Aiming at the band of rolling bearing vibration signals getting wider, based on the traditional Shannon-Nyquist sampling theorem for data collection, it will get a huge amount of data and produce difficult problems of storage, transmission and processing, we propose a data compression and acquisition method. Firstly, analyzed the sparsity(also called compressibility) of the vibration signal based on orthogonal dictionary Fourier domain; then integrated sparse structure information of the vibration signal into the design of measurement matrix and got its optimized measurement matrix and measured; Finally, based on the compression measurements used orthogonal matching pursuit algorithm to reconstruct the original vibration signal. Through simulation experiments, the results show that the proposed method can get a higher signal compression ratio and has precise signal reconstruction performance, in the case of without losing vibration information, greatly reducing the original amount of vibration data.