Study on Translation Transformation Wavelet Shrinkage De-noising Algorithm

摘要

In navigation field, the signals included jump discontinuities are always required to be de-noised. For this kind of signals, normal wavelet shrinkage method will result Pseudo-Gibbs phenomena nearby the jump discontinuities in the de-noised signals. In order to eliminate Pseudo-Gibbs phenomena, this paper presents a translation transformation wavelet shrinkage de-noising algorithm. This algorithm divides signal into several segments according to jump discontinuities and performs translation for each segment to get rid of jump discontinuities. Then wavelet shrinkage is used to de-noise the signal. Finally reverse translation is adopted to offset the translation. Simulation results and measurement data show that this method can effectively eliminate Pseudo-Gibbs phenomena and improve SNR.