Phase Difference Measurement Method for Sine Signals Based on Fast ICA

Noise is the key to cause phase measurement error of sine signals. Although there have been many attempts to suppress noise interference, high accuracy is still hard to meet in terrible noise environment. In order to improve reliability and accuracy of the measurement system, a robust method is proposed. In this paper, independent component analysis (ICA) is used to separate sine signal from noises. Firstly, an ICA model, which compensate the number of observed signals and its components is established. Then, the relationship of separation order of the independent components with measurement error is revealed, and a sorting algorithm based on priori knowledge is proposed to improve separation accuracy. Experimental results indicate the method is robust and accurate.