Decomposition Entropy Research Articles

Research on Denoising Method for Hydroelectric Unit Vibration Signal Based on ICEEMDAN-PE-SVD.

Vibration monitoring and analysis play a crucial role in the fault diagnosis of hydroelectric units. However, accurate extraction and identification of fault features from vibration signals are challenging because of noise interference. To address this issue, this study proposes a novel denoising method for vibration signals based on improved complementary ensemble empirical mode decomposition with adaptive noise (ICEEMDAN), permutation entropy (PE), and singular value decomposition (SVD). The proposed method is applied for the analysis of hydroelectric unit sway monitoring. Firstly, the ICEEMDAN method is employed to process the signal and obtain several intrinsic mode functions (IMFs), and then the PE values of each IMF are calculated. Subsequently, based on a predefined threshold of PE, appropriate IMFs are selected for reconstruction, achieving the first denoising effect. Then, the SVD is applied to the signal after the first denoising effect, resulting in the SVD spectrum. Finally, according to the principle of the SVD spectrum and the variation in the singular value and its energy value, the signal is reconstructed by choosing the appropriate reconstruction order to achieve the secondary noise reduction effect. In the simulation and case analysis, the method is better than the commonly used wavelet threshold, SVD, CEEMDAN-PE, and ICEEMDAN-PE, with a signal-to-noise ratio (SNR) improvement of 6.9870 dB, 4.6789 dB, 8.9871 dB, and 4.3762 dB, respectively, and where the root-mean-square error (RMSE) is reduced by 0.1426, 0.0824, 0.2093 and 0.0756, respectively, meaning that our method has a better denoising effect and provides a new way for denoising the vibration signal of hydropower units.

Fault-Diagnosis Method for Rotating Machinery Based on SVMD Entropy and Machine Learning

Rolling bearings and gears are important components of rotating machinery. Their operating condition affects the operation of the equipment. Fault in the accessory directly leads to equipment downtime or a series of adverse reactions in the system, which brings enormous pecuniary loss to the institution. Hence, it is of great significance to detect the operating status of rolling bearings and gears for fault diagnosis. At present, the vibration method is considered to be the most common method for fault diagnosis, a method that analyzes the equipment by collecting vibration signals. However, rotating-machinery fault diagnosis is challenging due to the need to select effective fault feature vectors, use appropriate machine-learning classification methods, and achieve accurate fault diagnosis. To solve this problem, this paper illustrates a new fault-diagnosis method combining successive variational-mode decomposition (SVMD) entropy values and machine learning. First, the simulation signal and the real fault signal are used to analyze and compare the variational-mode decomposition (VMD) and SVMD methods. The comparison results prove that SVMD can be a useful method for fault diagnosis. Then, these two methods are utilized to extract the energy entropy and fuzzy entropy of the gearbox dataset of Southeast University (SEU), respectively. The feature vector and multiple machine-learning classification models are constructed for failure-mode identification. The experimental-analysis results successfully verify the effectiveness of the combined SVMD entropy and machine-learning approach for rotating-machinery fault diagnosis.

Research on Ultra-Wideband Radar Life Detection Algorithm Based on SE and SSD

Using pulsed Ultra-Wideband Radar non-contact detection technology, when detecting vital sign signals, the weak vital sign echo signals are often covered by various noises, resulting in the inability to identify and locate human targets. To solve this problem, an Ultra-Wideband Radar life detection algorithm based on the combination of sample entropy and singular spectrum decomposition is proposed. The human target position information is determined according to the SE value of the radar received pulse. In addition, the respiration and heart rate information were obtained using the SSD method. 15 groups of experiments were carried out, and the echo radar signals of 5 volunteers at 3 distances were collected respectively. The measured data analysis results show that the proposed algorithm can accurately and effectively extract the distance and vital signs information of human targets and has extensive research value and application prospects.

A multi-head residual connection GCN for EEG emotion recognition

Electroencephalography (EEG) emotion recognition is a crucial aspect of human-computer interaction. However, conventional neural networks have limitations in extracting profound EEG emotional features. This paper introduces a novel multi-head residual graph convolutional neural network (MRGCN) model that incorporates complex brain networks and graph convolution networks. The decomposition of multi-band differential entropy (DE) features exposes the temporal intricacy of emotion-linked brain activity, and the combination of short and long-distance brain networks can explore complex topological characteristics. Moreover, the residual-based architecture not only enhances performance but also augments classification stability across subjects. The visualization of brain network connectivity offers a practical technique for investigating emotional regulation mechanisms. The MRGCN model exhibits average classification accuracies of 95.8% and 98.9% for the DEAP and SEED datasets, respectively, highlighting its excellent performance and robustness.

The microbiome of an outpatient rehabilitation clinic and predictors of contamination: A pilot study.

Understanding sources of microbial contamination in outpatient rehabilitation (REHAB) clinics is important to patients and healthcare providers. The purpose of this study was to characterize the microbiome of an outpatient REHAB clinic and examine relationships between clinic factors and contamination. Forty commonly contacted surfaces in an outpatient REHAB clinic were observed for frequency of contact and swiped using environmental sample collection kits. Surfaces were categorized based on frequency of contact and cleaning and surface type. Total bacterial and fungal load was assessed using primer sets specific for the 16S rRNA and ITS genes, respectively. Bacterial samples were sequenced using the Illumina system and analyzed using Illumina-utils, Minimum Entropy Decomposition, QIIME2 (for alpha and beta diversity), LEfSe and ANCOM-BC for taxonomic differential abundance and ADONIS to test for differences in beta diversity (p<0.05). Porous surfaces had more bacterial DNA compared to non-porous surfaces (median non-porous = 0.0016ng/μL, 95%CI = 0.0077-0.00024ng/μL, N = 15; porous = 0.0084 ng/μL, 95%CI = 0.0046-0.019 ng/μL, N = 18. p = 0.0066,DNA. Samples clustered by type of surface with non-porous surfaces further differentiated by those contacted by hand versus foot. ADONIS two-way ANOVA showed that the interaction of porosity and contact frequency (but neither alone) had a significant effect on 16S communities (F = 1.7234, R2 = 0.0609, p = 0.032). Porosity of surfaces and the way they are contacted may play an underestimated, but important role in microbial contamination. Additional research involving a broader range of clinics is required to confirm results. Results suggest that surface and contact-specific cleaning and hygiene measures may be needed for optimal sanitization in outpatient REHAB clinics.

Recurrence Interval Analysis of the US Bitcoin Market

We considered the daily price dynamics of the US Bitcoin market in the period from 2015 to 2022. In the first step, we used a singular value decomposition (SVD) entropy method for assessing time-varying informational efficiency over different time scales, from weeks to quarters. It was shown that the US Bitcoin market has been informationally efficient most of the time, except for some isolated periods where the returns exhibited deviations from the random behavior. The COVID-19 pandemic has not impacted the informational efficiency. This suggests that the Bitcoin market is unpredictable, and no reliable predictions can be obtained. A further analysis was carried out by considering the recurrence intervals for different positive and negative returns. We found that the distribution of recurrence intervals for positive and negative returns is asymmetric, with mean values higher for negative returns. We found that the distribution of recurrence intervals can be described by a stretching exponential distribution, such that the empirical and analytical hazard probabilities as functions of the elapsed time show good agreement.

Contribution of Nonlinear Dynamics to the Informational Efficiency of the Bitcoin Market

The recent decade has witnessed a surge of cryptocurrency markets as innovative financial systems based strongly on digital emission, interchange and coding. The main characteristic is that cryptocurrencies are not subjected to the regulation of governments and financial institutions (e.g., central banks), such that their dynamics are determined solely by non-centralized mechanisms. Informational efficiency is a key issue for cryptocurrency markets since its fulfillment guarantees that all participants have access to the same information quality and that arbitrage conditions are discarded. This study evaluated the contribution of nonlinearities to the informational efficiency of the Bitcoin market for the period 2014–2022. Singular value decomposition (SVD) entropy together with shuffled and phase-randomized data in a rolling-window framework was used to capture randomness and nonlinear dynamics in Bitcoin returns. It was found that the contribution of nonlinearities to informational efficiency increases with the time scale, with a mean contribution of about 7.25% for long-time scales. This means that the Bitcoin market is only affected by weak nonlinearities, although these effects should be considered for forecasting and valuation.

A Behavior Sequence Based Psychological Feature Extraction Model for Students in English Teaching

With the rapid development of science and technology, intelligent teaching has been gradually popularized in the campus, and the change of relevant teaching modes has also been carried out. In the new English teaching, it is necessary to analyze the students’ psychological behavior. In order to get accurate results, the research captures the students’ behavior as a behavior sequence, and uses the improved combined kernel support vector machine to extract features on the basis of approximate entropy and local mean decomposition. At the same time, it is compared with the traditional support vector machine, the traditional least square vector machine and the back-propagation neural network algorithm. The experimental results show that the improved algorithm has higher accuracy, recall and accuracy than the other three algorithms. The overall accuracy of the improved algorithm is 82.53%, the highest accuracy is 90.12%, the overall recall is 58.64% and the overall precision is 86.92%. The experimental results show that the improved algorithm has a significant improvement in performance, and can be applied to the extraction of students’ psychological characteristics. Therefore, the algorithm can be used in conventional blended English teaching, which is convenient for teachers to carry out English teaching more efficiently.

Robotic Milling Chatter Types Detection Based on Adaptive VariationalMode Decomposition and Difference of Power Spectral Entropy

Robotic Milling Chatter Types Detection Based on Adaptive VariationalMode Decomposition and Difference of Power Spectral Entropy

HURST EXPONENT ESTIMATION FOR SHORT-TIME SERIES BASED ON SINGULAR VALUE DECOMPOSITION ENTROPY

Complex time series appear commonly in a large diversity of the science, engineering, economy, financial and social fields. In many instances, complex time series exhibit scaling behavior over a wide range of scales. The traditional rescaled-range (R/S) analysis and the detrended fluctuation analysis (DFA) are commonly used to characterize the scaling behavior via the Hurst exponent. These methods perform well for long-time series. However, the performance may be poor for short times resulting from scarce measurements (e.g. less than a hundred). This work proposes an approach based on singular value decomposition (SVD) entropy for estimating the Hurst exponent for short-time series. In the first step, synthetic time series were used to find the relationship between Hurst exponent and SVD entropy. In the second step, an empirical relationship was proposed to estimate the Hurst exponent from SVD entropy computations of the time series. The performance of the approach was illustrated with two examples of real-time series (consumer price index (CPI) and El Niño Oceanic Index), showing that the estimated Hurst exponent provides valuable insights into the physical mechanisms involved in the generation of the time series.

Feature and texture distillation via neural network training

Encoded-diffraction hybrid systems—optical encoding and simple electronic decoding—offers feature distillation via model training. Additionally, the most faithfully reconstructed images are not the ones that are best classified. We parametrize our results with singular value decomposition (SVD) entropy, a proxy for image complexity.

An Iterative Warping and Clustering Algorithm to Estimate Multiple Wave-Shape Functions From a Nonstationary Oscillatory Signal

Nonsinusoidal oscillatory signals are everywhere. In practice, the nonsinusoidal oscillatory pattern, modeled as a 1-periodic wave-shape function (WSF), might vary from cycle to cycle. When there are finite different WSFs, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$s_{1},\ldots,s_{K}$</tex-math></inline-formula> , so that the WSF jumps from one to another suddenly, the different WSFs and jumps encode useful information. We present an iterative warping and clustering algorithm to estimate <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$s_{1},\ldots,s_{K}$</tex-math></inline-formula> from a nonstationary oscillatory signal with time-varying amplitude and frequency, and hence the change points of the WSFs. The algorithm is a novel combination of time-frequency analysis, singular value decomposition entropy and vector spectral clustering. We demonstrate the efficiency of the proposed algorithm with simulated and real signals, including the voice signal, arterial blood pressure, electrocardiogram and accelerometer signal. Moreover, we provide a mathematical justification of the algorithm under the assumption that the amplitude and frequency of the signal are slowly time-varying and there are finite change points that model sudden changes from one wave-shape function to another one.

Monitoring Level of Hypnosis using Stationary Wavelet Transform and Singular Value Decomposition Entropy with Feedforward Neural Network.

Classifying the patient's depth of anesthesia (LoH) level into a few distinct states may lead to inappropriate drug administration. To tackle the problem, this paper presents a robust and computationally efficient framework that predicts a continuous LoH index scale from 0-100 in addition to the LoH state. This paper proposes a novel approach for accurate LoH estimation based on Stationary Wavelet Transform (SWT) and fractal features. The deep learning model adopts an optimized temporal, fractal, and spectral feature set to identify the patient sedation level irrespective of age and the type of anesthetic agent. This feature set is then fed into a multilayer perceptron network (MLP), a class of feed-forward neural networks. A comparative analysis of regression and classification is made to measure the performance of the chosen features on the neural network architecture. The proposed LoH classifier outperforms the state-of-the-art LoH prediction algorithms with the highest accuracy of 97.1% while utilizing minimized feature set and MLP classifier. Moreover, for the first time, the LoH regressor achieves the highest performance metrics (R2 = 0.9, MAE = 1.5) as compared to previous work. This study is very helpful for developing highly accurate monitoring for LoH which is important for intraoperative and postoperative patients' health.

필터 다양화를 통한 합성곱 신경망의 표현력 향상

This paper aims to improve the feature representation by diversifying CNN filters inspired by niche concept in evolution. The singular value decomposition (SVD) entropy based efficient metric for diversity is proposed In the proposed approach, filters are clustered by groups and they are calculated as differences from the center values within the groups, rather than by entire rank based comparison. This provides an effective method for increasing the substantial diversity of filters. Furthermore, the filters with low diversity are adjusted by the diversity spreading framework for better diversity in the reconstruction process. The improvement of the filter representation by performing experiments on CIFAR 10/100 data for VGG16, and ImageNet for ResNet34 is provided. Because there are no similar studies, we compare our results with respect to those of relatively relevant pruning methods in terms of classification performance accuracy as well as the pruned rates and flops.

Entropy Approximation by Machine Learning Regression: Application for Irregularity Evaluation of Images in Remote Sensing

Approximation of entropies of various types using machine learning (ML) regression methods are shown for the first time. The ML models presented in this study define the complexity of the short time series by approximating dissimilar entropy techniques such as Singular value decomposition entropy (SvdEn), Permutation entropy (PermEn), Sample entropy (SampEn) and Neural Network entropy (NNetEn) and their 2D analogies. A new method for calculating SvdEn2D, PermEn2D and SampEn2D for 2D images was tested using the technique of circular kernels. Training and testing datasets on the basis of Sentinel-2 images are presented (two training images and one hundred and ninety-eight testing images). The results of entropy approximation are demonstrated using the example of calculating the 2D entropy of Sentinel-2 images and R2 metric evaluation. The applicability of the method for the short time series with a length from N = 5 to N = 113 elements is shown. A tendency for the R2 metric to decrease with an increase in the length of the time series was found. For SvdEn entropy, the regression accuracy is R2 &gt; 0.99 for N = 5 and R2 &gt; 0.82 for N = 113. The best metrics were observed for the ML_SvdEn2D and ML_NNetEn2D models. The results of the study can be used for fundamental research of entropy approximations of various types using ML regression, as well as for accelerating entropy calculations in remote sensing. The versatility of the model is shown on a synthetic chaotic time series using Planck map and logistic map.

Data analytics and throughput forecasting in port management systems against disruptions: a case study of Busan Port

Forecasting cargo throughput is an essential albeit challenging task in ensuring efficient seaport management. In this study, data analytics is employed to analyze the nonlinear dynamic behaviors, as well as disruptions in port throughputs. Further, nonlinear analytical methods, including the Lyapunov exponent (LE), information entropy, Hurst exponent, and wavelet decomposition, are employed to explore the complex dynamic behavior of port throughput under supply chain disruptions. By employing the discrete wavelet transform (DWT) and the long short-term memory (LSTM) network, we develop a novel hybrid model of port throughput forecasting. DWT is employed to decompose the original data into a finite set of frequency components, so that the various hidden features of cargo throughput can be extracted via different modes, such as the trend, residual, and seasonal components. Thereafter, each component, obtained from the DWT spectra, is predicted via a machine learning model. Additionally, hypothesis testing, model evaluation, and statistical significance tests are employed to comprehensively evaluate the introduced forecasting models. Regarding prediction accuracy and efficiency, our extensive simulation results confirm the superiority of the hybrid strategy over five benchmarked models. Finally, by employing business forecasting software, we show that the robust hybrid strategy achieves accurate predictions of port throughputs against market disruptions. Our findings can help decision-makers understand disruption mechanisms in port systems, thus enabling them to successfully achieve their business goals.

Are Torque-Induced Bone Texture Alterations Related to Early Marginal Jawbone Loss?

The reason why marginal bone loss (MBL) occurs after dental implant insertion without loading has not yet been clearly investigated. There are publications that confirm or reject the notion that there are factors that induce marginal bone loss, but no research investigates what exactly occurs in the bone surrounding the implant neck. In this study, 2196 samples of dental implant neck bone radiographs were analyzed. The follow-up period was 3 months without functional loading of the implant. Marginal bone loss was evaluated in relation to the torque used during the final phase of implant insertion. Radiographic texture features were also analyzed and evaluated. The analyses were performed individually for the anterior and posterior part of the alveolar crest in both the mandible and maxilla. After 3 months, an MBL relation with higher torque (higher than 40 Ncm; p < 0.05) was observed, but only in the lower jaw. The texture features Sum Average (SumAverg), Entropy, Difference Entropy (DifEntr), Long-Run Emphasis (LngREmph), Short-Run Emphasis (ShrtREmph), and discrete wavelet decomposition transform features were changed over time. This study presents that MBL is related to the torque value during dental implant insertion and the location of the procedure. The increasing values of SumAverg and LngREmph correlated with MBL, which were 64.21 to 64.35 and 1.71 to 2.01, respectively. The decreasing values of Entr, DifEntr, and ShrtREmph also correlated with MBL, which were 2.58 to 2.47, 1.11 to 1.01, and 0.88 to 0.84, respectively. The analyzed texture features may become good indicators of MBL in digital dental surgery.

Oil Spill Detection by CP SAR Based on the Power Entropy Decomposition

In recent years, marine oil spills have adversely affected the marine economy and ecosystem, and the detection of marine oil slicks has attracted great attention. Combining different polarimetric features for better oil spill detection is a topic that needs to be studied in depth. Previous studies have shown that the compact polarimetric (CP) synthetic aperture radar (SAR) can be effectively applied to the detection of sea surface oil spill due to its own ability, which is conducive to the extraction of sea surface oil slick. In this paper, we apply the power–entropy (PE) decomposition theory, which decomposes the total scattered power according to the entropy contribution of each cell in the response, to CP SAR data for oil spill detection. The purpose of this study is to enhance the oil slick and the separability of the sea. As a result, an oil spill detection method based on the low-entropy radiation amplitude parameter lesa is proposed. We compare lesa with the other five popular polarimetric features and validate by quantitative evaluation that lesa is superior to other types of polarization feature parameters under different band data. Moreover, the random forest classification is performed on the feature map and achieves the visualization results of oil spill detection. The experimental results show that the lesa can combine the information of the two polarimetric characteristic parameters of entropy and total scattering power, and can clearly indicate the oil slick information under different scenarios.

Distinction of Human and Mechanical Vibrations within Similar Frequency Bands Based on Wavelet Entropy Using Ultrawideband Radar

Construction machinery is necessary in postdisaster emergency rescue missions involving the destruction of ruins. However, their mechanical vibrations can interfere with the detection of human survivors using ultrawideband (UWB) radar. Traditional methods detect and identify humans by determining maximum energy and checking respiratory frequency. However, they lose effectiveness because mechanical vibration is associated with a frequency band which is similar to the human respiration band, but it has higher energy. This study proposes a novel method to distinguish human vibrations from mechanical vibrations. After preprocessing, wavelet entropy decomposition was implemented on the radar data. An improved, censored mean-level detector, constant false-alarm rate algorithm was utilized to automatically identify the position of human and mechanical vibrations. A novel feature is then extracted by calculating the half-height width of the target’s wavelet entropy. Finally, the results of two independent sample t-tests prove that there is a significant statistical difference between the feature values of humans and the mechanical vibrations (p &lt; 1.9 × 10−6), thus proving the effectiveness of the method. We envisage that the proposed method can be used in postdisaster rescue missions to improve the accuracy and speed of identifying human targets. Therefore, more survivors may be rescued.

Simulation of state transfer from atom to cavity radiation in atom-photon interaction in large detuning

In this research, atom-cavity entanglement in closed Jaynes – Cumming model has been used with the assumption of large detuning between atom and cavity, to simulate the state transfer from atom to cavity radiation. A two-level atom which is prepared in a half-half superposition between the ground and excited states, has been exposed to initially coherent radiation field which is far from resonance with the atom. Mixture of each subsystem, linear entropy, entanglement, Von Neumann entropy and Schmidt decomposition of composite system as a function of time, has been determined analytically, for the first time. Assuming large frequency differences between atom and radiation, the simulation is shown no energy is transferred between atom and radiation due to large detuning, even though atom-radiation entanglement took place. Also, it has been shown that a particular measurement on the atom in this composite entangled system, can project the composite system into a separable (non-entangled) state so that the radiation state is changed into a superposition of the coherent state which is called the cat state. The conditions required to transfer radiation into odd and even cat states have been particularly investigated.