25 January 2002


         RINEX: The Receiver Independent Exchange Format Version 2.10
         ************************************************************

                            Werner Gurtner
                        Astronomical Institute
                          University of Berne


Contents:
********

0.  Revision History
0.1 Revision Summary
0.2 First Revision
0.3 Later Revisions
0.4 Version 2.10 Modifications
0.5 Version 2.10 Revisions

1.  THE PHILOSOPHY OF RINEX

2.  GENERAL FORMAT DESCRIPTION

3.  DEFINITION OF THE OBSERVABLES

4.  THE EXCHANGE OF RINEX FILES:

5.  RINEX VERSION 2 FEATURES
5.1 Satellite Numbers:
5.2 Order of the Header Records:
5.3 Missing Items, Duration of the Validity of Values
5.4 Event Flag Records
5.5 Receiver Clock Offset

6.  ADDITIONAL HINTS AND TIPS
6.1 Version 1 / Version 2
6.2 Leading Blanks in CHARACTER fields
6.3 Variable-length Records
6.4 Blank Fields
6.5 2-Digit Years
6.6 Fit Interval
6.7 Satellite Health
6.8 Transmission Time of Message (Navigation message file)

7.  RINEX UNDER ANTISPOOFING (AS)

8.  GLONASS Extensions
8.1 RINEX Observation File
8.1.1 Time System Identifier
8.1.2 Pseudorange Definition
8.1.3 More Than 12 Satellites per Epoch
8.2 RINEX Navigation Files for GLONASS

9.  RINEX Extensions for Geostationary Satellites (GPS Signal Payloads)
9.1 RINEX Observation Files for GEO Satellites
9.2 RINEX Navigation Message Files for GEO Satellites

10. REFERENCES

11. RINEX VERSION 2.10 FORMAT DEFINITIONS AND EXAMPLES

    A 1   GPS OBSERVATION DATA FILE - HEADER SECTION DESCRIPTION
    A 2   GPS OBSERVATION DATA FILE - DATA RECORD DESCRIPTION
    A 3   GPS NAVIGATION MESSAGE FILE - HEADER SECTION DESCRIPTION
    A 4   GPS NAVIGATION MESSAGE FILE - DATA RECORD DESCRIPTION
    A 5   METEOROLOCICAL DATA FILE - HEADER SECTION DESCRIPTION
    A 5   METEOROLOGICAL DATA FILE - DATA RECORD DESCRIPTION
    A 7   GPS OBSERVATION DATA FILE - EXAMPLE
    A 8   GPS NAVIGATION MESSAGE FILE - EXAMPLE
    A 9   METEOROLOGICAL DATA FILE - EXAMPLE
    A10   GLONASS NAVIGATION MESSAGE FILE - HEADER SECTION DESCRIPTION
    A11   GLONASS NAVIGATION MESSAGE FILE - DATA RECORD DESCRIPTION
    A12   GLONASS NAVIGATION MESSAGE FILE - EXAMPLE
    A13   GLONASS OBSERVATION FILE - EXAMPLE
    A14   MIXED GPS/GLONASS OBSERVATION FILE - EXAMPLE
    A15   GEOSTATIONARY NAVIGATION MESSAGE FILE - HEADER SECTION DESCRIPTION
    A16   GEOSTATIONARY NAVIGATION MESSAGE FILE - DATA RECORD DESCRIPTION
    A17   MIXED GPS/GEO OBSERVATION FILE - EXAMPLE
    A18   GEO NAVIGATION MESSAGE FILE - EXAMPLE


0. REVISION HISTORY

0.1 Revision Summary

First Revision, April 1993
Clarification December 1993
Doppler Definition: January 1994
PR Clarification: October 1994
Wlfact Clarification: February 1995
Event Time Frame Clarification: May 1996
Minor errors in the examples A7/A8: May 1996
Naming convention for compressed met files; January 1997
Continuation line clarifications: April 1997
GLONASS Extensions: April 1997
Met sensor description and position records: April 1997
Wavelength factor clarifications: April 1997
Error in example A12: CORR TO SYSTEM TIME, April 1997
Redefinition of sv clock params in GLONASS Nav Mess Files: March 1998
Naming conventions for compressed RINEX obs files: March 1998
GPS week: No roll-over, continuous number: March 1998
Error in compressed DOS file naming convention: July 1998
Table A13 contained blank satellite identifiers: Sept 1998
Discrepancy between Tables A5 and A9 removed: Sept 1998
Phase data format overflow: Clarification: Oct 1998
Message frame time Table A11: Clarification: Oct 1998
RINEX Version 2.10 Modifications: July 1999
Typo in paragraph 0.4 (epoch flag >1): Nov 1999
Clarification regarding trailing blanks: Dec 1999
Clarification regarding units of ZD,ZT, URA(GEO)
Clarification regarding time system identifier of GEO obs files
Clarification regarding time system identifier in TIME OF LAST record: Feb 2000
Addition of GEO examples: February 2000
Clarification of epoch field for event flag records: May 2000
Table A6: Typos in format definition of epoch: May 2000
Clarification of the GLONASS satellite identifier: June 2001
Clarification of the floating point exponent format: January 2002

0.2 First Revision

The first documentation of the RINEX Version 2 Format was published by
W. Gurtner and G. Mader in the CSTG GPS Bulletin of September/October 1990.
The main reason for a revision is the new treatment of antispoofing data by
the RINEX format (see chapter 7). Chapter 4 gives a recommendation for data
compression procedures, especially useful when large amounts of data are
exchanged through computer networks. In Table A3 in the original paper the
definiton of the "PGM / RUN BY / DATE" navigation header record was
missing, although the example showed it. The redefinition of AODE/AODC to
IODE/IODC also asked for an update of the format description. For consistency
reasons we also defined a Version 2 format for the Meteorological Data files
(inclusion of a END OF HEADER record and an optional MARKER NUMBER record).

The slight modification (or rather the definition of a bit in the Loss of Lock
Indicator unused so far) to flag AS data is so small a change that we decided
to NOT increase the version number!


0.3 Later Revisions

* URA Clarification (10-Dec-93):

  The user range accuracy in the Navigation Message File did not contain
  a definition of the units: There existed two ways of interpretation:
  Either the 4 bit value from the original message or the converted value
  in meters according to GPS ICD-200. In order to simplify the interpretation
  for the user of the RINEX files I propose the bits to be converted into
  meters prior to RINEX file creation.

* GLONASS Extensions:

  In March 1997 a proposal for extensions to the current RINEX definitions
  based on experiences collected with GLONASS only and mixed GPS/GLONASS data
  files was circulated among several instrument manufacturers and software
  developers.

  The results of the call for comments have been worked into this document.
  A separate document (glonass.txt) summarizes just the necessary extensions.

* A blank satellite identifier is allowed in pure GPS files only

* Met sensor description and position records were added to facilitate the
  precise use of met values.

* Description and examples for wavelength factors and their temporary changes
  (bit 1 of LLI) clarified.

* The RINEX documentation distributed in spring 1997 contained definitions for
  the GLONASS satellite clock offset and drift with the intention to have them
  defined identically to the GPS values. Unfortunately the GLONASS Interface
  Document consulted had a sign error in one of the formulae.

  The values should be stored into the RINEX file as -TauN, +GammaN, -TauC.
  The original definition asked for -TauN, -GammaN, +TauC. See paragraph 8.2.

  To avoid problems with files created with the original definitions a real
  valued version number (2.01) has been introduced for GLONASS nav mess files.

* IGS decided to use the Hatanaka compression scheme for RINEX observation
  files. Below the corresponding RINEX file name conventions are included
  as recommendations. The DOS naming (extension .yyE) was wrongly set to
  .yyY in the March 1998 version of the document.

* GPS week: The GPS week number in all RINEX files is a continuous number
  not affected by the 1024 roll-over, it runs from 1023 over 1024 to 1025 etc.

* A descrepancy between the definition of the header line fields of met sensor
  description and position in Table A5 and the example in Table A9 was removed.
  The latter was correct.

* Clarification for phase data format overflows: Add or subtract a suitable
  number of cycles, set LLI flag.

* Clarification for the GLONASS satellite identifier: "Almanac number" was
  somewhat ambiguous. It has been replaced by "slot number" within the
  satellite constellation.


0.4 Version 2.10 Modifications

The modifications leading to Version 2.10 include:

- Fractional version number
- Zero padding of 2-digit year values (years 2000-2009 --> 00-09)
- Field length of time of first obs (1/10 microsecond resolution)
- Non-integer sampling rate (INTERVAL header record)
- Header records now allowed after all epoch flags >1                          |
- Additional obs types in obs files: S1, S2 (raw signal strength values)
- Receiver clock offset header line to clarify applied corrections
- Default wavelength factor header line mandatory
- Inmarsat GPS payloads: New satellite system definition, new nav mess files
- Curve fit interval in GPS nav mess file
- Redefinition of SV health value in GPS nav mess file
- Additional obs types in met files (ZD, ZT)


0.5 Version 2.10 Revisions

* "Header records now allowed after all epoch flags >2" in paragraph 0.4
  should read ">1"

* The original intention of the RINEX format was to allow for variable record
  lengths of the ASCII files to minimize the file size. Empty fields or
  unknown values can either be represented by zeroes or blank space. Most
  RINEX converters removed trailing blank to further reduce the file size. The
  documentation was not clear enough to explicitely allow for this practice
  (paragraphs 2, 5.3, 9.1).

* The time system identifier of GPS observations generated by GEO payloads
  defaults to GPS (explicitly stated now in paragraph 9.1)

* The time system identifier in the TIME OF LAST OBS header record has to
  be identical to the one in the TIME OF FIRST OBS record

* Clarification of Table A2 to be compatible with examples of Table A7:
  For event flags without significant epoch the epoch fields can be left blank.
  Table A6: Format for epoch contained obvious errors

* Clarification of the floating point exponent format in navigation message
  files (two digits, E,e,D,d letters)