To solve the problem of noise measurement in wide area measurement system

we proposed a low-frequency oscillation analysis method based on Stein unbiased risk estimate（SURE） wavelet threshold de-noising and modified complementary ensemble empirical mode decomposition and Hilbert-Huang transform（MCEEMD-HHT）. Furthermore

the proposed method is used to solve the problem of mode mixing and pseudo components in mode identification using Hilbert-Huang transform（HHT） in the analysis of low-frequency oscillation. Firstly

the low-frequency oscillation signal of power network measurement with strong noise is pre-processed by using SURE wavelet threshold de-noising. Secondly

the permutation entropy algorithm is introduced to improve CEEMD to form MCEEMD

which can effectively suppress the phenomenon of modal aliasing and pseudo components in EMD. Finally

HHT analysis is carried out on the real low-frequency modes obtained by MCEEMD decomposition. Through the composite signal tests

IEEE four-generator two-area system simulation

and an actual measured North American power grid data analysis

it is shown that the proposed method is effective in the analysis of power system low-frequency oscillation. In addition

compared with fast Fourier transform（FFT） and Prony algorithm analysis

the proposed method is more accurate in the extraction of modal parameters and does not need an artificial order determination.