“The spaceborne GNSS-R technology is widely used for observing the significant wave height on the sea surface. Based on the Convolutional Neural Network Gated Recurrent Unit (CNN-GRU) model, researchers use CNN to extract the spatial features of the delayed Doppler map (DDM) and GRU to extract the temporal features of one-dimensional parameters. The spatial and temporal features are concatenated and fused into a comprehensive feature through a fully connected network (FCN) to obtain the significant wave height value, providing a new solution for the inversion of significant wave height in high sea conditions.”