Reconstruction of Rainfall Field Using Earth–Space Links Network: A Compressed Sensing Approach

High-precision rainfall information is of great importance for the improvement accuracy numerical weather prediction and monitoring floods mudslides that affect human life. With rapid development satellite constellation networks, there potential reconstructing high-precision fields in large areas by using widely distributed Earth–space link (ESL) networks. In this paper, we have carried out research on an ESL network with compressed sensing (CS) method case a sparse distribution ESLs. Firstly, networks different densities are designed K-means clustering algorithm. The real then reconstructed CS, results compared inverse distance weighting (IDW) show root mean square error (RMSE) correlation coefficient (CC) CS lower than 0.15 mm/h higher 0.999, respectively, when density 0.05 links per kilometer, indicating capable under sampling. Additionally, performance superior to IDW