211 / 2023-09-21 16:14:43

地基GNSS-IR高度反演中的误差分析及其补偿模型

GNSS,反射遥感,误差分析,模型改正

地基GNSS多路径干涉反射测量（GNSS-IR）技术凭借其数据来源广泛、低成本、移动性好、高时空分辨率等优势，在多场景（海洋、陆地以及冰雪等）特征参数监测方面展现出巨大的潜力。精准的环境特征参数监测对于各种灾害监控和气候气象研究十分重要，当前的GNSS-IR反演研究中主要考虑到高度变化误差及观测信号质量误差等，这些误差可直接影响到GNSS-IR技术的反演精度和应用范围。为了进一步挖掘上述误差的特性、加深对该误差的认识，通过利用不同误差信号关联的有效高度值（Reflector Height, RH），验证了不同误差类型下反演误差的存在，并定量分析了不同卫星系统的误差大小。针对这些误差的各自特性，提出相关的误差补偿模型，从而易于实现在不同场景下高精度GNSS-IR反演模型的构建。此外，由于当前GNSS卫星信号已进入多模多频的新时代，改正后的融合反演值相较于未改正的反演值的精度获得了明显提升。这种精度的提升不仅得益于多模多频信号提供的大量冗余数据，也是正确处理多种误差信号(包括系统误差、粗差和随机误差)的结果。



The Ground-based GNSS-Interferometric Reflectometry (GNSS-IR) technology, leveraging its wide-ranging data sources, cost-effectiveness, excellent mobility, and high spatiotemporal resolution, presents significant potential in monitoring characteristic parameters across various settings such as ocean, land, and snow environments. Accurately monitoring the environmental parameters is crucial for disaster management and climate meteorological research. In the current GNSS-IR inversion research, the primary focus is on addressing errors related to changes in elevation and the quality of observed signal data. These errors can have direct impacts on the precision and applicability of GNSS-IR technology. To delve deeper into the characteristics of these errors and enhance our understanding of them, we utilized effective reflector height values (RH) associated with different error signals. This approach allowed us to validate the presence of inversion errors under various error types and quantitatively analyze the magnitude of errors across different satellite systems. In response to the unique attributes of these errors, we have proposed relevant error compensation models, making it easier to construct high-precision GNSS-IR inversion models in diverse scenarios. Furthermore, as GNSS satellite signals have entered a new era of multi-mode and multi-frequency signals, the accuracy of the corrected fusion inversion values has significantly improved when compared to uncorrected inversion values. This enhanced accuracy is not only attributable to the wealth of redundant data provided by multi-system and multi-frequency signals but also the result of effectively addressing various error signals, including system errors, outliers, and random errors.