memory (Bi‐LSTM) networks has seen substantial advances in complex EEG signal analysis as compared to ML‐based methods. The objective of this article is to design a firefly‐optimized discrete wavelet transform (DWT) and CNN‐Bi‐LSTM–based imagined speech recognition (ISR) system to interpret imagined speech EEG signals. This study utilizes two publicly available datasets. EEG signal is enhanced using firefly optimization algorithm (FOA)–based optimized soft thresholding of high‐frequency detail components obtained by DWT decomposition. The enhanced EEG signal is augmented using sliding window data augmentation to increase the training data. Frequency‐domain features like power spectral density (PSD), frequency band power (FBP), band ratios, peak frequency, mean frequency, median frequency, spectral entropy, and relative power are extracted from augmented EEG segments. The extracted feature vector is fed to the designed CNN‐Bi‐LSTM classifier such that the EEG data are classified into two‐class, three‐class, and four‐class categories. To achieve optimal performance, the CNN‐Bi‐LSTM model was optimized using the Keras tuner library. The designed CNN consists of one‐dimensional (1‐D) convolutional layers and max pooling layers for familiarizing local associations along with mining hierarchical connections, and the Bi‐LSTM network acquires long‐term dependencies from the features learned by the former CNN. Bi‐LSTM network improves the performance and acquires potentially more affluent representations by looking at the sequence in both forward and reverse ways to capture representations that might be left unexploited by the sequential‐order kind alone. The performance of the designed FOA‐DWT‐CNN‐Bi‐LSTM–based ISR system is assessed using four evaluation measures: accuracy, F1 score, recall, and precision. It is found that the proposed system achieves the highest classification accuracy of 99.43 ± 2.5%, 94.41 ± 3.31%, and 89.57 ± 4.3% for two‐class, three‐class, and four‐class categories, respectively.

Ambient Data-Driven Online Tracking of Electromechanical Modes Using Recursive Subspace Dynamic Mode Decomposition

The extraction of electromechanical modal parameters from ambient data is a useful and practical method that can monitor the modal properties of power system oscillations online. This paper proposes a recursive subspace dynamic mode decomposition (RSub-DMD) algorithm for online monitoring power system modes using wide-area synchronized ambient data. By introducing Givens rotation to the recursion process, the computational capability of subspace dynamic mode decomposition has been significantly improved without compromising accuracy. The shorter sliding data window enables the proposed RSub-DMD to quickly track the electromechanical modal parameters and participation factors (PFs), with low estimation variance. IEEE 16 generator 68 bus test system and measurement data from real system are used to verify the performance of the proposed RSub-DMD algorithm. The tracking results in different operating environments verify the effectiveness of the proposed algorithm.

Non-unit ultra-high-speed DC line protection method for HVDC grids using dynamic fitting algorithm with data reconstruction

Existing non-unit protection schemes are inevitably influenced by the high fault-resistance in MMC-HVDC grid, which is a serious problem affecting the rapidity and reliability of fault recognition. This study proposes a dynamic fitting of data reconstruction-based protection method to improve the accuracy and rapidity of HVDC grid protection. Due to the propagation characteristics of the initial voltage traveling wave being unaffected by the transition resistance, the self-adapting exponential fitting algorithm is used to recognize the fault waveform. Then, sliding data window with the voltage traveling wave fit by the optimal basis exponential fitting algorithm could avoid the influence of initial value selection on fitting convergence. Meanwhile, the original fault traveling wave is reconstructed by “Bayesian compressed sensing algorithm (BCSA)” in a low sampling rate. This method could reduce the operation response time of the proposed method to satisfy the ultra-high-speed requirement of MMC-HVDC grid and also adapt the limitation of sampling rate in existing protection devices. Practical case studies in PSCAD/EMTDC and protection prototype testing have demonstrated that the proposed method is effective under different fault locations, fault types, and high fault impedance.

A Wavelet Transform-Based Real-Time Filtering Algorithm for Fusion Magnet Power Signals and Its Implementation

With high real-time performance and a good filtering effect, the wavelet transform-based real-time filter is more commonly used in the fusion magnet power supply. Considering the signal characteristics of the fusion magnet power supply, the sliding time window method is widely used to satisfy the requirements of filtering efficiency. In this paper, an updated half-edge extension method is proposed to improve the boundary effect of sliding window data. Moreover, the hardware-based wavelet loopback structure is proposed to avoid the double computation of long sliding window data, thereby improving real-time performance and satisfying the demands of fusion magnet power supply filtering. The proposed half-edge extension method and the wavelet loopback structure are implemented by designed FPGA and experimented on the Experimental and Advanced Superconducting Tokamak (EAST) fusion magnet power supply system to verify the filtering efficiency. According to the experiment’s results, the proposed half-edge extension method has a better filtering effect, and the proposed wavelet loopback structure decreases the time by 67.4%, which can satisfy the requirements of the power supply system.

A Time Series Forecasting Model Selection Framework using CNN and Data Augmentation for Small Sample Data

The key to the accuracy of time series forecasting is to find the most appropriate forecasting method. Therefore, the forecasting model selection of time series has become a new research hotspot in the data analysis field. However, most of the existing meta learning forecasting model selection methods rely on manual selection of features, which leads to low efficiency and lack of objectivity. Therefore, this paper proposes an improved meta learning framework for deep learning time series forecasting model selection. Inspired by computer vision, we transform one -dimensional time series into two-dimensional images, and use convolution neural network to train and classify time series images (model selection). Moreover, in order to deal with the over fitting problem caused by small sample datasets, the sliding window data augmentation method is used to improve the accuracy of small datasets model selection. The large-scale empirical study on M3 data sets shows that the framework has better model selection accuracy and smaller forecasting error than the recurrent neural network (RNN), back propagation neural network and traditional time series image algorithms. In addition, compared with the traditional time series image method, RNN and BP, the classification rate (model selection accuracy) of this algorithm is improved by 6.5%, 4.4% and 3.2%, respectively.

Analisis Akurasi Prediksi Perubahan Aktivitas Pada Sistem Monitoring Aktivitas Jarak Jauh Pasien Isolasi Mandiri Berbasis IOT

Patients who contract the disease should avoid contact with other people. One way to do this is to self-isolate at home. The family of the patient who cares for the activities that are carried out in self-isolation to find out the condition of the patient's condition, his condition is improving or deteriorating. To avoid direct contact, the patient's activity, independently, can be monitored by remotely predicting changes in patient activity using an Internet of Things-based remote monitoring system for self-isolating patient activities. This cellular-based monitoring system uses an accelerometer sensor to retrieve data on changes in patient activity and analyzes the effect of several variations in the number of data samples and sliding-windows on the accuracy of the system in predicting changes in patient activity. Variations in the number of N samples tested were 4,6,8,10,20,30,40,50,60,70,80,90 and 100 samples, while the sliding-window N variation tested was 1 ,2,3,4,5,6,7,8,9 and 10 samples where there is a change in activity every 30 seconds for 330 seconds (10 changes in activity) for each number of N samples and N sliding windows. The results shown are N sample data = 6 providing the highest activity change prediction accuracy, amounting to 90.15%, while N sliding window data = 6 providing the highest activity change prediction accuracy, amounting to 92.72%.

A hybrid data-driven method for rapid prediction of lithium-ion battery capacity

The quick and accurate prediction of future capacity is a challenging issue in the field of battery health management. To solve this problem, the paper proposes a data-driven model based on advanced machine-learning techniques. First, wavelet transform is used to denoise the initial features, and the relatively important features are selected with the help of random forest. Second, the paper uses the approximate entropy theory to construct multi-time scale sliding window data to solve the problem that the sliding window size can not be determined in traditional methods. Third, in order to overcome the disadvantages of traditional methods, such as the comparative inefficiency in converging and using temporal contextual information, the paper proposes a novel mogrifier long short-term memory network denoted as QGA-ASM-LSTM, which is based on attention mechanism and similarity judgment mechanism, using quantum genetic algorithm to optimize its parameters. Finally, the proposed data-driven method is used to model the degradation of lithium-ion batteries and finish the fast capacity prediction. The results of experiments on different public datasets demonstrate that the performance of QGA-ASM-LSTM exceeds the state-of-the-art models in terms of prediction accuracy and training time.

Sliding-Window Pitch-Class Histograms as a Means of Modeling Musical Form

This study proposes a fine-tuned approach to modeling musical form in Classical repertoires by analyzing pitch-class distributions in symbolic data using machine learning algorithms. Results suggest that sliding-window histograms, which take the temporal component into account, palpably improve algorithms’ performance in assigning form labels to pieces. Pitch-class histograms were extracted from major-mode piano-sonata movements by W. A. Mozart and L. v. Beethoven according to two methods: whole-piece histograms and sliding-window histograms. In the latter method, richer features were obtained by calculating histograms for 9–90 partially overlapping windows per piece. Supervised learning methods, such as Support Vector Machines (SVMs) and Artificial Neural Networks (ANNs), yield a good separation of data points according to three distinct form labels: sonata form, ABA (ternary) form, and variation form. For both models, using sliding-window histograms significantly improves performance with regard to the whole-piece histogram data. Unsupervised clustering into three components employing a Gaussian Mixture Model (GMM), on the other hand, yields no successful results. Finally, we offer an in-depth exploration of our findings, and propose some directions for further research based on sliding-window data.

Research on Fan Operation Evaluation and Error State Judgment Relying on Improved Neural Network and Intelligent Computing

Fan, as the most commonly used mechanical equipment, is widely used. In order to solve the problem of fan bearing fault diagnosis, this paper analyzes the main factors affecting fan spindle speed and power generation in operation. The input and output parameters of the performance prediction model are determined. The performance prediction model of wind turbine is established by using generalized regression neural network, and the smoothing factor of GRNN is optimized by comparing the prediction accuracy of the model. Based on this model, the sliding data window method is used to calculate the residual evaluation index of wind turbine speed and power in real time. When the evaluation index continuously exceeds the pre-set threshold, the abnormal state of wind turbine can be judged. In order to obtain wind turbine blades with better aerodynamic performance, a blade aerodynamic performance optimization method based on quantum heredity is proposed. The B é zier curve control point is used as the design variable to represent the continuous chord length and torsion angle distribution of the blade, the blade shape optimization model aiming at the maximum power is established, and the quantum genetic algorithm is used to optimize the chord length and torsion angle of the blade under different constraints. The optimization results of quantum genetic algorithm and classical genetic algorithm are compared and analyzed. Under the same parameters and boundary conditions, the proposed blade aerodynamic optimization method based on quantum genetic optimization is better than the classical genetic optimization method, and can obtain better blade aerodynamic shape and higher wind energy capture efficiency. This method makes up for the shortcomings of traditional fault diagnosis methods, improves the recognition rate of fault types and the accuracy of fault diagnosis, and the diagnosis effect is good.

A Sliding Window Data Compression Method for Spatial-Time DOA Estimation

This paper presents a sliding window data compression method for spatial-time direction-of-arrival (DOA) estimation using coprime array. The signal model is firstly formulated by jointly using the temporal and spatial information of the impinging sources. Then, a sliding window data compression processing is performed on the array output matrix to realize fast calculation of time average function, and the computational burden has been reduced accordingly. Based on the concept of sum and difference co-array (SDCA), the vectorized conjugate augmented MUSIC is adopted, with which more sources than twice of the physical sensors can be resolved. Additionally, the sparse array robustness to sensor failure has been evaluated by introducing the concept of essential sensors. The theoretical analysis and numerical simulations are provided to confirm the effectiveness performance of the proposed method.

Gait cycle prediction model based on gait kinematic using machine learning technique for assistive rehabilitation device

The gait cycle prediction model is critical for controlling assistive rehabilitation equipment like orthosis. The human gait model has recently used statistical models, but the dynamic properties of human physiology limit the current approach. Current human gait cycle prediction models need detailed kinematic and kinetic data of the human body as input parameters, and measuring them requires special instruments, making them difficult to use in real-world applications. In our study, three separate machine learning algorithms were used to create a human gait model: Gaussian process regression, support vector machine, and decision tree. The algorithm used to create the model's input parameters are height, weight, hip and knee angle, and ground reaction force (GRF). For better gait cycle model prediction, the models produced were enhanced by incorporating different sliding window data. The best gait period prediction model was DT with sliding window data (t−3), which had a root mean square error of 3.3018 and the R-squared (R-Value) of 0.97. The projection model focused on hip and knee angle and GRF was a feasible solution to controlling assistive rehabilitation devices during the gait cycle.

Anomaly Detection Using a Sliding Window Technique and Data Imputation with Machine Learning for Hydrological Time Series

Water level data obtained from telemetry stations typically contains large number of outliers. Anomaly detection and a data imputation are necessary steps in a data monitoring system. Anomaly data can be detected if its values lie outside of a normal pattern distribution. We developed a median-based statistical outlier detection approach using a sliding window technique. In order to fill anomalies, various interpolation techniques were considered. Our proposed framework exhibited promising results after evaluating with F1-score and root mean square error (RMSE) based on our artificially induced data points. The present system can also be easily applied to various patterns of hydrological time series with diverse choices of internal methods and fine-tuned parameters. Specifically, the Spline interpolation method yielded a superior performance on non-cyclical data while the long short-term memory (LSTM) outperformed other interpolation methods on a distinct tidal data pattern.

Estimating the short-time rate of change in the trend of the Keeling curve

What exactly is the short-time rate of change (growth rate) in the trend of text {CO}_2 data such as the Keeling curve? The answer to this question will obviously depend very much on the duration in time over which the trend has been defined, as well as the smoothing technique that has been used. As an estimate of the short-time rate of change we propose to employ a very simple and robust definition of the trend based on a centered 1-year sliding data window for averaging and a corresponding centered 1-year difference (2-year data window) to estimate its rate of change. In this paper, we show that this simple strategy applied to weekly data of the Keeling curve (1974–2020) gives an estimated rate of change which is perfectly consistent with a more sophisticated regression analysis technique based on Taylor and Fourier series expansions. From a statistical analysis of the regression model and by using the Cramér–Rao lower bound, it is demonstrated that the relative error in the estimated rate of change is less than 5 %. As an illustration, the estimates are finally compared to some other publicly available data regarding anthropogenic text {CO}_2 emissions and natural phenomena such as the El Niño.

Self-Adaptive Gathering for Energy-Efficient Data Stream in Heterogeneous Wireless Sensor Networks Based on Deep Learning

Big data streams are available across the growing heterogeneous wireless sensor networks, with characteristics of vast volume and dynamic transmission. Energy efficiency improvement in big data stream gathering is becoming a challenge. In this article, a self-adaptive gathering algorithm for multisource heterogeneous big data streams with sliding windows is proposed, which can improve the energy efficiency for data stream processing due to adaptively adjusting the window size based on the difference of data probability distribution between adjacent windows. In order to save the spatial correlation of a heterogeneous data stream, the Gaussian Bernoulli Matrix Variable Restricted Boltzmann Machine (GBMVRBM) is proposed to deal with the multi-source data separately, and then a joint layer is used to fuse the data features of different modalities. The probability distribution of sliding window data is obtained by the energy function of the GBMVRBM, and the Hoeffding boundary is adopted to ensure that the probability distribution variation between the windows can be detected in time. The algorithm is tested on the Clemson University Audio Visual Experiments database, and it can be concluded that the algorithm proposed in this article can not only detect the data change in time, but also expand the window size to process the data efficiently.

Workload-aware wavelet synopses for sliding window aggregates

In this work, we study the problem of maintaining basic aggregate statistics over a sliding-window data stream under the constraint of limited memory. As in IoT scenarios the available memory is typically much less than the window size, queries are answered from compact synopses that are maintained in an online fashion. For the efficient construction of such synopses, we propose wavelet-based algorithms that provide deterministic guarantees and produce near exact results for a variety of data distributions. Furthermore, we show how accuracy can be further improved when workload information is known. For this purpose, we propose a workload-aware streaming system that trade-offs accuracy with synopsis’ construction throughput. The conducted experiments indicate that with only a $$15\%$$ penalty in throughput, the proposed system produces fairly accurate results even for the most adversarial distributions.