Mechanical Compound Fault Analysis Method Based on Shift Invariant Dictionary Learning and Improved FastICA Algorithm

Combined the improved fast independent component analysis (FastICA) algorithm with the singular value decomposition algorithm, a single-channel fetal electrocardiogram (fECG) extraction method is proposed. First, the improved FastICA algorithm is used to estimate the maternal ECG component from a single-channel abdominal signal of pregnant women using an overrelaxation factor. Then, a preliminary estimate of the fECG signal is obtained by subtracting from the single-channel abdominal signal. Subsequently, the singular value decomposition algorithm is used to denoise the preliminarily estimated fECG signal to obtain a high signal-to-noise ratio. In addition, in the singular value decomposition algorithm for fetal arrhythmia, an improved method for constructing the ECG signal reconstruction matrix is proposed. Finally, the fECG extraction experiments on synthetic abdominal signals and actual abdominal signals (data from 49 abdominal channels sourced from DAISY database and the non-invasive fECG database in PhysioNet) are carried out. The experimental results show that the method in this paper can effectively improve the signal-to-noise ratio and the accuracy of fECG signal extraction, and is suitable for maternal or fetal arrhythmias. Compared with the FastICA algorithm, the signal-to-noise ratio of the fECG signal extracted by the method in this paper is improved by about 5 dB, and the accuracy of fECG extraction in the PhysioNet database can reach 96.54%.

The convergence of fast independent component analysis (FastICA) algorithm based on the Newton iterative method was depended on initial value. So the different initial values could result in the different convergence speeds. To deal with this problem, this paper is proposed an improved FastICA algorithm based on symmetric orthogonalization. The algorithm selected initial value randomly, and used serial orthogonalization to get the suitable initial separating matrix firstly. Then it used symmetric orthogonalization to get the separating matrix. Finally, it could get the separated signals. Simulation results show that the proposed algorithm has faster convergence speed than the original and another improved FastICA algorithm with the same signal separation accuracy.

To solve the problem that vibration signal is easily affected by noise, which leads to the difficulty of fault feature extraction. This paper proposes a method for rolling bearing fault feature extraction which based on Cubic spline interpolation Intrinsic Time-scale Decomposition (CITD) algorithm and Fast Independent Component Analysis (FastICA) algorithm. Firstly, CITD method was used to decompose the original fault vibration signal, and a series of Proper Rotation (PR) components were obtained. Then the components with abundant fault information are selected by kurtosis criterion to reconstruct the observed signal, and the remaining components are reconstructed to obtain the virtual noise channel signal. FastICA algorithm is used to reduce the noise of reconstructed signal. The Teager-Kaiser Energy Operator (TKEO) is used to demodulate the signal after noise reduction. Finally, FFT transform is applied to the demodulated signal, the fault feature information of the original signal is extracted. Through comparative experiments, the results show that the proposed method can extract the fundamental frequency and frequency doubling characteristic information of rolling bearing faults more clearly and effectively.

A method based on a sparse auto-encoder (SAE) network for the estimation of spreading sequences in long-code direct-sequence code-division multiple access (LC-DS-CDMA) signals is proposed. First, a network classification model based on SAE and softmax classifier is established. Next, the effectiveness of the proposed method is verified by estimating Walsh sequences and m sequences. To estimate the spreading sequences, the LC-DS-CDMA signal is divided into fragments. Then, each user’s spreading sequence is separated by the fast independent component analysis (Fast-ICA) algorithm, and the amplitude fuzziness is eliminated by the delay-and-multiply method. Finally, the spreading sequences are estimated by the SAE model. Experimental results showed that the proposed algorithm could effectively estimate the spreading sequences of LC-DS-CDMA signals. Compared to the existing matching algorithm and Fast-ICA algorithm, the estimation time required by the proposed algorithm was shorter, and its estimation performance at low signal-to-noise ratios was superior.

In the case of multiple nonstationary independent source signals and linear instantaneous time-varying mixing systems, it is difficult to adaptively separate the multiple source signals. Therefore, the adaptive blind source separation (BSS) problem is firstly formally expressed and compared with tradition BSS problem. Then, we propose an adaptive blind identification and separation method based on the variable learning rate equivariant adaptive source separation via independence (EASI) algorithm. Furthermore, we analyze the scope and conditions of variable-learning rate EASI algorithm. The adaptive BSS simulation results also show that the variable learning rate EASI algorithm provides better separation effect than the fixed learning rate EASI and recursive least-squares algorithms.

<p>The exploitation of ever increasing Interferometric Synthetic Aperture Radar (InSAR) datasets to monitor the Earth surface deformation is an important goal of today’s geodesy. Surface geodetic deformation observations are often the result of the combination of a multitude of sources (either volcano-tectonic deformation associated with seismic events, post-seismic relaxation, aseismic transients, long-term creep loading, magma intrusions or non-tectonic deformation associated with hydrological loads, poroelastic rebound, anthropic activity and various sources of noise). In this regard, we are facing a so-called Blind Source Separation (BSS) problem. Natural approaches to tackle BSS problems are those multivariate statistical techniques which attempt to decompose the dataset into a limited number of statistically independent sources, under the assumption that the different physical mechanisms underlying the observations have independent footprints either in space or time. Multiple algorithms have been proposed to separate the various independent sources, and here we show the capabilities of a variational Bayesian Independent Component Analysis (vbICA) algorithm. In particular, we show through synthetic test cases its superiority with respect to other commonly used multivariate statistical techniques like the Principal Component Analysis (PCA) and the FastICA algorithm. Application of vbICA to InSAR time series from European Space Agency (ESA) Sentinel-1 satellite in the Central Valley and on the Central San Andreas Fault segment, California, spanning the time range 2015-2019, shows that the algorithm provides a viable way to separate elastic and inelastic deformation in response to the aquifer charge/discharge as well as creeping signal from seasonal loading.</p>

The impact of outliers on signal separation performance of an independent component analysis (ICA) algorithm is an important factor when selecting an ICA algorithm. If an ICA estimator has the property of B-robustness, the influence of an extreme point is bounded, leading to good separation performance in the presence of outliers. Since this property is binary, it does not give the degree of influence an outlier has on the separation performance. To address this issue, a Mamdani-type fuzzy inference, based on the location of a potential outlier and on the skewness of the data set, has been developed. It creates an outlier sensitivity map for an ICA algorithm. The implication of this work is a criterion to switch from one ICA algorithm to another in real time, as determined by the algorithms sensitivity to the data set under consideration. This paper describes estimation of the outlier impact on the separation performance of the non-B-robust FastICA algorithm using a Mamdani-type fuzzy inference. In simulations with data sets contaminated by outliers, the FastICA sensitivity map resembles the separation performance measured by the Amari performance index.

As one of the main optical wireless communication technologies, visible light communication (VLC) is considered a promising supplement in 5G/6G and the internet of things (IoT) networks. The multiple-input–multiple-output (MIMO) multiplexing scheme can significantly enhance a VLC system’s performance. To handle the blind MIMO demultiplexing, geometric vector extraction (GVE) algorithms are used in a mixture space (MS), which has not been presented in the extant literature. In a 2 × 2 VLC MIMO scenario, the singular value decomposition of an arbitrary MIMO channel matrix is introduced to discuss the one-to-one relationship between channel-matrix elements and the outline-shape conversion starting from the transmitted rectangular-shaped points in the MS. By extracting the vector features of the adjacent sides of the received parallelogram-shaped constellation points, we can obtain the elements of an unknown channel matrix and recover the superposed signals transmitted by two light-emitting diodes (LED). To obtain a vector from these constellation points, clustering is required to obtain the scattered points’ centroids in advance. To avoid falling into a local minimum, we present a clustering algorithm with an initial-position-expansion pre-processing stage to place the initial centroids with the greatest accumulated distance. To enhance the computational efficiency, we enhance the clustering algorithm via a single iteration to obtain the centroids, which we named the GVE-for-iteration-once (GVE–IO) algorithm. The enhanced algorithm has an equivalent or better MIMO demultiplexing performance and quick convergence performance compared to the fast independent component analysis (FastICA) algorithm. Moreover, the GVE–IO algorithm’s computational complexity and measured running time is reduced to 29% and 23%, respectively, of the FastICA algorithm’ time costs. Finally, the experimental investigation shows that the proposed scheme exhibits a significantly reduced computational complexity in all discussed voltage peak-to-peak (Vpp) cases of LED and a Q-value enhancement in two-thirds of the Vpp combination regions for 50 input points.

The multicarrier code division multiple access (MC-CDMA) technique has received considerable interest for its potential application to future wireless communication systems due to its high data rate. A common problem regarding the blind multiuser detectors used in MC-CDMA systems is that they are extremely sensitive to the complex channel environment. Besides, the perturbation of colored noise may negatively affect the performance of the system. In this paper, a new coherent detection method will be proposed, which utilizes the modified fast independent component analysis (FastICA) algorithm, based on approximate negentropy maximization that is subject to the second-order cone programming (SOCP) constraint. The aim of the proposed coherent detection is to provide robustness against small-to-medium channel estimation mismatch (CEM) that may arise from channel frequency response estimation error in the MC-CDMA system, which is modulated by downlink binary phase-shift keying (BPSK) under colored noise. Noncoherent demodulation schemes are preferable to coherent demodulation schemes, as the latter are difficult to implement over time-varying fading channels. Differential phase-shift keying (DPSK) is therefore the natural choice for an alternative modulation scheme. Furthermore, the new blind differential SOCP-based ICA (SOCP-ICA) detection without channel estimation and compensation will be proposed to combat Doppler spread caused by time-varying fading channels in the DPSK-modulated MC-CDMA system under colored noise. In this paper, numerical simulations are used to illustrate the robustness of the proposed blind coherent SOCP-ICA detector against small-to-medium CEM and to emphasize the advantage of the blind differential SOCP-ICA detector in overcoming Doppler spread.

The economic and environmental losses due to serious leakage in the urban water supply network have increased the effort to control the water leakage. However, current methods for leakage estimation are inaccurate leading to the development of ineffective leakage controls. Therefore, this study proposes a method based on the blind source separation theory (BSS) to calculate the leakage of water supply network. The method uses fast independent component analysis (FastICA) algorithm to separate flow signal of laboratory and practical measuring area, adopts trend similarity to solve the uncertainty of separation sequence to get hourly change curve of user usage and physical leakage, and embeds the leakage model into amplitude optimization model to solve amplitude uncertainty to obtain physical leakage value. The study found that the estimation of leakage level using the blind source separation is reasonably accurate and facilitates the identification of the subsequent reduction in water leakage. This can provide scientific evidence for leakage reduction and the investment of pressure relief devices in the next stage.

A new blind image source separation algorithm based on feedback sparse component analysis

Fetal electrocardiograms (FECGs) provide important clinical information for early diagnosis and intervention. However, FECG signals are extremely weak and are greatly influenced by noises. FECG signal extraction and detection are still challenging. In this work, we combined the fast independent component analysis (FastICA) algorithm with singular value decomposition (SVD) to extract FECG signals. The improved wavelet mode maximum method was applied to detect QRS waves and ST segments of FECG signals. We used the abdominal and direct fetal ECG database (ADFECGDB) and the Cardiology Challenge Database (PhysioNet2013) to verify the proposed algorithm. The signal-to-noise ratio of the best channel signal reached 45.028 dB and the issue of missing waveforms was addressed. The sensitivity, positive predictive value and F1 score of fetal QRS wave detection were 96.90%, 98.23%, and 95.24%, respectively. The proposed algorithm may be used as a new method for FECG signal extraction and detection.

Denoising of MRI images is very essential for the effective diagnosis of various brain diseases. In this paper, a new hybrid ROFTV-Fast ICA algorithm is proposed to enhance the MRI images corrupted by Gaussian noise. The original MRI image is subjected to Gaussian noise. The corrupted brain image is denoised by the combination of both Rudin-Osher- Fatemi (ROF) Total variation filter and Fast-Independent Component Analysis (ICA) algorithms. The total variation in the noisy brain images is minimized by using ROFTV filter. Again, the recovered image is denoised further by Fast ICA algorithm, by separating the noise and noiseless components in the image. The performance of this hybrid ROFTV -Fast ICA filter is evaluated by means of Peak Signal to Noise Ratio (PSNR). The proposed method is also compared with Adaptive Median Filter (AMF), Progressive Switching Median Filter (PSMF) and Bilateral filter (BF). The result shows that the proposed hybrid algorithm outperforms rest of the filters and smoothens the MRI images very well also preserving the edges and corners.

This paper introduces the basic theory of ICA and the FastICA algorithm. In order to increase the algorithm convergence rate and reduce the running time, the paper which is based on FastICA algorithm proposes an improved independent component analysis algorithm with Newton's iteration- the “Newton Type” iteration algorithm with third-order convergence. The simulation results of the image signal separation show that the improved algorithm have the same separate effect as conventional FastICA algorithm and can reduce the times of iterations.

An algorithm to improve direction of arrival (DOA) estimation accuracy with an extended sensor array in the presence of multiple coherent signal sources is proposed. The algorithm uses virtual element theory to extend the sensor array, estimates virtual element information via linear prediction and expands the array aperture in practical sense; the sparsity of the target orientation in angle space is exploited to establish an over-complete dictionary and a reception model for the array signal in sparse space; the received array data is preprocessed using singular value decomposition (SVD) method and target DOA estimation is realized by calculating the best atoms. The algorithm improves DOA estimation accuracy by extended array and uses SVD to control computational complexity effectively, which ensures the accuracy and efficiency. Computer simulation shows that the proposed algorithm is able to accurately estimate the DOA for both single-target and closely spaced multi-target cases in low signal-to-noise ratio environments, and also has excellent DOA estimation performance in the presence of multiple coherent signal sources.