With the proposal of the "double carbon" goal, a higher proportion of renewable energy and power electronic equipment will be connected to the grid in the future. This will cause the problem of wide-band oscillation. Therefore, according to the characteristics of "strong noise" and "wide band", a monitoring method of power system wide-band oscillation based on variational mode decomposition and compressive sensing is proposed. The variational mode decomposition (VMD) method is improved to adaptively determine the number of mode decompositions, reduce noise and monitor the effective information of the oscillation signal. If the wide-band oscillation is detected, the wide-band oscillation data after noise reduction is uploaded by a compressed sensing (CS) method. The compressed data is reconstructed in the dispatching center to accurately recover the wide-band oscillation signal. This is convenient for the subsequent analysis of the dispatching station. Numerical examples show that the proposed method can maintain the quality of wide-band oscillation monitoring when there is high-intensity random noise, overcome the limitation of data transmission bandwidth after high-speed sampling, and has a good application to wide-band oscillation signal monitoring of an actual power system. This work is supported by the Youth Fund of National Natural Science Foundation of China (No. 52107113).