Abstract: In order to solve the problem of the poor timeliness and the difficulty of accurate noise estimation of EEMD threshold de-noising method, a circular strategies signal extraction method based on CEEMD permutation entropy was proposed. The original signal was processed by CEEMD, and the reconstruction error caused by white noise was suppressed by superimposing the opposite white noise on the signal. The noise component and signal component were determined by calculating permutation entropy of the intrinsic mode functions. Considering the unknown prior knowledge of noise of the signal, the signal extraction method of circular strategies was established on the basis of singular value decomposition without prior knowledge of signal and noise, the useful signal could be well preserved while noise was suppressed. Through the noise reduction of simulation signal and chaotic signal, the results showed that the proposed method not only effectively eliminated the noise interference, but also reduced the loss of useful details.