> Except that TLEs & identifications for the non-secret objects allow their elimination from consideration - what's left are the secret satellites which somebody might want to hide from, or whatever. Presumably the secret satellites that are currently in track would remain in track, so the problem would come with new launches and UNIDs in general. That's a question I'd like to see the experienced observers/analysts discuss a bit: Undoubtedly the large files resulting from OIG compilations are very useful in IDing UNIDs, but are they absolutely indispensable? After all, a TLE resulting from visual obserations is, in essence, a vector in a seven-dimensional feature space that isn't all that densily populated -- that gives a lot of discrimination capability right there. Add visual magnitude, light curve, maybe color, azimuth and times of known launches, etc., and I think the exclusion part of the ID problem may be less formidable that it first seems. ------------------------------------------------------------------------- Subscribe/Unsubscribe info, Frequently Asked Questions, SeeSat-L archive: http://www.satobs.org/seesat/seesatindex.html
This archive was generated by hypermail 2b29 : Mon Aug 09 2004 - 11:44:57 EDT