Tensor-based blind source separation

We proposed novel tensor-based BSS methods, namely TenSOFO and TCBSS. The former is designed for a joint individual differences in scaling (INDSCAL) decomposition, addressing instantaneous (linear) BSS tasks; while the latter efficiently performs a constrained block term decomposition (BTD), aligning with the design of convolutive BSS.

Demo

Please run

• demo_joint_decomposition.m: To see the performance of the proposed joint INDSCAL decomposition of two tensors
• demo_type2_BTD.m: To see the performance of the proposed type2-BTD decomposition
• demo_source_separation.m: To illustrate the performance of tensor-based source seperation algorithms
• demo_EMG.m: To see the performance of our algorithm for EMG source separation, download file "SynthMUAP.mat" from the Release

Reference

This code is free and open source for research purposes. If you use this code, please acknowledge the following paper.

[1] L.T. Thanh, K. Abed-Meraim, P. Ravier, O. Buttelli, A. Holobar. "Tensorial Convolutive Blind Source Separation". Proc. 49th IEEE ICASSP, 2024.

[2] L.T. Thanh, K. Abed-Meraim, P. Ravier, O. Buttelli, A. Holobar. "Joint INDSCAL Decomposition Meets Blind Source Separation". Proc. 49th IEEE ICASSP, 2024.

[3] L.T. Thanh, K. Abed-Meraim, P. Ravier, O. Buttelli, A. Holobar. "Tensor decomposition meets blind source separation". Signal Process., 2024.