<html>
<body>
<font color="#0000FF">At 12:17 PM 8/23/2005, Keith Hogie wrote:<br>
<blockquote type=cite class=cite cite="">3 - The protocol stack diagram
is nice but I'm not sure about the AOS boxes. Normally AOS is used
for data coming down from space.</font></blockquote><br>
Not so - the original architecture that was developed for ISS was
completely symmetric - the AOS frame was intended to be used as either a
unidirectional or a bi-directional space link protocol. See
<a href="http://public.ccsds.org/publications/archive/701x0b3.pdf" eudora="autourl">
http://public.ccsds.org/publications/archive/701x0b3.pdf</a> page
2-3.<br><br>
<blockquote type=cite class=cite cite=""><font color="#0000FF">It also
has Reed-Solomon coding on it. The diagram uses AOS in both
directions. Are we proposing the use of AOS and Reed-Solomon coding
both ways. This would require R/S encoders at ground stations and
decoders installed on spacecraft?</font></blockquote><br>
The CCSDS space link protocols are cleanly layered:
<a href="http://public.ccsds.org/publications/archive/130x0g1.pdf" eudora="autourl">
http://public.ccsds.org/publications/archive/130x0g1.pdf<br><br>
</a>In particular, the current AOS Space Link Protocol as defined
in:<br>
<a href="http://public.ccsds.org/publications/archive/732x0b1.pdf" eudora="autourl">
http://public.ccsds.org/publications/archive/732x0b1.pdf</a> is decoupled
from its underlying coding layers.<br><br>
The coding layer that underlies AOS or conventional TM space links is
defined in:
<a href="http://public.ccsds.org/publications/archive/131x0b1.pdf" eudora="autourl">
http://public.ccsds.org/publications/archive/131x0b1.pdf</a> and note
that it is NOT confined to R-S coding. It may be also expected to evolve
as new codes (such as LDPC) mature.<br><br>
<blockquote type=cite class=cite cite=""><font color="#0000FF"> Do
we want to get into the details of lower layer protocols to use for
two-way links </font></blockquote><br>
No; other CCSDS WGs (by charter) take care of these. The Cislunar WG may
want to form SIGs to correspond with them, though, if the Chairman thinks
that this is appropriate.<br><br>
<blockquote type=cite class=cite cite=""><font color="#0000FF">If this
group is not dealing with lower layers, should the diagrams just leave
them blank and start with the network layer?</font></blockquote><br>
Chris's Section 8.1.1 is clearly identified as an "example". As
such, it is fully appropriate that it could show a representative set of
underlying standard protocols. Since there's nothing technically
incorrect about the diagram, as an example it is probably
useful.<br><br>
Best regards<br>
<x-sigsep><p></x-sigsep>
Adrian J. Hooke<br>
Chairman, CCSDS Engineering Steering Group (CESG)<br>
</body>
</html>