Cramér–Rao Bounds for Complex-Valued Independent Component Extraction: Determined and Piecewise Determined Mixing Models

Kautský, Václav; Koldovsky, Zbynek; Tichavský, Petr; Zarzoso, Vicente

doi:10.1109/tsp.2020.3022827

Cited by 14 publications

(23 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The bound in (5) has order O(N −1 ). This is in contrast to the conventional static mixture-based BSE where data are independently processed in each block, which ensures time-variant demixing on one hand, but on the other hand yields ISR whose lower bound is obviously O(N −1 b ) [11]. Moreover, there is the ambiguity problem that a different source other than the SOI can be extracted in each block when they are processed independently.…”

Section: Csv Mixing Modelmentioning

confidence: 95%

“…The reduced parameterization gives the CSV an advantage over the models with time-variant separating vectors. To justify, it has been shown in [11] that the achievable residual interference-to-signal ratio (ISR) after the extraction based on CSV (for one data-set only, i.e., K = 1), is lower-bounded in the sense that…”

Section: Csv Mixing Modelmentioning

confidence: 99%

“…FastDIVA currently appears, to our best knowledge, to be the fastest method for the dynamic case T > 1. Performance analysis of the algorithm provided in [13] has shown that the asymptotic mean ISR achieved within the kth data set is equal to the Cramér-Rao-induced lower bound from [11] provided that the true score function of the SOI is used.…”

Section: One-unit Fastdivamentioning

confidence: 99%

“…The model, referred to as CSV (Constant Separating Vector) assumes time-varying mixing and time-invariant separating parameters in the determined model [10]. Some advantageous theoretical properties as well as practical usefulness of CSV have been reported in [11][12][13]. However, the practical potential of CSV has still not been fully explored.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Blind Extraction of Moving Sources via Independent Component and Vector Analysis: Examples

Amor

Cmejla

Kautský

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper is devoted to the recently proposed mixing model with constant separating vector (CSV) for Blind Source Extraction of moving sources using the FastDIVA algorithm, which is an extension of the famous FastICA and FastIVA for static mixtures. The benefits due to the CSV model and FastDIVA are demonstrated in three new applications. First, the extraction of a moving speaker in a noisy reverberant environment using a dense array of 48 MEMS microphones is considered. Second, a case study on the blind extraction of moving brain activity from visually evoked potentials in electroencephalogram is reported. Third, a simulation of block-byblock online extraction of a moving source is demonstrated. In these examples, the CSV and FastDIVA show their new potential and good performance in handling the blind moving source extraction problem.

show abstract

Section: Csv Mixing Modelmentioning

confidence: 95%

Section: Csv Mixing Modelmentioning

confidence: 99%

Section: One-unit Fastdivamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Blind Extraction of Moving Sources via Independent Component and Vector Analysis: Examples

Amor

Cmejla

Kautský

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Recently, an alternative approach was proposed based on the constant separating vector (CSV) model [5], which allows for changes of mixing parameters within the processed block of data. The CSVbased method can thus operate with longer intervals and achieves, in theory [6], higher accuracy compared the block-wise approach.…”

Section: Introductionmentioning

confidence: 99%

Blind Extraction of Moving Audio Source in a Challenging Environment Supported by Speaker Identification Via X-Vectors

Málek

Janský

Kounovsky

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

We propose a novel approach for semi-supervised extraction of a moving audio source of interest (SOI) applicable in reverberant and noisy environments. The blind part of the method is based on independent vector extraction (IVE) and uses the recently proposed constant separating vector (CSV) mixing model. This model allows for changes of mixing parameters within the processed interval of the mixture, which potentially leads to higher accuracy of SOI estimation. The supervised part of the method concerns a pilot signal, which is related to the SOI and ensures the convergence of the blind method towards the SOI. The pilot is based on robust detection of frames where SOI is dominant via speaker embeddings called X-vectors. Robustness of the detection is achieved through augmentation of the data for the supervised training of the X-vectors. The pilot-supported extraction yields significantly better performance compared to its unsupervised counterpart identifying SOI solely using the initialization.

show abstract

Analysis of asymptotic and exact Cramér-Rao lower bounds on the parameters of a noisy periodic signals

Belega

Petri

2023

Digital Signal Processing

View full text Add to dashboard Cite

Cramér–Rao Bounds for Complex-Valued Independent Component Extraction: Determined and Piecewise Determined Mixing Models

Cited by 14 publications

References 52 publications

Blind Extraction of Moving Sources via Independent Component and Vector Analysis: Examples

Blind Extraction of Moving Sources via Independent Component and Vector Analysis: Examples

Blind Extraction of Moving Audio Source in a Challenging Environment Supported by Speaker Identification Via X-Vectors

Analysis of asymptotic and exact Cramér-Rao lower bounds on the parameters of a noisy periodic signals

Contact Info

Product

Resources

About