Single channel chaotic signal blind source separation method

Abstract

The invention provides a single channel chaotic signal blind source separation method, and belongs to the technical field of electronic information. The separation method is characterized in that modules used in the method comprise an ensemble empirical mode decomposition module, a correlation analysis module, a principal component analysis module and an independent component analysis module. The separation method comprises the steps that single channel signals formed by mixing multichannel chaotic signals are decomposed to mutichannel intrinsic mode function components through the empirical mode decomposition module; first time dimensionality reduction is performed on pivot elements of the multichannel intrinsic function components by using correlation analysis; then second time dimensionality reduction is performed on the pivot elements of the multichannel intrinsic function components by using a main component analysis method; finally blind source separation of the mixture chaotic signals is completed by using independent component analysis technics, and source signals are restored. The separation method has the advantages that the number of transmission channels can be effectively decreased, the purposes of reducing the complexity of hardware and saving equipment cost are achieved, and the source signals can be restored quickly and effectively.

Description

technical field [0001] The invention belongs to the technical field of electronic information, and in particular relates to a single-channel chaotic signal blind source separation method. Background technique [0002] The waveform of the chaotic signal is very irregular. It looks like noise on the surface, but in fact it is produced by deterministic rules, which are sometimes very simple. It is this simple rule that produces complex waveforms that has sparked so much interest in it. Chaos has been a new branch of science since the 1970s. [0003] For the transmission of multi-channel chaotic signals, some use multiple interfaces and connections, but the cost is high; or use multiplexing and demultiplexing techniques, but the disadvantage of these methods is that the equipment complexity is high. Therefore, in real life, due to conditions and cost constraints, the single-channel blind source separation method is often used, that is, after the multi-channel source signals ar...

AI Technical Summary

Problems solved by technology

[0005] Overall Empirical Mode Decomposition (Ensemble Empirical Mode Decomposition, EEMD) is to decompose a single-channel mixed signal into multiple intrinsic mode functions (IMFs), and construct a virtual multi-channel. The separation effect has been significantly improved, and a good separation and recovery effect can be obtained for signals with overlapping spectrums. However, in actual experiments, it is found that when the IMFs component is directly processed by ICA, the number of algorithm iterations is too high and the convergence speed is slow. question

Therefore, in the process of using the EEMD_ICA method, people need to select signals based on experience, but the intelligence is not high.

Images