Abstract:According to the theory of single channel blind source separation (SCBSS), the algorithm based on virtual channel expansion must be established in a known source number, and most algorithms can only separate two source signals. When separating multiple source signals, the performance will deteriorate sharply. Since the existing methods of this kind use only a single algorithm for virtual channel expansion, they cannot retain all the source signals' valuable information and effectively separate the multiple source signals. From the perspective of making the constructed virtual multi-channel signal contain enough information of the source signals as much as possible, this paper proposes a SCBSS algorithm based on improved wavelet packet and variational mode decomposition (IWP-VMD-SCBSS). Firstly, the source number is estimated according to the interval sampling method and the minimum description length (MDL) criterion. Secondly, the signal reconstruction method based on improved wavelet packet decomposition (IWPD) is used to reconstruct multiple purer virtual signals. Then the virtual signals are combined with the first intrinsic mode function (IMF) of two-level variational mode decomposition (VMD) and the original single-channel observed signal to constitute a virtual multi-channel signal. Finally, the joint approximate diagonalization of eigen-matrices (JADE) algorithm is used to process the virtual multi-channel observed signal to achieve BSS and obtain estimated source signals. The simulation results indicate that the IWP-VMD-SCBSS algorithm can achieve a lower symbol error rate (SER) than existing algorithms and lower computational complexity. It can solve the SCBSS problem of multiple communication signals effectively under an unknown source number.

Research on Single-channel Blind Deconvolution Algorithm for Multi-source Signals

Multi-frame Blind Deconvolution Using Sparse Priors

Algorithm for Multichannel Blind Deconvolution Based on Subspace Decomposition

Blind Deconvolution Using an Improved L0 Sparse Representation

Multinoise-type Blind Denoising Using a Single Uniform Deep Convolutional Neural Network.

Single-Channel Signal Separation and Deconvolution with Generative Adversarial Networks

Identifiability of Multichannel Blind Deconvolution and Nonconvex Regularization Algorithm

A smart universal single-channel blind source separation method and applications

Multi-Task Learning for Blind Source Separation

Multichannel blind speech source separation with a disjoint constraint source model

Blind Source Separation with Compressively Sensed Based on SBL

Blind Source Separation of Single-Channel Mixtures via Multi-Encoder Autoencoders

Particle filtering algorithms for single-channel blind separation of convolutionally coded signals

A single-channel blind source separation algorithm based on improved wavelet packet and variational mode decomposition

Blind Separation of Noisy Mixed Speech Based on Independent Component Analysis and Neural Network

Joint Multichannel Deconvolution and Blind Source Separation

Multichannel Sparse Blind Deconvolution on the Sphere

Single-Channel Blind Image Separation Based on Transformer-Guided GAN

A New Mixing Matrix Identification Algorithm for Underdetermined Blind Source Separation

Noisy blind source separation based on adaptive noise removal

Research on Blind Source Separation Based on Natural Gradient Algorithm