Fast Independent Component Analysis Research Articles

Topological Comparison of Some Dimension Reduction Methods Using Persistent Homology on EEG Data

In this paper, we explore how to use topological tools to compare dimension reduction methods. We first make a brief overview of some of the methods often used in dimension reduction such as isometric feature mapping, Laplacian Eigenmaps, fast independent component analysis, kernel ridge regression, and t-distributed stochastic neighbor embedding. We then give a brief overview of some of the topological notions used in topological data analysis, such as barcodes, persistent homology, and Wasserstein distance. Theoretically, when these methods are applied on a data set, they can be interpreted differently. From EEG data embedded into a manifold of high dimension, we discuss these methods and we compare them across persistent homologies of dimensions 0, 1, and 2, that is, across connected components, tunnels and holes, shells around voids, or cavities. We find that from three dimension clouds of points, it is not clear how distinct from each other the methods are, but Wasserstein and Bottleneck distances, topological tests of hypothesis, and various methods show that the methods qualitatively and significantly differ across homologies. We can infer from this analysis that topological persistent homologies do change dramatically at seizure, a finding already obtained in previous analyses. This suggests that looking at changes in homology landscapes could be a predictor of seizure.

Establishment of original liquor discrimination using fluorescence spectroscopy combined with chemometric analysis and its application in Chinese mild-flavor liquors

Original liquor quality control and grade identification are important for the quality improvement of finished liquor. For the purpose of this study, a meticulous collection of 209 samples of original Chinese mild-flavored liquors was conducted. The grading of these samples was determined through sensory tasting, which is the prevailing method used in major liquor factories. These samples were scanned by three-dimensional fluorescence spectroscopy to obtain the respective excitation-emission matrices (EEM). The suitability of EEM was evaluated for the classification of these samples using principal component analysis (PCA), Discrete Cosine Transform (DCT), Fast Independent Component Analysis (Fast-ICA) combined with linear discriminant analysis (LDA) and support vector machine (SVM). In general, it was found that a fast and labor-saving model for quality identification of original liquor was obtained, PCA was a more suitable algorithm for reducing the number of variables. The classification accuracy of the original liquor obtained by the SVM method after PCA dimensionality reduction was the highest, the training set accuracy was 89.81%. Therefore, the PCA-SVM model was selected, and after testing, its prediction accuracy was 84.62%. The model has undergone five-fold cross-validation (CV), and the CV accuracy is 81.53%, which proves that the model is sufficiently stable.

A high-precision and efficient algorithm for space-based ADS-B signal separation

AbstractSpace-based automatic dependent surveillance-broadcast (ADS-B) receivers can cover thousands of aircraft, each transmitting 6 ⋅ 2 signals per second. As a result, ADS-B signals are very prone to overlap. When the number of aircraft covered by a receiver reaches 3,000, about 90 % of the signals will be overlapping. Overlapped signals can reduce the decoding accuracy of receivers, so that aircraft information cannot be accurately transmitted to the air traffic control (ATC) surveillance system, hence threatening aviation flight safety. It is necessary to propose signal separation algorithms for space-based ADS-B systems. An orthogonal projection linear constrained minimum variance (OPLCMV) algorithm is proposed, which can separate two signals simultaneously based on the linearly constrained minimum variance algorithm by exploiting the characteristics of overlapped signals. Compared with the state-of-the-art extended projection algorithm and the fast independent component analysis algorithm, the OPLCMV method has a higher decoding accuracy for multiple overlapping signals with a small direction difference of arrival or frequency shift. Moreover, the OPLCMV algorithm has a low computational complexity when the number of overlapped signal sources is less than seven.

A New Method for Separating Temperature Effect of Bridge Strain Monitoring

Temperature has a significant influence on bridge strain monitoring data. To improve the accuracy of temperature effect separation in strain monitoring data, this paper proposes a temperature effect separation method comprising variational nonlinear chirp mode decomposition (VNCMD), principal component analysis (PCA) and blind source separation. Firstly, VNCMD was used to decompose the monitoring data of strain and temperature, and the intrinsic mode functions (IMF) of strain and temperature signals were obtained. Secondly, PCA was used to reduce the dimension of IMF, and the false components were eliminated to select the optimal components. After reducing the dimension, the components were used as the input of fast independent component analysis model for blind source separation. Finally, the feasibility and accuracy of the method was verified via the signal-to-noise ratio (SNR) in the simulated signal, and the separation results were evaluated using the Pearson correlation coefficient between the strain component and the corresponding temperature component in real bridge monitoring data. The proposed method performed better than the empirical mode decomposition (EMD) method. The signal-to-noise ratio (SNR) of VNCMD improved 51.80% for daily temperature difference effect and 32.41% for annual temperature difference effect in the numerical study, respectively; the correlation coefficients of VNCMD improved 52.90% for daily temperature difference effect and 4.26% for annual temperature difference effect in practical verification, respectively.

Rolling Bearing Composite Fault Diagnosis Method Based on Enhanced Harmonic Vector Analysis.

Composite fault diagnosis of rolling bearings is very challenging work, especially when the characteristic frequency ranges of different fault types overlap. To solve this problem, an enhanced harmonic vector analysis (EHVA) method was proposed. Firstly, the wavelet threshold (WT) denoising method is used to denoise the collected vibration signals to reduce the influence of noise. Next, harmonic vector analysis (HVA) is used to remove the convolution effect of the signal transmission path, and blind separation of fault signals is carried out. The cepstrum threshold is used in HVA to enhance the harmonic structure of the signal, and a Wiener-like mask will be constructed to make the separated signals more independent in each iteration. Then, the backward projection technique is used to align the frequency scale of the separated signals, and each fault signal can be obtained from composite fault diagnosis signals. Finally, to make the fault characteristics more prominent, a kurtogram was used to find the resonant frequency band of the separated signals by calculating its spectral kurtosis. Semi-physical simulation experiments are conducted using the rolling bearing fault experiment data to verify the effectiveness of the proposed method. The results show that the proposed method, EHVA, can effectively extract the composite faults of rolling bearings. Compared to fast independent component analysis (FICA) and traditional HVA, EHVA improves separation accuracy, enhances fault characteristics, and has higher accuracy and efficiency compared to fast multichannel blind deconvolution (FMBD).

An efficient and secure key management with the extended convolutional neural network for intrusion detection in cloud storage

SummaryCloud computing aids users for storing and recovering their information everywhere in the world. Security and efficiency are the two main issues in cloud service. Numerous intrusion detection techniques for the cloud computing environment were proposed, but those techniques do not effectively and accurately detect the attacks. Hence, an efficient and secure key management using extended convolutional neural network, that is, hybrid Enhanced Elman Spike convolutional Neural Network optimized with improved COOT optimization algorithm (Hyb EESCCNN) is proposed for intrusion detection in cloud system. Furthermore, the novel Adaptive Tangent Brakerski‐Gentry Vaikuntanathan Homomorphic Encryption (ATBGVHE) method is proposed for providing the security of the system. At first, SHA‐512 is used for authenticating cloud users to store its own information in to the cloud server. Then for Intrusion Detection (ID) the input data from the NSL‐KDD, UNSWNB15, CICIDS2018, and ToN‐IoT datasets are pre‐processed. The most relevant features are extracted using Fast Independent Component Analysis (Fast ICA) from the pre‐processed output. These extracted data are classified into malicious and non‐malicious data using Hyb EESCCNN. After classification, the non‐malicious data is secured using an ATBGVHE technique. The outcomes of the proposed methods shows that the NSL‐KDD datasets attains 99.9% higher accuracy, UNSW‐NB15 datasets offer 99.89% higher accuracy, CSE‐CIC‐IDS2018 datasets attains 99.8% higher accuracy, ToN‐IoT datasets attains 99.8% higher accuracy and 0.02 s lower encryption time compared with existing methods. Finally, case study with real time applications is also analyzed to prove the efficiency of the proposed method.

A novel non-contact heart rate measurement method based on EEMD combined with FastICA

Objective. The imaging photoplethysmography (IPPG), which is applied to heart rate measurement, is susceptible to interference caused by illumination variations during photography. To improve the accuracy, robustness, and real-time performance of IPPG, a new method combined with modified ensemble empirical mode decomposition (EEMD) and fast independent component analysis (FastICA) was proposed. Approach. On the modified EEMD, to reduce the error caused by the residual white noise of the algorithm, it was characterized by recurrently updating zero-mean white noise for ensemble averaging based on the standard deviation of the input signal. On the modified FastICA, its constructed nonlinear function was superseded by Huber’s approximation function to improve the robustness and running speed. Main results. Comparison experiments were conducted between the MAHNOB-HCI database and own collected data. In the hybrid natural light and computer screen light scenario, the mean absolute error (MAE) of heart rate amounted to 0.93 beats per minute and the correlation coefficient reached 0.85. In the experiments of MAHNOB-HCI database, the MAE amounted to 6.03 beats per minute and the correlation coefficient reached 0.75. Furthermore, the modified method decreased the running time by approximately ten times compared to the original algorithms. Significance. Various experimental results demonstrated that the proposed method significantly improves the accuracy, timeliness, and interference resistance.

Application of Underdetermined Blind Source Separation Algorithm on the Low-Frequency Oscillation in Power Systems

The timely discovery of low-frequency oscillations in power systems and accurate identification of their modal parameters is critical in numerous applications. Therefore, we investigated the feasibility of using multi-channel signals and established a relative theory. An algorithm based on the underdetermined blind source separation (UBSS) algorithm was proposed using this theory. First, the energy ratio function was used to determine the fault occurrence time. Then, the Bayesian information criterion was used to estimate the number of fault sources, and the boundary conditions were set to determine the number of fault sources. Next, the UBSS algorithm was used to analyze raw data, extract individual components that characterize faults, and subsequently measure low-frequency oscillation modal parameters through the Hilbert transform. Finally, the fast independent component analysis (FastICA) algorithm was used to separate noise signal from raw data. This separation considerably reduced noise disturbance and ensured the stability of the proposed method. Model simulation was conducted in MATLAB and experimental measurement revealed that the proposed method effectively reduced noise disturbance and could be applied to conditions with considerable disturbance.

Operating modal analysis from sub-sampled vibration response based on compressed sensing and FastICA

In order to identify more high order modal from a small amount of vibration response signal sampling, this paper proposes a method for identifying the OMA of sub-sampled vibration response signals based on compression sensing (CS) and fast independent component analysis (FastICA). Firstly, the method subsamples a large number of vibrational signals to retain the original key information of the signal; Secondly, the subsampled signal is restored by using the compressive sampling matching pursuit (CoSaMP) algorithm of the compression-aware reconstruction algorithm. Finally, the source signal is separated by FastICA to obtain the multi-order natural frequencies and modal mode shapes of the structure. According to Shannon’s theorem, if the analog signal is reproduced without distortion, the sampling frequency should be no less than 2 times the highest frequency in the analog signal spectrum. This method can reduce the amount of original signal acquisition and the requirements for the frequency of the sampled signal, and the recovery signal has a high similarity with the original signal and the natural frequency is not changed. It can effectively identify the higher-order modes of the structure on the basis of breaking through the Nyquist sampling frequency. Experimental results of the subsampled vibration response signal on the uniform steel cantilever beam show that the method can reconstruct the original signal from a small number of subsampled vibration response signals, and identify the high-order nature frequency and modal mode shape that break through the Nyquist sampling frequency.

Parameter Identification of Wheel Polygonization Based on Effective Signal Extraction and Inertial Principle

Violent interactions between wheel and rail caused by the polygonal wheels can strongly influence the operational comfort and safety of railway vehicles. It is an effective tool to dynamically detect wheel defects with the axle box acceleration signals. However, since the onboard vibration signals of the axle box are mixed with abundant noise from track irregularity and vehicle components, it is difficult to ensure the detection accuracy of the polygonal wheel. A novel strategy is proposed to extract the effective signal components and quantitatively identify the polygonal wheel parameters. First, the genetic mutation particle swarm optimization (GMPSO) algorithm is developed to search for optimization parameters of variational mode decomposition (VMD). Specifically, the minimum approximate entropy is adopted for the objective function, and the over-decomposition issue of VMD is addressed by the criterion of center frequencies ratio. Then, the independent components (ICs) are separated from VMD modes and the original signal by fast independent component analysis (FastICA). The dominated signal components related to the polygonal wheels can be decided by the proposed criterion. Finally, the order and amplitude of the wheel polygonal wear are obtained from the effective IC based on the inertial principle. The effectiveness of the proposed parameter identification algorithm is verified by the dynamics simulation and field test, and the estimation accuracy of wear amplitudes outperforms the comparison methods.

Fetal electrocardiogram signal extraction and analysis method combining fast independent component analysis algorithm and convolutional neural network

Fetal electrocardiogram (ECG) signals provide important clinical information for early diagnosis and intervention of fetal abnormalities. In this paper, we propose a new method for fetal ECG signal extraction and analysis. Firstly, an improved fast independent component analysis method and singular value decomposition algorithm are combined to extract high-quality fetal ECG signals and solve the waveform missing problem. Secondly, a novel convolutional neural network model is applied to identify the QRS complex waves of fetal ECG signals and effectively solve the waveform overlap problem. Finally, high quality extraction of fetal ECG signals and intelligent recognition of fetal QRS complex waves are achieved. The method proposed in this paper was validated with the data from the PhysioNet computing in cardiology challenge 2013 database of the Complex Physiological Signals Research Resource Network. The results show that the average sensitivity and positive prediction values of the extraction algorithm are 98.21% and 99.52%, respectively, and the average sensitivity and positive prediction values of the QRS complex waves recognition algorithm are 94.14% and 95.80%, respectively, which are better than those of other research results. In conclusion, the algorithm and model proposed in this paper have some practical significance and may provide a theoretical basis for clinical medical decision making in the future.

Blind Estimation of Spreading Code Sequence of QPSK-DSSS Signal Based on Fast-ICA

Most of the existing estimation methods of spreading code sequence are not suitable for the QPSK-DSSS. We propose a spreading code sequence estimation method based on fast independent component analysis (Fast-ICA). It mainly includes signal preprocessing, calculations of separation matrix, and spreading code sequence. Firstly, the received signal is segmented according to the period of the spreading code sequence, and the covariance matrix can be calculated. Then, the signal subspace and corresponding eigenvalues are obtained by eigenvalue decomposition of the covariance matrix. Subsequently, the received signal matrix needs to be whitened. Finally, the Fast-ICA algorithm is used to find the separation matrix to estimate the in-phase and orthogonal spreading code sequence. The experiment result shows that the estimation of the spreading code sequence can be carried out based on Fast-ICA under a low SNR of −12 dB. Compared with the constant modulus algorithm (CMA) and the decomposition method for the real part of the self-covariance matrix (EVD-R), this method has a better performance.

Packaged butter adulteration evaluation based on spatially offset Raman spectroscopy coupled with FastICA

Optical detection technology has been widely used in unpackaged food adulteration detection. However, due to the interference of packaging materials on the internal food optical signal, including signal occlusion, mixing and overlap precluded the accurate detection of internal food quality. In this study, a method of packaged butter adulteration evaluation based on spatially offset Raman spectroscopy (SORS) combined with fast independent component analysis (FastICA) was proposed. The adulterated butter from 0% to 100% w/w margarine at 10% intervals was covered with packaging sheets as test samples. A line-scan Raman hyperspectral imaging system was used to obtain a scattering spectral image of the packaged butter samples. The region of interest of the scattering image is extracted as the input of FastICA model to separate the internal butter signals. The extracted butter Raman features were input into four quantitative analysis models to assess the content of butter adulteration. The results showed that the ensemble model Extra-tree has the best performance with RMSEp, Rp2, and RPD values of 0.6, 0.93, and 4.73, respectively. Additionally, the applicability of the method was validated with four types of packaging materials. This rapid non-destructive testing method is beneficial to the effective testing method of packaged butter and other products industry.

Research on the evaluation of rectal function after LAR based on CEEMDAN‐Fast‐ICA algorithm

AbstractRectal cancer is one of the most common lower gastrointestinal diseases worldwide. Currently, the common treatment is low anterior resection (LAR) of the rectum, which preserves the anus of the patient. However, it is easy to cause low anterior resection syndrome after surgery, which has a significant negative impact on the life of patients, and there is no unified evaluation standard for postoperative rectal function. To solve this problem, a multi‐sensor fusion rectal information acquisition system is designed in this paper, and a rectal signal processing method is proposed to theoretically evaluate the rectal function of postoperative patients. The method uses the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) to decompose the one‐dimensional rectal signal to solve the underdetermined ICA problem, uses the Fast independent component analysis (Fast‐ICA) to separate the pure rectal signal, uses the wavelet packet to extract features, and uses the particle swarm optimization optimizes support vector machine (PSO‐SVM) to classify and evaluate postoperative function. According to the experimental results, the rectal signal preprocessing effect is good, the evaluation prediction rate is 99.5565%, and the algorithm classification results are accurate, which provides a certain preliminary theoretical basis and reference value for the evaluation of rectal function after LAR.

Deep learning aided cyclostationary feature analysis for blind modulation recognition in massive MIMO systems

Blind modulation recognition (BMR) has been proposed as a promising approach for massive multiple-input multiple-output (M-MIMO) systems to support massive user equipment (UE). In this work, we devise a new uplink BMR algorithm based on a deep learning (DL) aided cyclostationary feature (CF) analysis approach. Firstly, we propose the so-called minimum description length (MDL) method aided complex fast independent component analysis (CFICA) algorithm, to separate the multi-user (MU) signals subjected to arbitrary, different modulation schemes with an unknown number of transmitters. Then, we tailor a convolutional deep belief network (CDBN) to analyze the relation between CF and modulation type (MT) for generating the BMR results. As the proposed BMR process is invoked blindly, it is particularly suitable for non-cooperative MU communication scenarios. Simulation results demonstrate the effectiveness of the proposed technique.

Concise method for high-precision vibration recovery of Φ-OTDR based on the GF-FastICA algorithm.

To efficiently restore the vibration signals of a phase-sensitive optical time domain reflectometer (Φ-OTDR), the GF-FastICA joint algorithm is proposed, which combines guided filtering with fast independent component analysis (FastICA). The marked region of vibration is precisely located by guided filtering. FastICA deals with the optimal phase mixing matrix of the marked region to separate the vibration signals from the noise-containing phase signals. The experimental results show that the GF-FastICA achieves a correlation coefficient of 0.998 for 5-Hz vibration signal recovery from a 14.3-km sensing fiber, verifying the potency of the algorithm. Compared with the traditional method and FastICA only, GF-FastICA improves the root mean square error (RMSE) metric by an order of magnitude, which is approaching an experience value of 10-3.

MUNIX repeatability evaluation method based on FastICA demixing.

To enhance the reproducibility of motor unit number index (MUNIX) for evaluating neurological disease progression, this paper proposes a negative entropy-based fast independent component analysis (FastICA) demixing method to assess MUNIX reproducibility in the presence of inter-channel mixing of electromyography (EMG) signals acquired by high-density electrodes. First, composite surface EMG (sEMG) signals were obtained using high-density surface electrodes. Second, the FastICA algorithm based on negative entropy was employed to determine the orthogonal projection matrix that minimizes the negative entropy of the projected signal and effectively separates mixed sEMG signals. Finally, the proposed experimental approach was validated by introducing an interrelationship criterion to quantify independence between adjacent channel EMG signals, measuring MUNIX repeatability using coefficient of variation (CV), and determining motor unit number and size through MUNIX. Results analysis shows that the inclusion of the full (128) channel sEMG information leads to a reduction in CV value by $1.5 \pm 0.1$ and a linear decline in CV value with an increase in the number of channels. The correlation between adjacent channels in participants decreases by $0.12 \pm 0.05$ as the number of channels gradually increases. The results demonstrate a significant reduction in the number of interrelationships between sEMG signals following negative entropy-based FastICA processing, compared to the mixed sEMG signals. Moreover, this decrease in interrelationships becomes more pronounced with an increasing number of channels. Additionally, the CV of MUNIX gradually decreases with an increase in the number of channels, thereby optimizing the issue of abnormal MUNIX repeatability patterns and further enhancing the reproducibility of MUNIX based on high-density surface EMG signals.

Anomaly Detection of UAV State Data Based on Single-Class Triangular Global Alignment Kernel Extreme Learning Machine

Unmanned Aerial Vehicles (UAVs) are widely used and meet many demands in military and civilian fields. With the continuous enrichment and extensive expansion of application scenarios, the safety of UAVs is constantly being challenged. To address this challenge, we propose algorithms to detect anomalous data collected from drones to improve drone safety. We deployed a one-class kernel extreme learning machine (OCKELM) to detect anomalies in drone data. By default, OCKELM uses the radial basis (RBF) kernel function as the kernel function of the model. To improve the performance of OCKELM, we choose a Triangular Global Alignment Kernel (TGAK) instead of an RBF Kernel and introduce the Fast Independent Component Analysis (FastICA) algorithm to reconstruct UAV data. Based on the above improvements, we create a novel anomaly detection strategy FastICA-TGAK-OCELM. The method is finally validated on the UCI dataset and detected on the Aeronautical Laboratory Failures and Anomalies (ALFA) dataset. The experimental results show that compared with other methods, the accuracy of this method is improved by more than 30%, and point anomalies are effectively detected.

ScSSA: A clustering method for single cell RNA-seq data based on semi-supervised autoencoder

Single cell sequencing is a technology for high-throughput sequencing analysis of genome, transcriptome and epigenome at the single cell level. It can improve the shortcomings of traditional methods, reveal the gene structure and gene expression state of a single cell, and reflect the heterogeneity between cells. Among them, the clustering analysis of single-cell RNA data is a very important step, but the clustering of single-cell RNA data is faced with two difficulties, dropout events and dimension curse. At present, many methods are only driven by data, and do not make full use of the existing biological information. In this work, we propose scSSA, a clustering model based on semi-supervised autoencoder, fast independent component analysis (FastICA) and Gaussian mixture clustering. Firstly, the semi-supervised autoencoder imputes and denoises the scRNA-seq data, and then get the low-dimensional latent representation. Secondly, the low-dimensional representation is reduced the dimension and clustered by FastICA and Gaussian mixture model respectively. Finally, scSSA is compared with Seurat, CIDR and other methods on 10 public scRNA-seq datasets. The results show that scSSA has superior performance in cell clustering on 10 public datasets. In conclusion, scSSA can accurately identify the cell types and is generally applicable to all kinds of single cell datasets. scSSA has great application potential in the field of scRNA-seq data analysis. Details in the code have been uploaded to the website https://github.com/houtongshuai123/scSSA/.

Compressed spike-triggered averaging in iterative decomposition of surface EMG

Background and ObjectiveAnalysis of motor unit activity is important for assessing and treating diseases or injuries affecting natural movement. State-of-the-art decomposition translates high-density surface electromyography (HDsEMG) into motor unit activity. However, current decomposition methods offer far from complete separation of all motor units. MethodsThis paper proposes a peel-off approach to automatic decomposition of HDsEMG into motor unit action potential (MUAP) trains, based on the Fast Independent Component Analysis algorithm (FastICA). The novel steps include utilizing compression by means of Principal Component Analysis and spike-triggered averaging, to estimate surface MUAP distributions with less noise, which are iteratively subtracted from the HDsEMG dataset. Furthermore, motor unit spike trains are estimated by high-dimensional density-based clustering of peaks in the FastICA source output. And finally, a new reliability measure is used to discard poor motor unit estimates by comparing the variance of the FastICA source output before and after the peel-off step. The method was validated using reconstructed synthetic data at three different signal-to-noise levels and was compared to an established deflationary FastICA approach. ResultsBoth algorithms had very high recall and precision, over 90%, for spikes from matching motor units, referred to as matched performance. However, the peel-off algorithm correctly identified more motor units for all noise levels. When accounting for unidentified motor units, total recall was up to 33 percentage points higher; and when accounting for duplicate estimates, total precision was up to 24 percentage points higher, compared to the state-of-the-art reference. In addition, a comparison was done using experimental data where the proposed algorithm had a matched recall of 97% and precision of 85% with respect to the reference algorithm. ConclusionThese results show a substantial performance increase for decomposition of simulated HDsEMG data and serve to validate the proposed approach. This performance increase is an important step towards complete decomposition and extraction of information of motor unit activity.