Convolutive Transfer Function-Based Multichannel Nonnegative Matrix Factorization for Overdetermined Blind Source Separation

Wang, Taihui; Yang, Feiran; Yang, Jun

doi:10.1109/taslp.2022.3145304

Cited by 19 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sparse coding was applied for each

to detect the SSPs. Two assumptions were made to estimate the mixing matrix, i.e., (A1) in the mixing matrix

, any column vector is linearly independent; (A2) there are some TF points

for each source

in which only

is dominant,

[ 19 ]. Based on these assumptions, for each source signal

at any TF point such as

where only one source is active,

is written as

where

is an SSP corresponding to

.…”

Section: Mixing Matrix Estimationmentioning

confidence: 99%

“…Much work has been carried out on what is called SCA [ 14 , 15 , 16 , 17 , 18 ]. Several techniques for obtaining sparsity in the transform domain have been developed thus far, including the short-time Fourier Transform (STFT) and the wavelet packet transform [ 19 , 20 , 21 , 22 , 23 ]. UBSS techniques are highly dependent on the sparsity of the source signal.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Sparse Component Analysis (SCA) Based on Adaptive Time—Frequency Thresholding for Underdetermined Blind Source Separation (UBSS)

Hassan

Ramli

2023

Sensors

View full text Add to dashboard Cite

Blind source separation (BSS) recovers source signals from observations without knowing the mixing process or source signals. Underdetermined blind source separation (UBSS) occurs when there are fewer mixes than source signals. Sparse component analysis (SCA) is a general UBSS solution that benefits from sparse source signals which consists of (1) mixing matrix estimation and (2) source recovery estimation. The first stage of SCA is crucial, as it will have an impact on the recovery of the source. Single-source points (SSPs) were detected and clustered during the process of mixing matrix estimation. Adaptive time–frequency thresholding (ATFT) was introduced to increase the accuracy of the mixing matrix estimations. ATFT only used significant TF coefficients to detect the SSPs. After identifying the SSPs, hierarchical clustering approximates the mixing matrix. The second stage of SCA estimated the source recovery using least squares methods. The mixing matrix and source recovery estimations were evaluated using the error rate and mean squared error (MSE) metrics. The experimental results on four bioacoustics signals using ATFT demonstrated that the proposed technique outperformed the baseline method, Zhen’s method, and three state-of-the-art methods over a wide range of signal-to-noise ratio (SNR) ranges while consuming less time.

show abstract

“…Sparse coding was applied for each

to detect the SSPs. Two assumptions were made to estimate the mixing matrix, i.e., (A1) in the mixing matrix

, any column vector is linearly independent; (A2) there are some TF points

for each source

in which only

is dominant,

[ 19 ]. Based on these assumptions, for each source signal

at any TF point such as

where only one source is active,

is written as

where

is an SSP corresponding to

.…”

Section: Mixing Matrix Estimationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Sparse Component Analysis (SCA) Based on Adaptive Time—Frequency Thresholding for Underdetermined Blind Source Separation (UBSS)

Hassan

Ramli

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…In addition, we noticed that processing multi-channel NMR signals falls within the realm of handling composite signals from multiple detectors. In this domain, blind source separation (BSS) stands out as an exceptional model with diverse applications, including speech recognition 34 , image processing 35 , and bio-medical signal processing 36 . BSS methods, under appropriate assumptions such as the independent sources condition 37 , not only identifies signals from multiple sources, but also estimates the mixing matrix, which reveals the inter-detector coupling information.…”

Section: Introductionmentioning

confidence: 99%

A digital twin for parallel liquid-state nuclear magnetic resonance spectroscopy

He,

Faderl,

MacKinnon

et al. 2024

Commun Eng

View full text Add to dashboard Cite

One approach to increasing nuclear magnetic resonance measurement sample throughput is to implement multiple, independent detection sites. However, the presence of radio frequency interference poses a challenge in multi-detector systems, particularly in unshielded coil arrays lacking sufficient electrical isolation. This issue can lead to unwanted coupling of inductive coils, resulting in excitation pulse interference and signal transfer among multiple detection sites. Here we propose a theoretical framework that combines electromagnetic simulation with spin-dynamic calculations. This framework enables the evaluation of coil coupling effects, the design of parallel pulse sequences to mitigate inter-channel coupling, and the separation of composite free induction decays obtained from multiple detectors. The parallel pulse compensation scheme was validated by a 2-channel parallel spectroscopy experiment. These results provide valuable insights for the design of parallel nuclear magnetic resonance hardware and for exploring the limits of parallelization capacity within a fixed magnet system.

show abstract

“…In the past few decades, various community detection methods have emerged, including methods based on graph partitioning [7,8], methods based on spectral clustering [9], methods based on modularity optimization [10], methods based on The technique of label propagation [11,12] and the way based on deep learning [13]. It is worth pointing out that non-negative matrix factorization (NMF) is also widely used in community discovery due to its simplicity, easy scalability, and strong interpretability [14]. At present, many community discovery methods based on NMF have been proposed, for example, the community method based on naive non-negative matrix factorization [15], the community discovery method SNMF based on symmetric NMF [16], the community discovery method SNMF-SS [17] based on semi-supervised NMF, NMF community discovery method based on node similarity and modularity M-NMF [18] and so on.…”

Section: Introductionmentioning

confidence: 99%

Graph Convolutional Networks based Non-Negative Matrix Factorization aware Community detection

Scholar

Professor

2023

Preprint

View full text Add to dashboard Cite

Nonnegative matrix factorization (NMF) is widely used in community discovery because of its effectiveness and easy interpretability. However, most of the existing NMF-based community detection methods are linear. They cannot effectively deal with the nonlinear characteristics of complex networks, resulting in further improvement in community detection performance. Aiming at this problem, a convolution graph network (GCN) enhanced nonlinear NMF community discovery method NMFGCN is proposed. NMFGCN consists of two main modules: GCN and NMF, where GCN is used to learn network node representations, and NMF uses node representations as input to obtain network community representations. In addition, a joint optimization method is proposed to train NMFGCN, which enables NMFGCN to have nonlinear feature representation capabilities and enables GCN and NMF to promote each other and obtain better community segmentation results. Many experiments on artificial synthetic networks and entire networks show that NMFGCN is superior to current NMF-based community discovery methods , thus proving that NMFGCN can improve the performance of NMF community discovery methods. In addition, NMFGCN also outperforms Deep Walk and LINE standard graph representation learning methods.

show abstract

Convolutive Transfer Function-Based Multichannel Nonnegative Matrix Factorization for Overdetermined Blind Source Separation

Cited by 19 publications

References 48 publications

Sparse Component Analysis (SCA) Based on Adaptive Time—Frequency Thresholding for Underdetermined Blind Source Separation (UBSS)

Sparse Component Analysis (SCA) Based on Adaptive Time—Frequency Thresholding for Underdetermined Blind Source Separation (UBSS)

A digital twin for parallel liquid-state nuclear magnetic resonance spectroscopy

Graph Convolutional Networks based Non-Negative Matrix Factorization aware Community detection

Contact Info

Product

Resources

About