Extraction of independent components from hybrid mixture: KuicNet learning algorithm and applications

Kung, Sun‐Yuan; Mejuto, C.

doi:10.1109/icassp.1998.675488

Cited by 18 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[21]) and of blind source separation for real-valued signal mixtures (cf. [6,22,23]). In blind deconvolution tasks, there is only one kurtosis value κ i = κ in ( 23), which simplifies the optimization strategy for achieving a deconvolved sequence.…”

Section: On the Extraction Of A Single Complex-valued Sourcementioning

confidence: 99%

“…As stated previously, the relations in ( 23)-( 24) are identical in form to those in the real-valued blind source separation case, where the roles of the real-valued amplitudes {A i } in the complex-valued separation case play identical roles to those of the real-valued combined system coefficients {c i } in the real-valued separation case. Thus, we directly borrow from existing proofs in the literature, such as [22], where it has already been shown that maximization of J(b) under unit-output-power constraints occurs only at points corresponding to an extracted source, such that A i is nonzero for a single index i ∈ {1, ≤, m}. The constraint A i = 1 then follows from the unit-power constraint and (24).…”

Section: Theorem 4 Consider the Single-unit Extraction Criterionmentioning

confidence: 99%

See 1 more Smart Citation

Fixed-Point Algorithms for the Blind Separation of Arbitrary Complex-Valued Non-Gaussian Signal Mixtures

Douglas

2007

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

We derive new fixed-point algorithms for the blind separation of complex-valued mixtures of independent, noncircularly symmetric, and non-Gaussian source signals. Leveraging recently developed results on the separability of complex-valued signal mixtures, we systematically construct iterative procedures on a kurtosis-based contrast whose evolutionary characteristics are identical to those of the FastICA algorithm of Hyvarinen and Oja in the real-valued mixture case. Thus, our methods inherit the fast convergence properties, computational simplicity, and ease of use of the FastICA algorithm while at the same time extending this class of techniques to complex signal mixtures. For extracting multiple sources, symmetric and asymmetric signal deflation procedures can be employed. Simulations for both noiseless and noisy mixtures indicate that the proposed algorithms have superior finitesample performance in data-starved scenarios as compared to existing complex ICA methods while performing about as well as the best of these techniques for larger data-record lengths.

show abstract

Section: On the Extraction Of A Single Complex-valued Sourcementioning

confidence: 99%

Section: Theorem 4 Consider the Single-unit Extraction Criterionmentioning

confidence: 99%

Fixed-Point Algorithms for the Blind Separation of Arbitrary Complex-Valued Non-Gaussian Signal Mixtures

Douglas

2007

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

show abstract

“…Instead of directly minimizing the MSE, which requires training signal or analog channel estimation, [7] has proposed minimizing the mean dispersion of the sampler output signal, i.e., (18) where is a constant corresponding to the source statistics. ( ; for details, see [11].)…”

Section: A Dm-timing Offsetsmentioning

confidence: 99%

“…However, this result cannot be applied to the timing offset problem here, since the channel vectors parameterized by timing offsets are constrained. In fact, the geometrical arguments used to show the near-optimal MSE performance of dispersion minimization [17], [18] fail for the constrained cases [10]. On the other hand, due to the nonlinearity of the cost functions for DM-timing and MMSE-timing, it is a virtually impossible task to establish a rigorous MSE bound for a general channel.…”

Section: Performance Of Dm-timingmentioning

confidence: 99%

Timing phase offset recovery based on dispersion minimization

Chung

Sethares

Johnson

2005

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

Abstract-This paper presents a blind timing phase offset recovery scheme that attempts to optimize the baud spaced equalizer output mean square error (MSE) for a realistic equalizer length that is usually shorter than the ideal length. Among the existing blind timing recovery schemes, few are designed for equalizer output MSE optimization, and none are designed for the realistic case when the equalizer is short. The proposed algorithm (that is based on a cost function that minimizes the dispersion of the received signal) attempts to minimize the MSE of a one-tap equalizer output. It also exhibits good performance for relatively short equalizers. Conditions for the unimodality of the dispersion minimization cost are investigated, and a geometric relationship to the minimum MSE (MMSE) timing offset is shown qualitatively. The detailed MSE performance of the algorithm is investigated for the representing classes of channels by comparing existing blind timing offset estimation schemes.

show abstract

“…Alternatively, if J (W) is a contrast function, then (5) and (6) forms the framework for contrast-based blind source separation of prewhitened instantaneous mixtures [33][34][35][36].…”

Section: Grassmann and Stiefel Manifoldsmentioning

confidence: 99%

Gradient Adaptive Paraunitary Filter Banks for Spatio-Temporal Subspace Analysis and Multichannel Blind Deconvolution

Douglas

Амари

Kung

2004

The Journal of VLSI Signal Processing-Systems for Signal, Image

View full text Add to dashboard Cite

Abstract. Paraunitary filter banks are important for several signal processing tasks, including coding, multichannel deconvolution and equalization, adaptive beamforming, and subspace processing. In this paper, we consider the task of adapting the impulse response of a multichannel paraunitary filter bank via gradient ascent or descent on a chosen cost function. Our methods are spatio-temporal generalizations of gradient techniques on the Grassmann and Stiefel manifolds, and we prove that they inherently maintain the paraunitariness of the multichannel adaptive system over time. We then discuss the necessary practical approximations, modifications, and simplifications of the methods for solving two relevant signal processing tasks: (i) spatio-temporal subspace analysis and (ii) multichannel blind deconvolution. Simulations indicate that our methods can provide simple, useful solutions to these important problems.

show abstract

Extraction of independent components from hybrid mixture: KuicNet learning algorithm and applications

Cited by 18 publications

References 4 publications

Fixed-Point Algorithms for the Blind Separation of Arbitrary Complex-Valued Non-Gaussian Signal Mixtures

Fixed-Point Algorithms for the Blind Separation of Arbitrary Complex-Valued Non-Gaussian Signal Mixtures

Timing phase offset recovery based on dispersion minimization

Gradient Adaptive Paraunitary Filter Banks for Spatio-Temporal Subspace Analysis and Multichannel Blind Deconvolution

Contact Info

Product

Resources

About