Convolutive blind source separation based on joint block Toeplitzation and block-inner diagonalization

Xu, Xianfeng; Feng, Da‐Zheng; Zheng, Wei Xing; Zhang, Hua

doi:10.1016/j.sigpro.2009.05.021

Cited by 14 publications

(11 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, we focused on the task of spike detection; thus, the complete separation of all sources is not required as it is in the existing approaches. The special structure induced by spareness and convolutive filters is currently still being investigated and only first attempts have been made to fully incorporate it into algorithm design [50,51].…”

Section: Resultsmentioning

confidence: 99%

An Unsupervised and Drift-Adaptive Spike Detection Algorithm Based on Hybrid Blind Beamforming

Natora

Obermayer

2010

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

In the case of extracellular recordings, spike detection algorithms are necessary in order to retrieve information about neuronal activity from the data. We present a new spike detection algorithm which is based on methods from the field of blind equalization and beamforming and which is particularly adapted to the specific signal structure neuronal data exhibit. In contrast to existing approaches, our method blindly estimates several waveforms directly from the data, selects automatically an appropriate detection threshold, and is also able to track neurons by filter adaptation. The few parameters of the algorithm are biologically motivated, thus, easy to set. We compare our method with current state-of-the-art spike detection algorithms and show that the proposed method achieves favorable results. Realistically simulated data as well as data acquired from simultaneous intra/extracellular recordings in rat slices are used as evaluation datasets.

show abstract

Section: Resultsmentioning

confidence: 99%

An Unsupervised and Drift-Adaptive Spike Detection Algorithm Based on Hybrid Blind Beamforming

Natora

Obermayer

2010

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Hence the proposed algorithms have the same quantity level of complexity as convolutive FastICA [22], an extended algorithm of FastICA in instantaneous case. In contrast, another second-order technique called joint block Toeplitzation and block-inner diagonalization (JBTBID) [23], which converges slower than convolutive FastICA, requires 3 3…”

Section: Theoretical Performance Concerning Iterative Separatorsmentioning

confidence: 99%

“…Figure 4 shows the average values of r and ISR obtained by three kinds of iterative separators for circular 4QAM sources, when different initial values of the correct support estimation ratio r 1 are adopted. Also shown are the theoretical approximations given by (34-36) and the corresponding ISR defined by (23)(24) and (29). One can observe that, after several iterations the correct support estimation ratio r converges around 1.0.…”

Section: Performance Of Iterative Algorithmsmentioning

confidence: 99%

Convolutive Independent Component Analysis Based on First-order Statistics for Complex-valued Source Extraction

Xu¹,

Shen²

2016

IJSIP

View full text Add to dashboard Cite

show abstract

“…Jacobi like algorithms were suggested De Lathauwer et al 2002). Then, different alternative methods achieving the same task but for a non unitary joint-block diagonaliser have been proposed (Cherrak et al 2013b;Ghennioui et al 2007;Ghennioui et al 2010;Ghennioui et al 2007b;Lahat et al 2012a;Nion 2011;Tichavsky and Koldovsky 2012;Xu et al 2010;; for video event recognition and object recognition)).…”

Section: Introductionmentioning

confidence: 99%

Preconditioned optimization algorithms solving the problem of the non unitary joint block diagonalization: application to blind separation of convolutive mixtures

Cherrak

Ghennioui

Thirion-Moreau

et al. 2017

Multidim Syst Sign Process

View full text Add to dashboard Cite

This article addresses the problem of the Non Unitary Joint Block Diagonalization (NU − JBD) of a given set of complex matrices for the blind separation of convolutive mixtures of sources. We propose new different iterative optimization schemes based on Conjugate Gradient, Preconditioned Conjugate Gradient, Levenberg-Marquardt and Quasi-Newton methods. We perform also a study to determine which of these algorithms offer the best compromise between efficiency and convergence speed in the studied context. To be able to derive all these algorithms, a preconditioner has to be computed which requires either the calculation of the complex Hessian matrices or the use of an approximation to these Hessian matrices. Furthermore, the optimal stepsize is also computed algebraically to speed up the convergence of these algorithms. Computer simulations are provided in order to illustrate the behavior of the different algorithms in various contexts: when exactly block-diagonal matrices are considered but also when these matrices are progressively perturbed by an additive Gaussian noise. Finally, it is shown that these algorithms enable solving the blind separation of the convolutive mixtures of sources problem. Keywords

show abstract

Convolutive blind source separation based on joint block Toeplitzation and block-inner diagonalization

Cited by 14 publications

References 24 publications

An Unsupervised and Drift-Adaptive Spike Detection Algorithm Based on Hybrid Blind Beamforming

An Unsupervised and Drift-Adaptive Spike Detection Algorithm Based on Hybrid Blind Beamforming

Convolutive Independent Component Analysis Based on First-order Statistics for Complex-valued Source Extraction

Preconditioned optimization algorithms solving the problem of the non unitary joint block diagonalization: application to blind separation of convolutive mixtures

Contact Info

Product

Resources

About