新兴GNSS陆面参数遥感研究备受关注，但空地联合与Multi-GNSS协同植生参数感知研究还很缺乏。本项目面向北斗卫星导航系统，围绕1个核心、4项任务、3类建模算法和2种典型应用中涉及的科学问题予以深入研究。以双基雷达模式GNSS信号感知植生参数理论方法为核心目标，拟重点研究①Space-surface联合模式Multi-GNSS机会信号时空可用性，②Multi-GNSS植生参数感知建模机器学习算法（支持向量回归机、SOM-BP级联神经网络），③LEO/UAV/Land联合Multi-GNSS信号感知植生参数智能融合方法（改进的EM+自适应混合高斯滤波），④典型场景（农田、煤矿开采影响区）、典型作物（小麦、水稻、土豆）、典型地类（低矮稀疏林地、草地）植生参数时序特征提取。项目成果将为BDS陆面参数感知创新研究与应用提供关键技术支撑，具有重要的理论意义和实用价值。

It is promising and essential to study on Global Navigation Satellite System (GNSS) signals as sources of opportunity for remotely sensing biophysical parameters of vegetation or crop. However, there are still some deficiencies in retrieving high temporal resolution land surface parameters with multi-GNSS and space-surface bistatic radar. Focusing on one core objective, four research contents, three modelling algorithms and two representative applications, this project intends to study on alternative methodology for reflectometry of biophysical parameters. In particular, it attempts to study on the availability of opportunistic signals with respect to LEO/UAV/Land-borne platforms, machine learning modelling algorithms of biophysical parameters such as Support Vector Regression Machine, SOM-BP neural network, adaptive Gaussian Mixture Model filter supported by improved EM for fusing biophysical parameters by GNSS space-surface bistatic radar sensing, as well as feature retrieval of time series biophysical parameters within context of cropland or coal mined farmland, lower and sparse canopy or grassland. The entire study is expected to advance research on land surface biophysical parameter reflectometry of BeiDou Navigation Satellite System.