Independent Source Separation of Multichannel Electroencephalogram Based on Neural Network

A neural network method for independent source separation (ISS) of multichannel electroencephalogram (EEG) is proposed in this paper.Using the denoising function of wavelet multiscale decomposition,the high-frequency noises are removed from the original (raw) EEGs.Then the multichannel EEGs are treated as the weighted mixtures and the expression of weight vector is obtained by seeking the local extrema of the fourth-order cumulants (i.e.kurtosis coefficients) of the mixtures.After these process steps,the weighted mixtures are used as the input of neural network,so the independent source of EEGs can be separated one by one.The experimental results show that our method is effective for ISS of multichannel EEGs.

RESEARCH OF QUANTUM GENETIC ALGORITH AND ITS APPLICATION IN BLIND SOURCE SEPARATION

This letter proposes two algorithms: a novel Quantum Genetic Algorithm (QGA)based on the improvement of Han's Genetic Quantum Algorithm (GQA) and a new Blind Source Separation (BSS) method based on QGA and Independent Component Analysis (ICA). The simulation result shows that the efficiency of the new BSS method is obviously higher than that of the Conventional Genetic Algorithm (CGA).

Measurment of gas-liquid two-phase slug flow with a Venturi meter based on blind source separation

We propose a novel flow measurement method for gas–liquid two-phase slug flow by using the blind source separation technique. The flow measurement model is established based on the fluctuation characteristics of differential pressure(DP) signals measured from a Venturi meter. It is demonstrated that DP signals of two-phase flow are a linear mixture of DP signals of single phase fluids. The measurement model is a combination of throttle relationship and blind source separation model. In addition, we estimate the mixture matrix using the independent component analysis(ICA) technique. The mixture matrix could be described using the variances of two DP signals acquired from two Venturi meters. The validity of the proposed model was tested in the gas–liquid twophase flow loop facility. Experimental results showed that for most slug flow the relative error is within 10%.We also find that the mixture matrix is beneficial to investigate the flow mechanism of gas–liquid two-phase flow.

Realization of a source-device-independent quantum random number generator secured by nonlocal dispersion cancellation

Quantum random number generators(QRNGs)can provide genuine randomness by exploiting the intrinsic probabilistic nature of quantum mechanics,which play important roles in many applications.However,the true randomness acquisition could be subjected to attacks from untrusted devices involved or their deviations from the theoretical modeling in real-life implementation.We propose and experimentally demonstrate a source-device-independent QRNG,which enables one to access true random bits with an untrusted source device.The random bits are generated by measuring the arrival time of either photon of the time–energy entangled photon pairs produced from spontaneous parametric downconversion,where the entanglement is testified through the observation of nonlocal dispersion cancellation.In experiment,we extract a generation rate of 4 Mbps by a modified entropic uncertainty relation,which can be improved to gigabits per second by using advanced single-photon detectors.Our approach provides a promising candidate for QRNGs with no characterization or error-prone source devices in practice.

Source-Independent Amplitude-Semblance Full-Waveform Inversion using a Hybrid Time-and Frequency-Domain Approach

Full-waveform inversion is a promising tool to produce accurate and high-resolution subsurface models.Conventional full-waveform inversion requires an accu-rate estimation of the source wavelet,and its computational cost is high.We develop a novel source-independent full-waveform inversion method using a hybrid time-and frequency-domain scheme to avoid the requirement of source wavelet estimation and to reduce the computational cost.We employ an amplitude-semblance objective function to not only effectively remove the source wavelet effect on full-waveform inver-sion,but also to eliminate the impact of the inconsistency of source wavelets among different shot gathers on full-waveform inversion.To reduce the high computational cost of full-waveform inversion in the time domain,we implement our new algorithm using a hybrid time-and frequency-domain approach.The forward and backward wave propagation operations are conducted in the time domain,while the frequency-domain wavefields are obtained during modeling using the discrete-time Fourier trans-form.The inversion process is conducted in the frequency domain for selected frequen-cies.We verify our method using synthetic seismic data for the Marmousi model.The results demonstrate that our novel source-independent full-waveform inversion pro-duces accurate velocity models even if the source signature is incorrect.In addition,our method can significantly reduce the computational time using the hybrid time-and frequency-domain approach compared to the conventional full-waveform inversion in the time domain.

An efficient equivariant adaptive separation via independence algorithm for acoustical source separation and identification

To balance the convergence rate and steadystate error of blind source separation(BSS) algorithms, an efficient equivariant adaptive separation via independence(Efficient EASI) algorithm is proposed based on separating indicator, which was derived from the convergence condition of EASI, and can be used to evaluate the separation degree of separated signals. Furthermore, a nonlinear monotone increasing function between suitable step sizes and separating indicator is constructed to adaptively adjust step sizes, and forgetting factor is employed to weaken effects of data at the initial stage. Numerical case studies and experimental studies on a test bed with shell structures are provided to validate the efficiency improvement of the proposed method. This study can benefit for vibration & acoustic monitoring and control, and machinery condition monitoring and fault diagnosis.