中国近岸RFI源对星载盐度计影响机制分析及动态递归抑制研究

The Radio Frequency Interference (RFI) sources as an important influence factor for satellite observations of sea surface salinity in China's coastal waters, the transmitted interfering noise signals seriously affect the observation accuracy of L-band space-borne ocean salinity meter. Since the RFI weak signal strength is equivalent to the natural radiation signal level, and the aliased signals of multiple RFI sources vary randomly with time, the conventional RFI detection and mitigation methods are difficult to completely suppress the interference of terrestrial RFI sources to the sea surface radiation signals. This project uses a combination of polarization feature detection and statistical analysis to perform land-based RFI source detection and positioning algorithms and sea surface RFI signal detection algorithms, explores the impact of terrestrial RFI source emission intensity on L-Band satellite-borne microwave radiometer observations of sea surface polarization brightness temperature data,studies the multi-source relationship eigenfunction model of the two to reveal the impact mechanism of terrestrial RFI sources on the sea surface. In order to accurately invert the sea surface salinity in the South China Sea, we study the RFI suppression and sea surface salinity inversion algorithm based on the dynamic recurrent neural network model, assess the applicability of the algorithm model, Establishing a sea surface salinity inversion model based on the localization of China’s inshore bureaus to improve the satellite inversion accuracy of sea surface salinity in China. It provides a theoretical basis for the detection and suppression of RFI detection and sea surface salinity inversion by independent ocean salinity satellite in China and has important scientific significance.

无线射频干扰（RFI）源作为中国近岸海域海表盐度卫星观测的重要影响因子，其发射的干扰噪声信号严重影响了L波段星载盐度计的观测精度，由于RFI弱信号强度与自然辐射信号水平相当，且多个RFI源的混叠信号随时间随机变化，传统的RFI检测及减缓方法难以完全抑制陆地RFI源对海面辐射信号的干扰。本项目采用极化特征检测和统计分析相结合的方法进行陆地RFI源检测定位和海面RFI信号检测算法研究，探索陆地RFI源发射强度对L波段星载微波辐射计观测海面极化亮温数据影响特征，研究二者的多源关系特征函数模型，揭示陆地RFI源对海面的影响机理，并以精确反演中国南海海表盐度为目标，研究基于动态递归神经网络模型的RFI抑制及海表盐度反演算法，评估算法的适用性，提出基于中国近岸局地化的海表盐度反演模型，提高我国海域海表盐度卫星反演精度，为我国自主盐度星的RFI检测、抑制及海表盐度反演研究提供理论基础，具有重要科学意义。

星载L波段微波辐射计工作频段是1.413GHz，在全球范围内，该受保护频段目前正在遭受大量无线射频干扰的污染。本项目基于L波段微波辐射计长时间序列交叉极化数据作为数据源，采用极化特征检测和统计分析相结合的方法进行陆地RFI源检测定位和海面RFI信号检测算法研究，构建了一套能够实现RFI检测、聚类、识别及定位的自动化处理系统与算法，能够初步实现对非线性变化的陆地RFI源的信号检测和定位。同时，分析了陆地RFI源发射强度对L波段星载微波辐射计观测海面极化亮温数据影响特征，研究二者的多源关系特征函数模型及陆地RFI源对海面的影响机理，为我国自主盐度星的RFI检测、抑制及海表盐度反演研究提供理论基础，具有重要科学意义。