Blind source separation of convolutive nonlinear mixtures by flexible spline nonlinear functions

Milani, Fabio; Solazzi, M.; Uncini, Aurelio

doi:10.1109/icassp.2002.5744933

Cited by 5 publications

(11 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nonlinear distortion is suppressed in the first stage assuming some prior conditions. Spline nonlinear functions or spline neural networks have been applied to the linearization process [3], [4]. Furthermore, a maximum likelihood estimator has been applied [5].…”

Section: Introductionmentioning

confidence: 99%

A blind source separation cascading separation and linearization for low-order nonlinear mixtures

Nishiwaki

Nakayama

Hirano

2004 IEEE International Conference on Acoustics, Speech, and Signal Processing

View full text Add to dashboard Cite

A network structure and its learning algorithm have been proposed for blind source separation applied to nonlinear mixtures. Nonlinearity is expressed by low-order polynomials, which are acceptable in many practical applications. A separation block and a linearization block are cascaded. In the separation block, the cross terms are suppressed, and the signal sources are separated in each group, which include its high-order components. The high-order components are further suppressed through the linearization block. A learning algorithm minimizing the mutual information is applied to the separation block. A new learning algorithm is proposed for the linearization block. Simulation results, using 2-channel speech signals, instantaneous mixtures, and 2nd-order post nonlinear functions, show good separation performance.

show abstract

Section: Introductionmentioning

confidence: 99%

A blind source separation cascading separation and linearization for low-order nonlinear mixtures

Nishiwaki

Nakayama

Hirano

2004 IEEE International Conference on Acoustics, Speech, and Signal Processing

View full text Add to dashboard Cite

show abstract

“…The nonlinear mixture model is the same as in [19], [20], [21]. First, the signal sources s i are mixed through linear combination resulting in u j .…”

Section: Network Structurementioning

confidence: 99%

“…In a separation block, a linearization process and a linear separation process are arranged in this order. Spline nonlinear functions or spline neural networks have been applied to the linearization process [19], [20]. Furthermore, a maximum likelihood estimator has been applied [21].…”

Section: Introductionmentioning

confidence: 99%

A cascade form blind source separation connecting source separation and linearization for nonlinear mixtures

Nakayama

Hirano

Nishiwaki

Proceedings of the International Joint Conference on Neural Networks, 2003.

View full text Add to dashboard Cite

A network structure and its learning algorithm have been proposed for blind source separation applied to nonlinear mixtures. The network has a cascade form consists of a source separation block and a linearization block in this order. The conventional learning algorithm is employed for the separation block. A new learning algorithm is proposed for the linearization block assuming 2nd-order nonlinearity. After, source separation, the outputs include the nonlinear components for the same signal source. This nonlinearity is suppressed through the linearization block. Parameters in this block are iteratively adjusted based on a process of solving a 2nd-order equation of a single variable. Simulation results, using 2-channel speech signals and an instantaneous nonlinear mixing process, show good separation performance.

show abstract

“…Now a large number of papers explore solutions to Post Nonlinear Mixing problem (PNL) [6][3] [2]. It is possible to find only few results of the convolutive post nonlinear problem ( [9][10]) and of some static nonlinear problems [5] more complex than PNL. In [1] Jutten and Karhunen review the recent advances in BSS of nonlinear mixing models.…”

Section: Introductionmentioning

confidence: 99%

Nonlinear ICA solutions for convolutive mixing of PNL mixtures

Vigliano

Uncini

Parisi

2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512)

View full text Add to dashboard Cite

This paper introduces an ICA approach to a novel nonlinear convolutive BSS problem. The mixing model considered here is an evolution of the Post Nonlinear one: it is the convolutive mixing of a PNL mixture. The main aim of this paper is to enlarge the set of blind sources separation problems that can be approached by Nonlinear ICA with some stricter mixing environments than the one just widely described in literature. The Flexibility of the algorithm is given by the on line estimation of the score function performed by Spline Neurons.

show abstract

Blind source separation of convolutive nonlinear mixtures by flexible spline nonlinear functions

Cited by 5 publications

References 6 publications

A blind source separation cascading separation and linearization for low-order nonlinear mixtures

A blind source separation cascading separation and linearization for low-order nonlinear mixtures

A cascade form blind source separation connecting source separation and linearization for nonlinear mixtures

Nonlinear ICA solutions for convolutive mixing of PNL mixtures

Contact Info

Product

Resources

About