Re: Sorting satellite elements

Jim Varney (
Thu, 1 Aug 1996 22:31:33 -0700

>David Bishop wrote:
>>>>  One thing that I need is a good satellite elements digester....
>>>1)  Read a free formatted file with elements in it, and pars them out
>>>2)  Save only the most current elements to a file with JUST elements in it.
>>>3)  Disregard elements which are over X months old
>>>4)  Sort the elements in the output file by NORAD #
>>>5)  Recognize Ted Molczan's element files, keeps the identifier line
>>>6)  Check QUICKSAT.MAG, if there is no entry for it, and .. add it.
>>>7)  Fix any checksum errors it finds.
>>>If anybody has something better I am all ears.

Bjoern Gimle wrote:

>And I don't want to throw away any elsets - they can be needed for
>for analysis of photographs, observations by others, and decays /
>maneouvres / sudden accelerations. All old elsets (preferrably
>without duplicates) should be written to a history file.
>I will write a utility for this, using CA-Realizer for Windows.

Regarding developing new programs for handling satellite elements,
I am working on a new program as well.

The program I am writing is called "SatManager."  It will have a
distributable database engine that will have a capacity of 2 Gigabytes
of elements (your hard drive permitting).  SatManager has this model:

 +------+        +----------+           +----------------+
 | .tle | parse  |SatManager|    out    | User-selected  |
 | files|------->| Database |---------->| elsetss based  |
 +------+        +----------+           | on selection   |
                      |                 | flags/criteria |
                      |                 +----------------+ 

Many elset-handling program handle elsets as one blob of text and
just sorts them by catalog number.  SatManager extracts each element
of data from the elset and stores them in a separate database field.
By treating each element (such as mean motion, B*star, etc.) as a
numeric field, not a text field, analysis becomes possible.

SatManager is configured now to support "history files" of multiple
elsets per catalog number.

It will have "name control" so the user can impose consistent common
names.  It will also support extraction of elsets from text files, strip
headers, etc., as well as perform automatic Unix to MS-DOS linefeed

You'll find a screen shot of the database interface at

I hope to get this thing done before the year digit changes
from "6" to "7"  :)

Jim Varney
Sacramento, Calif

   What we know is not much; what we do not know is immense.
                                                       - Laplace