Independent component analysis based blind source separation algorithm and its application in the gravity and magnetic signal processing
Nian Zhang, Jing Nie · 2015
The paper defines the blind source separation (BSS) problem, demonstrates the mathematic model and conditions for its solving, and explains the uncertainty of its solving. The paper analyses the algorithm of independent component analysis (ICA), proving EFFICA more suitable for engineering data processing. The paper illustrates the application of EFFCIA to the gravity and magnetic data processing and simulated experiments. The results show the advantage of ICA application in denoising and extraction of weak signal over the traditional gravity and magnetic data processing, worth further research.