Abstract: Electrostatic monitoring is a highly sensitive monitoring method, which can monitor the occurrence of system performance degradation earlier, but the electrostatic signal is weak, and in the actual complex environment is easily disturbed by changes in working conditions and reduces its monitoring capability. In the process of rolling bearing electrostatic monitoring, the effective electrostatic signal is easily drowned by noise, and the fault characteristics are difficult to extract. In order to solve the above problems, a method based on the sparse representation of the clustering shrinkage segmental orthogonal matching tracking algorithm (CcStOMP) is proposed for rolling bearing electrostatic signal fault feature extraction. The method adds a clustering shrinkage mechanism to the segmented orthogonal matching tracking algorithm (StOMP), performs secondary filtering on the selected atoms during the atomic search, updates the support set, and finally solves for the weights and updates the residuals to reconstruct the original electrostatic signal to extract the rolling bearing fault feature components, maintaining fast convergence while improving the accuracy of sparse recovery. By monitoring the measured signals of rolling bearing outer ring and bearing roller faults, a comparison with the StOMP algorithm shows that the CcStOMP algorithm has the advantage of accurately extracting the fault characteristic components of the rolling bearing electrostatic monitoring signals.