cFit Compressed FitACF data format

RFC: 0007

Title

cFit Compressed FitACF data format

Author

R.J.Barnes

Summary

A description of the compressed FitACF data format "cFit".

Description

1. Introduction

The cFit format is a compressed format for storing a limited sub-set of data from the output of the FitACF algorithm. The format does not store all of the radar operating parameters or the full set of derived values of the algorithm, instead it stores sufficient paramaters to characterize the operation of the radar and the power, spectral width, and velocity, and the derived errors, from the analysis.

The advantage of cFit files is that they are considerably smaller and more convenient to work with than full fit or fitacf files. Typically a cFit file is a few megabytes in length, whereas a fit file can be over a 100 megabytes.

2. The Format

A cFit file is divided up into blocks or records, each record has the same format but can have varying length depending on the scatter observed by the radar.Each record is divided up into a header and a data section. The structure of a record is given below:

Byte Offset	Size (Bytes)	Type	Content
0	8	64-bit double float	Time in seconds from UNIX epoch (00:00:00 UTC, January 1, 1970).
8	2	16-bit short integer	Station identifier
10	2	16-bit short integer	Scan flag
12	2	16-bit short integer	Program identifier
14	2	16-bit short integer	Beam number
16	4	32-bit float	Beam azimuth
20	2	16-bit short integer	Channel
22	2	16-bit short integer	Integral component of integration time in seconds
24	4	32-bit integer	Fractional component of integration time in micro-seconds
28	2	16-bit short integer	Distance to first range in kilometers
30	2	16-bit short integer	Range separation in kilometers
32	2	16-bit short integer	Receiver rise time in micro-seconds
34	2	16-bit short integer	Operating frequency in kilohertz
36	4	32-bit integer	Noise Level
40	2	16-bit short integer	Attenuation
42	2	16-bit short integer	Number of averages
44	2	16-bit short integer	Number range gates
46	1	byte	Number of stored ranges
47	m	m bytes	Range table
47+m	n	n bytes	Data table

The remainder of the block consists of the data values for the stored ranges. The range table indicates which range a data value corresponds to. For example, if the first entry in the range table contains a value of 23, that indicates that the first set of data values in the data table came from range 23.

Each entry in the data table has the same format shown below:

Byte Offset	Size (Bytes)	Type	Content
0	1	byte	Ground scatter flag.
1	4	32-bit float	Lag-zero power
5	4	32-bit float	Lag-zero power error
9	4	32-bit float	Velocity
13	4	32-bit float	Lambda power
17	4	32-bit float	Lambda spectral width
21	4	32-bit float	Velocity error
25	4	32-bit float	Lambda power error
29	4	32-bit float	Lambda spectral width error

References

None

History

2007/03/26 Update to reflect changes to file format
2004/07/27 Initial Revision.