Second-order Blind Identification Research Articles

An automated operational modal analysis algorithm and its application to concrete dams

With the increasing application of long-term dynamic monitoring systems to civil engineering structures, automated operational modal analysis (OMA) has become a fascinating research area for structural health monitoring. In recent years, different automated approaches based on the existing parametric or nonparametric OMA algorithms have been developed. However, they usually perform less well in large complex structures with weak excitation because of inherent limitations of those algorithms. In this work, a three-stage automated OMA algorithm based on a combination of the second-order blind identification (SOBI) and the covariance-driven stochastic subspace identification (SSI-COV) is proposed, which takes full advantage of both parametric and nonparametric algorithms while overcoming the limitations. The automated approach avoids the ambiguous work related to interpreting stabilization diagrams without losing accuracy and thus obtains more reliable modal parameters. The proposed method is fully validated by simulated data from an 8-DOF system and data recorded on a concrete arch-gravity dam in view of single and continuous monitoring. The results show that all physical modes can be well-separated and reliably identified by the proposed method with high accuracy, confirming the excellent performance of the method, especially for continuous monitoring.

High-speed ocular artifacts removal of multichannel EEG based on improved moment matching

Objective. The excellent signal-to-noise ratio (SNR) is the premise of electroencephalogram (EEG) research and applications. This study aims to use innovative method to swiftly remove the ocular artifacts (OAs) from multichannel EEG to enhance the SNR. Approach. The moment matching method which is prevalently used to removing stripe noise from hyperspectral images is adapted and improved to deduct OAs from EEG. This modified approach regards sampling points of multichannel EEG as pixels in images. It utilizes the propagation characteristics of EEG to correct the potential shift caused by OAs. Main results. By using mathematical derivation and empirical corroboration, the results suggest that the improved moment matching (IMM) is capable of reducing OAs effectively and reserving the EEG waveform information on the greatest extent compared to existing methods, such as independent component analysis (ICA) and second-order blind identification. In the frontal region heavily affected by OAs, the SNR increased by 138.1% compared with ICA, the whole SNR increased by an average of 58.7%. Moreover, low latency superiority is provided for real-time and offline processing. IMM is effective for OAs removal and it is helpful to improve SNR of multichannel EEG. Significance. IMM affords option offering preponderance for enhancement of SNR of multichannel EEG.

FMRI activations via low-complexity second-order inverse-sparse-transform blind separation

Since functional Magnetic Resonance Imaging (fMRI) signals are a group of sparse signals, and its autocorrelation matrix contains limited information, it is difficult to accurately locate the brain activation area directly using traditional blind separation algorithms. For the issue, this paper proposes a method with inverse-sparse transform and second order blind identification (SOBI) for the separation of the activations. The contribution of this paper is to achieve the separation of sparse brain map signals and have lower computational complexity than higher-order statistical BSS. In experiments, we use both simulated and measured fMRI data to evaluate our method. The experimental results show that the proposed method's running time is only 1/30 of a higher-order statistical independent component analysis (ICA) algorithm, while its separation errors is close to ICA and less than half of a traditional SOBI algorithm.

Low observable group targets detection based on space–frequency cascaded adaptive processing for MIMO OTH radar

The spread of sea clutter in the Doppler domain caused by multi-mode propagation would make it difficult for over-the-horizon radar to detect low observable group targets. To solve this problem, a multi-mode clutter suppression method based on space–frequency cascaded adaptive processing is proposed by utilizing the spatial diversity characteristics of multiple-input–multiple-output radar. Then, the low observable group targets can be detected. The proposed method adopts the idea of space–frequency stepwise processing. First, the multi-stage Wiener filter ESPRIT algorithm is exploited to estimate the direction of departure and direction of arrival of different path signals, and then the multi-mode clutter is adaptively transformed into single-mode clutter based on the estimated angles. Second, the atomic norm minimization algorithm and Vandermonde decomposition are used to estimate the steering vector of the first order Bragg frequency of sea clutter, whose path contains the group targets. Finally, the least squares algorithm is used to realize the adaptive suppression of sea clutter in the frequency domain. Because there is no need to estimate the signal covariance matrix and its eigenvalue decomposition and conduct spectral peak searching, the proposed method greatly reduces the computation of the angle estimation. Compared with the existing second-order blind identification and spatial smoothing SOBI algorithms, the proposed algorithm does not need the information of the first order Bragg frequency of sea clutter, so it can improve the output signal-to-clutter-noise ratio when there is an error in the first order Bragg frequency of sea clutter. Therefore, the proposed algorithm is more conducive to the detection of low observable group targets. Simulation results are given to demonstrate the effectiveness of the proposed method.

A review of second-order blind identification methods.

Second‐order source separation (SOS) is a data analysis tool which can be used for revealing hidden structures in multivariate time series data or as a tool for dimension reduction. Such methods are nowadays increasingly important as more and more high‐dimensional multivariate time series data are measured in numerous fields of applied science. Dimension reduction is crucial, as modeling such high‐dimensional data with multivariate time series models is often impractical as the number of parameters describing dependencies between the component time series is usually too high. SOS methods have their roots in the signal processing literature, where they were first used to separate source signals from an observed signal mixture. The SOS model assumes that the observed time series (signals) is a linear mixture of latent time series (sources) with uncorrelated components. The methods make use of the second‐order statistics—hence the name “second‐order source separation.” In this review, we discuss the classical SOS methods and their extensions to more complex settings. An example illustrates how SOS can be performed.This article is categorized under:Statistical Models > Time Series ModelsStatistical and Graphical Methods of Data Analysis > Dimension ReductionData: Types and Structure > Time Series, Stochastic Processes, and Functional Data

Validation of SOBI-DANS method for automatic identification of horizontal and vertical eye movement components from EEG.

Neurophysiological investigation of neural processes are hindered by the presence of large artifacts associated with eye movement. Although blind source separation (BSS)-based hybrid algorithms are useful for separating, identifying, and removing these artifacts from EEG, it remains unexamined to what extent neural signals can remain mixed with these artifact components, potentially resulting in unintended removal of critical neural signals. Here, we present a novel validation approach to quantitatively evaluate to what extent horizontal and vertical saccadic eye movement-related artifact components (H and V Comps) are indeed ocular in origin. To automate the identification of the H and V Comps recovered by the second-order blind identification (SOBI), we introduced a novel Discriminant ANd Similarity (DANS)-based method. Through source localization, we showed that over 95% of variance in the SOBI-DANS identified H and V Comps' scalp projections were ocular in origin. Through the analysis of saccade-related potentials (SRPs), we found that the H and V Comps' SRP amplitudes were finely modulated by eye movement direction and distance jointly. SOBI-DANS' component selection was in 100% agreement with human experts' selection and was 100% successful in component identification across all participants indicating a high cross-individual consistency or robustness. These results set the stage for future work to transform the to-be-thrown-away artifacts into signals indicative of gaze position, thereby providing readily co-registered eye movement and neural signal without using a separate eye tracker.

Second order blind identification algorithm with exact model order estimation for harmonic and interharmonic decomposition with reduced complexity

The power distribution network is susceptible to several Power Quality (PQ) disturbances. Among those, the harmonic and interharmonic distortions should be highlighted due to their high proliferation. This work proposes the utilization of signal processing techniques to decompose the electrical voltage and/or current signals into its harmonic and interhamonic component waveforms through a Blind Source Separation (BSS) algorithm named Second Order Blind Identification (SOBI). This algorithm is normally applied to a multivariate data set, what implies in a necessity of multiple measurements in different points of the system that will be analyzed. However, Single-Channel Blind Source Separation (SCBSS) method will be proposed in this work to estimate the components via SOBI using only one measured signal point. The method works as a set of adaptive filters whose coefficients are blindly obtained via SOBI and is responsible for the components separation. An Exact Model Order (EMO) algorithm will be used to improve the performance of the SOBI algorithm in order to estimate the correct number of components to be separated. Also, the EMO will be helpful to reduce the computational complexity of the SOBI. The performance of the proposed SCBSS method will be compared to that of the SCICA (Single-Channel Independent Component Analysis) based on the well-known FastICA algorithm, which employs Higher Order Statistics (HOS). It will be shown that the proposed SCBSS overtakes the SCICA for harmonic and interharmonic decomposition in performance and in reduced computational complexity. Also, the proposed SCBSS method will be compared to EMO-ESPRIT algorithm, where will be shown that the SCBSS achieved better results in noisy and time-varying scenarios. Finally, the proposed SCBSS will be applied for the analysis of a voltage signal acquired from the simulation of a power system containing wind generation.

Fault Diagnosis of Rotating Machinery Based on Multi-Sensor Signals and Median Filter Second-Order Blind Identification (MF-SOBI)

Feature extraction plays a crucial role in the diagnosis of rotating machinery faults. However, the vibration signals measured are inherently complex and non-stationary and the features of faulty signals are often submerged by noise. The principle and method of blind source separation are introduced, and we point out that the blind source separation algorithm is invalid in an environment of strong impulse noise. In order to solve the problem of fast separation of multi-sensor signals in an environment of strong impulse noise, first, the window width of the median filter (MF) is calculated according to the sampling frequency, so that the impulse noise and part of the white noise can be effectively filtered out. Next, the filtered signals are separated by the improved second-order blind identification (SOBI) algorithm. At the same time, the method is tested on the strong pulse background noise and rub impact dataset. The results show that this method has higher efficiency and accuracy than the direct separation method. It is possible to apply the method to real-time signal analysis due to its speed and efficiency.

ImpediBands: Body Coupled Bio-Impedance Patches for Physiological Sensing Proof of Concept.

Continuous and robust monitoring of physiological signals is essential in improving the diagnosis and management of cardiovascular and respiratory diseases. The state-of-the-art systems for monitoring vital signs such as heart rate, heart rate variability, respiration rate, and other hemodynamic and respiratory parameters use often bulky and obtrusive systems or depend on wearables with limited sensing methods based on repetitive properties of the signals rather than the morphology. Moreover, multiple devices and modalities are typically needed for capturing various vital signs simultaneously. In this paper, we introduce ImpediBands: small-sized distributed smart bio-impedance (Bio-Z) patches, where the communication between the patches is established through the human body, eliminating the need for electrical wires that would create a common potential point between sensors. We use ImpediBands to collect instantaneous measurements from multiple locations over the chest at the same time. We propose a blind source separation (BSS) technique based on the second-order blind identification (SOBI) followed by signal reconstruction to extract heart and lung activities from the Bio-Z signals. Using the separated source signals, we demonstrate the performance of our system via providing strong confidence in the estimation of heart and respiration rates with low RMSE (HRRMSE = 0.579 beats per minute, RRRMSE = 0.285 breaths per minute), and high correlation coefficients (rHR = 0.948, rRR = 0.921).

A deep wavelet sparse autoencoder method for online and automatic electrooculographical artifact removal

Electrooculographical (EOG) artifacts are problematic to electroencephalographical (EEG) signal analysis and degrade performance of brain–computer interfaces. A novel, robust deep wavelet sparse autoencoder (DWSAE) method is presented and validated for fully automated EOG artifact removal. DWSAE takes advantage of wavelet transform and sparse autoencoder to become a universal EOG artifact corrector. After being trained without supervision, the sparse autoencoder performs EOG correction on time–frequency coefficients collected after brain wave signal wavelet decomposition. Corrected coefficients are then used for wavelet reconstruction of uncontaminated EEG signals. DWSAE is compared with five other methods: second-order blind identification, information maximization, joint approximation diagonalization of eigen-matrices, wavelet neural network (WNN) and wavelet thresholding (WT). Experimental results on a visual attention task dataset, a mental state recognition dataset and a semi-simulated contaminated EEG dataset show that DWSAE is capable of suppressing EOG artifacts effectively, while preserving the nature of background EEG signals. The mean square error of signals before and after correction by DWSAE on a semi-simulated contaminated EEG segment of 30 s is the lowest (65.62) when compared to the results produced by WNN and WT. DWSAE addresses limitations posed by these methods in three ways. First, DWSAE can be performed automatically and online in a single channel of EEG data; this has advantages over independent component analysis-based methods. Second, its results are robust and stable in comparison with those of other wavelet-based methods. Third, as an unsupervised learning scheme, DWSAE does not require the off-line training that is necessary for WNN and other supervised learning machine learning-based methods.

Bridge modal identification using acceleration measurements within moving vehicles

Vehicles commuting over bridge structures respond dynamically to the bridge’s vibrations. An acceleration signal collected within a moving vehicle contains a trace of the bridge’s structural response, but also includes other sources such as the vehicle suspension system and surface roughness-induced vibrations. This paper introduces two general methods for the bridge system identification using data exclusively collected by a network of moving vehicles. The contributions of the vehicle suspension system are removed by deconvolving the vehicle response in frequency domain. The first approach utilizes the vehicle transfer function, and the second uses ensemble empirical modal decomposition (EEMD). Next, roughness-induced vibrations are extracted through a novel application of second-order blind identification (SOBI) method. After these two processes, the resulting signal is equivalent to the readings of mobile sensors that scan the bridge’s dynamic response. Structural modal identification using mobile sensor data has been recently made possible with the extended structural modal identification using expectation maximization (STRIDEX) algorithm. The processed mobile sensor data is analyzed using STRIDEX to identify the modal properties of the bridge. The performance of the methods are validated on numerical case studies of a long single-span bridge with a network of moving vehicles collecting data while in motion. The analyses consider three road surface roughness patterns. Results show that for long-span bridges with medium- to high-ongoing traffic volume, the proposed algorithms are successful in extracting pure bridge vibrations, and produce accurate and comprehensive modal properties of the bridge. The study shows that the proposed transfer function method can efficiently deconvolve the linear dynamics of a moving vehicle. EEMD method is able to extract vehicle dynamic response without a priori information about the vehicle. In addition, proposed identification methods provide secondary information about the roughness pattern and the vehicle. This study is the first proposed methodology for complete bridge modal identification, including operational natural frequencies, mode shapes and damping ratios using moving vehicles sensor data.

A Multi-Model Based Approach for the Detection of Subtle Structural Damage Considering Environmental Variability

Minor structural damages like incipient cracks are difficult to detect as they alter the structural stiffness marginally. It is difficult to extract the features of minor damage, from the measured time history responses, which are usually contaminated by measurement noise. Also, the effect of environmental/operational variabilities often misleads the damage diagnostic process, especially for subtle damages. To tackle all the above challenges in detecting and locating the minor incipient damage, an automated multi-model based data-driven technique is proposed in this paper. It is based on the fact that a subtle damage alters only some structural modes, while the others remain unaltered. Hence, the proposal here is to decompose the measured time-history response into the modal components and then reconstruct the signal using only the modal components with the damage sensitive features. The reconstructed signal is used in damage diagnosis. As an improved version of the second-order blind identification, the blind source separation technique is proposed in this paper for signal decomposition and a crisp automated algorithm is presented for isolating the modal components with damage sensitive features. The autoregressive moving average with exogenous input model with the cepstral distance as the damage index is employed for localizing the damage. The proposed multi-model approach is completely automated. Numerical studies have been carried out to demonstrate the effectiveness of the proposed algorithm. Also, experimental studies have been conducted to ensure the practicality of the proposed technique.

Prediction of random dynamic loads using second-order blind source identification algorithm

Traditional load identification methods are based on the frequency response function matrix. However, in some cases, it is impossible to measure the frequency response functions directly, where only the measured structural dynamic response data are available. In this paper, a novel frequency domain method based on second-order blind source identification (SOBI) algorithm is proposed for identifying the random dynamic loads from some dynamic responses of limited test points. Firstly, the SOBI algorithm is applied to identify the modal parameters from the time histories of the measured displacement responses and then the modal loads are estimated by the identified modal parameters and modal responses in the modal space; finally, the random dynamic loads can be identified in the frequency domain. In order to control the error propagation, the theoretical formulas of the regularization process have been deduced, and the regularization parameters are selected by the generalized cross-validation method. A numerical simulation and an eight-storey spatial frame experimental model are studied to validate the proposed method; the comparison results show a good agreement between the identified random dynamic loads and the actually exerted loads.

Arduino Due Implementation of an Algorithm for Blind Source Separation using Matlab Simulink

This work describes the implementation of an algorithm of the blind source separation in the Adruino Due card as free and low cost material. The algorithm must be developed to adapt the inputs and outputs of the Arduino board to the actual installation. Using reference signals generated by two generators (GBF), this work has proved that the algorithm Second Order Blind Identification (SOBI) is the most efficient to analyze our signal database, so the separation is performed on the Arduino Due using the SOBI algorithm to validate its performance in real time.

An Improved Second-Order Blind Identification (SOBI) Signal De-Noising Method for Dynamic Deflection Measurements of Bridges Using Ground-Based Synthetic Aperture Radar (GBSAR)

Ground-based synthetic aperture radar (GBSAR) technology has been widely used for bridge dynamic deflection measurements due to its advantages of non-contact measurements, high frequency, and high accuracy. To reduce the influence of noise in dynamic deflection measurements of bridges using GBSAR—especially for noise of the instantaneous vibrations of the instrument itself caused by passing vehicles—an improved second-order blind identification (SOBI) signal de-noising method is proposed to obtain the de-noised time-series displacement of bridges. First, the obtained time-series displacements of three adjacent monitoring points in the same time domain are selected as observation signals, and the second-order correlations among the three time-series displacements are removed using a whitening process. Second, a mixing matrix is calculated using the joint approximation diagonalization technique for covariance matrices and to further obtain three separate signal components. Finally, the three separate signal components are converted in the frequency domain using the fast Fourier transform (FFT) algorithm, and the noise signal components are identified using a spectrum analysis. A new, independent, separated signal component matrix is generated using a zeroing process for the noise signal components. This process is inversely reconstructed using a mixing matrix to recover the original amplitude of the de-noised time-series displacement of the middle monitoring point among three adjacent monitoring points. The results of both simulated and on-site experiments show that the improved SOBI method has a powerful signal de-noising ability.

Rotor-Mechanical Coupled Fault Feature Extraction Based on Second-order Blind Identification

Noise reduction usually is conducted before analysis of mechanical fault feature, which could damage effective signals. This article proposes an algorithm of blind source separation based on the second-order statictics. The method focuses on noise separation rather than noise removal. So there are no harms to effective signals. This idea might provide a new way for noise reduction. The algorithm of blind source separation based on the second-order statistics blind identification is applied to seismic data. The results show that the algorithm is effect, noises are separated and re-moved, and the rotor fault feature is picked up.

Remove Diverse Artifacts Simultaneously From a Single-Channel EEG Based on SSA and ICA: A Semi-Simulated Study

Electroencephalogram (EEG) signals are often contaminated with diverse artifacts, such as electromyogram (EMG), electrooculogram (EOG), and electrocardiogram (ECG) artifacts. These artifacts make subsequent EEG analysis inaccurate and prevent practical usage. Recently, the use of wearable EEG devices in ambulatory systems has been developed. For practical reasons, these systems usually contain a single EEG channel. Several studies have proposed to combine single-channel decomposition methods with blind source separation (BSS) methods to denoise the single-channel EEG. However, the existing methods have their own limitations since most of them only focus on removing one single kind of artifacts. Unfortunately, the EEG is prone to be contaminated by various kinds of artifacts simultaneously. Yet to our knowledge, there are no existing methods to remove diverse artifacts simultaneously from the single-channel EEG. To address this issue, we propose an effective method to remove diverse artifacts simultaneously for the single-channel EEG case. This method is a combination of singular spectrum analysis (SSA) and second-order blind identification (SOBI) method. We conduct a semi-simulated study to investigate all possible cases of the single-channel EEG been contaminated by EMG, EOG, and ECG artifacts. The results show that the proposed method can successfully remove diverse artifacts from the single-channel EEG. It is a promising tool for biomedical signal processing applications.

Driver Drowsiness Detection Using Multi-Channel Second Order Blind Identifications

It is well known that blink, yawn, and heart rate changes give clue about a human's mental state, such as drowsiness and fatigue. In this paper, image sequences, as the raw data, are captured from smart phones which serve as non-contact optical sensors. Video streams containing subject's facial region are analyzed to identify the physiological sources that are mixed in each image. We then propose a method to extract blood volume pulse and eye blink and yawn signals as multiple independent sources simultaneously by multi-channel second-order blind identification (SOBI) without any other sophisticated processing, such as eye and mouth localizations. An overall decision is made by analyzing the separated source signals in parallel to determine the driver's driving state. The robustness of the proposed method is tested under various illumination contexts and a variety of head motion modes. Experiments on 15 subjects show that the multi-channel SOBI presents a promising framework to accurately detect drowsiness by merging multi-physiological information in a less complex way.

Let home nursing assistant robots see your heart rate

Purpose This paper aims to design a vision-based non-contact real-time accurate heart rate (HR) measurement framework for home nursing assistant. Design/methodology/approach The study applied Second-Order Blind Signal Identification (SOBI) algorithm to extract remote HR signal and analyzed it with Fast Fourier Transform (FFT). Multiple regions of interest are chosen and analyzed to obtain a more accurate result. Findings An accurate non-contact hear rate (HR) measurement framework is proposed and proved to be efficient. Originality/value The contributions of this HR measurement framework are as follows: accurate measurement of HR, real-time performance, robust under various scenes such as conversation, lightweight computation which is suitable and necessary for home nursing assistance. This framework is designed to be flexibly used in various real-life scenes such as domestic health assistance and affectively intelligent agents and is proved to be robust under such scenes.

A Computationally Efficient Blind Source Extraction Using Idempotent Transformation Matrix

Blind source separation (BSS) problem is an open area of research that requires further investigations. Various algorithms were presented in the literature based on second-order statistics and higher-order statistics. The computational complexity of those methods is a challenging task and must be carefully considered to produce fast BSS algorithms. In blind source extraction (BSE) using linear predictors, the adaptive filter update requires complex computations that need consideration. This work focus on new BSE using the idempotent transformation matrix. New algorithm is presented in this work to compute the matrix with less computational complexity as compared with the standard singular value decomposition method. New optimization problem was defined according to the proposed matrix equation, and solved by an iterative algorithm with low computational complexity. The proposed method is tested using speech and white Gaussian signals. The performance measures used in this work are the signal-to-interference ratio, signal-to-distortion ratio, and signal-to-artifact ratio. Simulation results show that the proposed algorithm significantly separate the source signals with better performance measures as compared with the state-of-the-art approaches such as second-order blind identification and fast independent component analysis.