一种适用于单频GNSS数据的多周跳探测与修复方法
A Multi-cycle Slip Detection and Repair Method for Single-Frequency GNSS Receiver
徐天扬 1章浙涛 1何秀凤 1袁海军1
作者信息
- 1. 河海大学地球科学与工程学院,江苏 南京,211100
- 折叠
摘要
随着低成本接收机的广泛使用,单频GNSS(global navigation satellite system)已成为主要应用场景之一.单站单频GNSS通常由于观测环境较为复杂,其周跳率往往较高.虽然传统的无几何(geometry-free,GF)组合可探测部分大周跳,但对小周跳不敏感;几何(geometry-based,GB)组合可有效解决该问题,但在多个周跳同时发生时,GB组合的探测成功率会大幅下降.因此,提出了一种将GF组合与GB组合相结合的周跳探测修复方法.首先利用GF组合将大周跳探测并剔除,然后采用GB周跳探测方法探测小周跳,并在选权迭代的过程中不断修复发生大周跳的观测值.实验结果表明,所提方法可有效提高多周跳发生时的探测修复率,在极端情况下与传统方法相比成功率可提高约50％.
Abstract
Objectives:With the rapid popularization of low-cost receivers,single-frequency global naviga-tion satellite system(GNSS)is now widely used.The cycle slip rate of a single-frequency GNSS receiver is commonly high in a complex situation.Although geometry-free(GF)model can detect some large cycle slips,it is not sensitive to small cycle slips.The geometry-based(GB)model is efficient for cycle slip detec-tion and repair,but its success rate of detection descends correspondingly with an excessive number of ab-normal satellites.Methods:By combining the advantages of GF and GB models,this paper proposes a new cycle slip detection and repair method.The proposed method firstly detects and rejects the abnormal satel-lites with large cycle slips by GF model,then uses GB model to detect small cycle slips and repair large cy-cle slips.Results:The results show that the proposed cycle slip detection and repair method can enhance the success rate if many cycle slips occur simultaneously.The success rate can be improved by approximate-ly 50％even when the abnormal satellites reach the half or more of total numbers simultaneously.Conclu-sions:Compared with the traditional GB model,the proposed GF+GB model can take advantage of each model to improve the performance when multi-cycles occur.
关键词
单频观测值/周跳探测与修复/无几何组合/几何组合Key words
single frequency observation/cycle slip detection and repair/geometry-free model/geome-try-based model引用本文复制引用
基金项目
国家自然科学基金(41830110)
国家自然科学基金(42004014)
江苏省自然科学基金青年基金(BK20200530)
出版年
2024
NETL
NSTL
万方数据