Borre, un geodeta che ha fondato e diretto il Centro danese GPS per molti anni, rilasciò la prima collezione di GPS Easy Suite su RINEX versione 2.10. Nel corso degli anni ha ricevuto molte richieste per un codice per il formato RINEX versione 3.03. Nel tempo anche Matlab era cambiato e ora sono inclusi i nuovi comandi rilevanti.
Borre ha deciso di aggiornare i file con un codice basato su Matlab versione 8.5 (R2015a).
L’Università Statale Aerospaziale di Samara ha sostenuto gli sforzi di Borre, che sono state intraprese nel periodo da Agosto 2015 al Febbraio di quest'anno.
The History of the Easy Suite
The first easy files (easy1-easy10) were published in the periodical GPS Solutions in 2003, 7:47-51. At that time RINEX version 2.10 was widely used. This first easy-suite was augmented by a continuation (easy11-easy18) in InsideGNSS in the years 2009-2010. During recent years numerous requests for an update to RINEX version 3.03 arrived. This is natural since 25 years have passed between the early and the present RINEX version.
A comprehensive description of the individual easy-files can be found in Chapters 9 and 10 of the textbook by Kai Borre & Gilbert Strang (2012) Algorithms for Global Positioning, Wellesley-Cambridge Press.
The Matlab programming language has undergone some developments, too. The new code is based on Matlab version 8.5 (R2015a).
The update involves a renumbering of a few scripts in order to avoid the file identification being a combination of a number and a letter. The update was a major undertaking as RINEX version 3.03 uses a more flexible format than was allowed for version 2.10.
I tried to find an optimum way of coding. Optimum in the sense that it still is useful as a teaching code which is easy to modify, and also not exploiting too sophisticated tricks.
Many users unfamiliar to Matlab complained that this or that file was missing. One solution to this situation is to avoid using functions too often or include them at the bottom of the script files. Again this results in more code lines, but hopefully the difficulty with seemingly missing functions is eliminated.
Somebody would likely appreciate that I used cells and structures more extensively in the code. My argument is that I still want to keep the newcomer in focus and I judge the mentioned efficient augmentation only would complicate the learning process.
Again the suite is described in the textbook Kai Borre & Gilbert Strang (2012) Algorithms for Global Positioning. Wellesley-Cambridge Press.
This new package of Easy-files must not be mixed with any earlier code. Some of the scripts and functions carry the same name, but most often they have been modified.
When introducing new code, it is inevitable that it contains bugs. The earlier easy-suite had reached a stable stage. I no longer receive hints on repairing code. So either the code is perfect, or no one uses it.
New Data Set
The present version is based on a new data set which is collected with two Javad Delta-3 receivers. About 15 minutes of data from two antennas 0:6 meter apart and using 1 second epoch intervals. The site is on top of the main building of Samara State Aerospace University. The approximate coordinates are
latitude φ = 53o12’43.2’’ , longitude λ = 50o10’39.5’’ , and ellipsoidal height h = 150 meter.
The data include only GPS observations. For a newcomer it would be too complicated to include for example GLONASS. However, we work on publishing a book on Matlab receivers for GPS, GLONASS, BeiDou, and Galileo, mainly single-frequency receivers. Other useful material in the book is a snap-shot receiver; all receivers are coded to start from output as coming from a common front-end that can be purchased.
Professor, Senior researcher
Copia qui lo "short link" a questo articolo