In the spectral processing of wind profile radar data, there often appear clutters, intermittent clutters, and the mixture of atmospheric echoes and geomagnetic clutters. In order to effectively restrain and remove the clutters and increase the wind profile radar detection range and accuracy, the effective spectral moment estimation must be conducted on the spectral moments of atmospheric echo spectrums. Through the power spectral analysis of wind profile radar Doppler echoes, a method of wind profiler radar echo spectral data estimation is presented based on the maximum likelihood method. The MATLAB simulation analysis is conducted, and compared with the conventional method, the feasibility and effectiveness of the algorithm is verified. The spectrum moment estimation has been improved significantly by sampling from the lowerlevel volume.