To address the issue of weak features related to faulty rotating parts in Instantaneous Angular Speed（IAS） signal， this study proposes a Average Down-Sampling Multi-Period Differential Means（ADSMPDM） scheme to enhance fault features. First? ly， based on the estimation characteristics of the IAS， the average down-sampling of the IAS signal is studied and its features of suppressing random noise are obtained. Secondly， the ADSMPDM scheme is proposed to enhance the features related to the fault in the IAS signal based on the advantages of the average down-sampling （such as noise suppression， low computational cost and low storage space） and accumulative characteristic of multi-period differential means. Finally， the features related to the fault are re? vealed by order spectrum analysis. By using Simulations and experiments and comparing with fast kurtogram， multipoint optimal minimum entropy deconvolution adjusted， discrete random separation and spectral amplitude modulation， the effectiveness and ad? vantages of the ADSMPDM algorithm in enhancing gear and bearing fault feature components are verified.