An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering

Bingchuan Niu; Chundong Qi

doi:10.1109/EITCE70137.2026.11634568

An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering

Bingchuan Niu
, Chundong Qi^*

^*Corresponding author for this work

Beijing Institute of Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

For single-channel underdetermined blind source separation, an adaptive VMD method guided by k-means clustering is proposed. The method exploits differences in higher-order cumulants to design a cyclic iterative strategy for adaptive optimization of the mode number K, and completes modal reconstruction through a second clustering. Simulations show that when the SNR exceeds 0 dB, the unique correspondence matching rate (UCMR) remains above 94% and the average correlation coefficient exceeds 0.85. Comparative experiments confirm that the method overcomes the dependence on linear mixing models, achieves balanced fidelity between the two reconstructed signals, and exhibits superior robustness.

Original language	English
Title of host publication	2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	69-73
Number of pages	5
ISBN (Electronic)	9798331563967
DOIs	https://doi.org/10.1109/EITCE70137.2026.11634568
Publication status	Published - 2026
Externally published	Yes
Event	10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026 - Guangzhou, China Duration: 12 Jun 2026 → 14 Jun 2026

Publication series

Name	2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026

Conference

Conference	10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026
Country/Territory	China
City	Guangzhou
Period	12/06/26 → 14/06/26

Keywords

blind source separation
higher-order cumulants
k-means clustering
variational mode decomposition

Access to Document

10.1109/EITCE70137.2026.11634568

Cite this

Niu, B., & Qi, C. (2026). An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering. In 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026 (pp. 69-73). (2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/EITCE70137.2026.11634568

Niu, Bingchuan ; Qi, Chundong. / An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering. 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026. Institute of Electrical and Electronics Engineers Inc., 2026. pp. 69-73 (2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026).

@inproceedings{05d521541033467485311f4e1a1821e8,

title = "An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering",

abstract = "For single-channel underdetermined blind source separation, an adaptive VMD method guided by k-means clustering is proposed. The method exploits differences in higher-order cumulants to design a cyclic iterative strategy for adaptive optimization of the mode number K, and completes modal reconstruction through a second clustering. Simulations show that when the SNR exceeds 0 dB, the unique correspondence matching rate (UCMR) remains above 94\% and the average correlation coefficient exceeds 0.85. Comparative experiments confirm that the method overcomes the dependence on linear mixing models, achieves balanced fidelity between the two reconstructed signals, and exhibits superior robustness.",

keywords = "blind source separation, higher-order cumulants, k-means clustering, variational mode decomposition",

author = "Bingchuan Niu and Chundong Qi",

note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026 ; Conference date: 12-06-2026 Through 14-06-2026",

year = "2026",

doi = "10.1109/EITCE70137.2026.11634568",

language = "English",

series = "2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "69--73",

booktitle = "2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026",

address = "United States",

}

Niu, B & Qi, C 2026, An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering. in 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026. 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026, Institute of Electrical and Electronics Engineers Inc., pp. 69-73, 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026, Guangzhou, China, 12/06/26. https://doi.org/10.1109/EITCE70137.2026.11634568

An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering. / Niu, Bingchuan; Qi, Chundong.
2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026. Institute of Electrical and Electronics Engineers Inc., 2026. p. 69-73 (2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering

AU - Niu, Bingchuan

AU - Qi, Chundong

PY - 2026

Y1 - 2026

N2 - For single-channel underdetermined blind source separation, an adaptive VMD method guided by k-means clustering is proposed. The method exploits differences in higher-order cumulants to design a cyclic iterative strategy for adaptive optimization of the mode number K, and completes modal reconstruction through a second clustering. Simulations show that when the SNR exceeds 0 dB, the unique correspondence matching rate (UCMR) remains above 94% and the average correlation coefficient exceeds 0.85. Comparative experiments confirm that the method overcomes the dependence on linear mixing models, achieves balanced fidelity between the two reconstructed signals, and exhibits superior robustness.

AB - For single-channel underdetermined blind source separation, an adaptive VMD method guided by k-means clustering is proposed. The method exploits differences in higher-order cumulants to design a cyclic iterative strategy for adaptive optimization of the mode number K, and completes modal reconstruction through a second clustering. Simulations show that when the SNR exceeds 0 dB, the unique correspondence matching rate (UCMR) remains above 94% and the average correlation coefficient exceeds 0.85. Comparative experiments confirm that the method overcomes the dependence on linear mixing models, achieves balanced fidelity between the two reconstructed signals, and exhibits superior robustness.

KW - blind source separation

KW - higher-order cumulants

KW - k-means clustering

KW - variational mode decomposition

UR - https://www.scopus.com/pages/publications/105048927416

U2 - 10.1109/EITCE70137.2026.11634568

DO - 10.1109/EITCE70137.2026.11634568

M3 - Conference contribution

AN - SCOPUS:105048927416

T3 - 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026

SP - 69

EP - 73

BT - 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026

Y2 - 12 June 2026 through 14 June 2026

ER -

Niu B, Qi C. An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering. In 2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026. Institute of Electrical and Electronics Engineers Inc. 2026. p. 69-73. (2026 10th International Conference on Electronic Information Technology and Computer Engineering, EITCE 2026). doi: 10.1109/EITCE70137.2026.11634568

An Adaptive VMD Method for Single-Channel Underdetermined Blind Source Separation Guided by K-Means Clustering

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this