Bibliografie VÚGTK
Optimizing high-resolution troposphere estimates using PPP method and Benchmark data set
Václavovič, Pavel ; Douša, Jan ; COST ES1206/GNSS4SWEC 3rd Workshop



Publication type: prezentace
Extent13 slidu

Link: http://knihovna.vugtk.cz/record/193068/files/2016-dousa-optimizing--GNSS4SWEC-proc_opt.pdf
Fulltext: https://www.researchgate.net/publication/297739140_Optimizing_high-resolution_troposphere_estimates_using_PPP_method_and_Benchmark_data_set

Annotation:
Compared to the network processing mode, the PPP is the preferred method for new products because of its high efficiency when running in a fully autonomous mode, optimal exploiting available multi-GNSS observations, use of undifferenced observations related to tropospheric delays in the absolute sense, and easy support of ultra-fast (real-time) tropospheric production. Moreover, the Kalman forward filtering together with the backward smoothing enable parameters to be estimated in a high resolution compared to the batch least squares adjustment. The backward smoothing is usually employed in the post-processing mode providing results with an accuracy comparable to the traditional least squares adjustment. We utilized the smoothing in near-real-time estimates and studied the dependence of the processing delay on estimated ZTD. In case of real-time, ZTD reached the standard deviation of 10 mm, which decreased to 6 mm when the processing was 4 hours delayed. Most important advantage of this approach is, that all high-resolution parameters were estimated with the same accuracy. Another optimizations of the troposphere estimates using the Benchmark data set, such as a random walk, multi-GNSS and use of different precise products, has been studied.

The record appears in these collections:
Focus on VÚGTK > Researchers > Pavel Václavovic
Focus on VÚGTK > Researchers > Jan Douša
Documents of VÚGTK > Articles VÚGTK
Annotations

 Record created 2016-09-14, last modified 2017-06-06


Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)