Abstract: Sea clutter is the sea surface electromagnetic scattering echo collected by the radar on the ocean surface. Affected by marine environmental factors(wind speed, wind direction, wave height, wave direction, etc.) and radar parameters, its amplitude has fluctuations with time and the accuracy of sea clutter amplitude prediction helps to increase target detection accuracy. Combined with the non-Gaussian and nonlinearity characteristics of sea clutter, the sea clutter amplitude prediction method based on gated feedback recurrent neural network is proposed. Through the prediction analysis of the sea clutter data, the results show that the proposed method has higher prediction accuracy than the traditional prediction methods.