GNSS-R海面风速反演算法

针对目前GNSS-R海面风速反演方法观测量单一，精度较低的问题，提出一种基于时延多普勒相关功率图（Delay Doppler map，DDM）的分割多普勒波形匹配反演算法。该方法利用新的DDM观测量，经过训练后，可以反演海面风速。使用该算法根据给定条件建立风场反演模型，以CYGNSS数据作为仿真输入量。仿真结果表明，利用该方法在海面风速20 m/s以下，反演风速的均方根误差为2.01 m/s，结果优于传统GNSS-R风速反演使用的峰值功率法和前缘坡度匹配法（Leading edge slope matching method，LES）。

Retrieval Algorithm for GNSS-R Sea Surface Wind Velocity

To tackle the problem of single measurement and low precision of GNSS-R sea surface wind speed inversion method, a segmentation Doppler waveform integral matching inversion algorithm based on Delay Doppler Map (DDM) is proposed. The method utilizes a new DDM observation parameter and, after training it can invert the sea surface wind speed. The algorithm is used to establish a wind field inversion model based on the given conditions, using CYGNSS data as the simulation input. The simulation results show that the root mean square error of the inversion wind speed is 2.01 m/s at the wind speed of sea surface of 20 m/s, which is better than the peak power method and leading edge slope (LES) matching method used in the traditional GNSS-R wind speed inversion.