An Inversion Method of Significant Wave Height Based on Radial Basis Function Neural Network

In view of the question that traditional significant wave height inversion method of ocean wave don´t have high precision and its applicable scope is limited, a significant wave height inversion method based on radial basis function neural network is proposed. Assume significant wave height has a linear relationship with the radar image signal-to-noise ratio´s square root, radial basis function neural network is adopt to study and to establish relational function between the two, thereby realizing the significant wave height inversion. The network architecture is designed, data center selection network weight setup and network learning method are discussed in detail. The simulation result shows, compared with the traditional inversion method, a better serviceability and the higher significant wave height inversion precision are obtained in this paper.