To characterize the motion statuses of spin- stabilized, three- axis stabilized and“tumbling”for non- resolved satellites, a method of the motion status indentification based on time series visible multispectral signal is proposed. The corresponding relationship is analyzed between motion statuses and timevariant types of satellite surface material as well as the correlation between material characteristics and the clustering information on the time series dual- color ratio. A discrimination algorithm on types of surface material and an identification algorithm on motion status are proposed on the basis of the time series dualcolor ratio feature in Gaussian mixture clustering. Considering characteristics of background, materials, structure and orbit of the satellite, mathematical simulations for time series multi- spectral signals generated from satellites with the above three motion statuses are operated. The identification algorithm is validated based on the simulated data, and results show that the proposed algorithm is effective.