"Eric Travis" <[email protected]> wrote:
> As the press release (and associated product links) so clearly state,
> the mission data *is* indeed disseminated to subscribers via multicast
> UDP/IP/DVB but this is only *after* the Envisat delivers the data to the
> ground non-IP related mechanisms;
>
Our group (unisal/gcs) got involved quite late in the Envisat project and
only into the "distribution" aspect - mostly since we had delivered a
similar delivery system for ESA earlier. Unfortunately we were never
involved in the spacecraft-earth link which our software could handle quite
well.
> The Envisat spacecraft itself does not appear to be using DVB or IP,
> Envisat seems to be only using CCSDS link protocols (probably nothing
> more at layers 3 or above):
>
The SiMPLE/DataCast software is quite independent of the link layer and has
been operating over a number of different technologies including VSAT links
and the MPEG-2/DVB MCPC and SCPC channels. The latter is quite popular now
since it is very cost effective with equipment costs (for the receiving
station) of a few hundred US$; the sending station for MPEG-2/DVB is,
however, more expensive but also very flexible.
All higher layer protocols run on top of IP/UDP and it would not be a major
task to run it e.g. over a CCSDS link. Our main motivation for pursuing this
development was the realization that TCP is a very good protocol but (a) for
point to point applications, and (b) being based on ARQ suffers from the
well known "long fat pipe/network" effects which are inherent to ARQ
protocols. Hence we decided to use FEC on the transport level (which also
supports multicast) in combination with a voluntary NAK strategy. This
protocol (RRMP) can operate over simplex, duplex or separate forward/return
links and can accordinly run in a mu,ber of different modes ranging from
"best effort" (simplex channel, UDP + FEC at the receiver) to fully reliable
(forward plus return link and NAK strategy combined with a timeout window
for retransmission requests) and it supports point to point as well as point
to multipoint transmissions.
We would most certainly be quite intersted in using this system for a direct
spacecraft to earth distribution service and are convinced it would work
very well when compared to TCP-based solutions.
Regards,
--Horst Clausen
(on leave from University of Salzburg
-at New Mexico Tech )
----- Original Message -----
From: To: "Keith Hogie" <[email protected]>
Cc: "William Ivancic" <[email protected]>; <[email protected]>
Sent: Friday, July 05, 2002 11:15 AM
Subject: Re: UDP versus SCPS-TP, TCP etc.
> Keith,
>
> It seems that I must correct myself regarding your message
> of June 26th...
>
> While I stand by my previous comments, they missed the point that
> (it is now clear) you were making:
>
> Envisat's DDS (*NOT* the Envisat platform itself) is an excellent
> example of proper use of UDP/IP by a space mission.
>
> (The point I missed: The Envisat spacecraft itself is not using
> UDP/IP - or even DVB)
>
> To this, I fully agree - and apologize for missing it the first
> few times around.
>
> To further discuss, the following elements of your message:
>
> Keith Hogie wrote:
>
> <snip>
>
> > What we see for missions using IP is a very heavy use of UDP and not
> > TCP.
>
> <snip>
>
> > For an example of other UDP use see the following:
> >
> > http://www.esa.int/export/esaCP/ESALS5OED2D_index_0.html
> > http://www.gcs-salzburg.at/
> > http://www.gcs-salzburg.at/products/p_odg.html
> > http://www.simple.at/
> > http://www.datacast.at/technology.html
>
> <snip>
>
> > The bottom line is we want to use standard IP/UDP/TCP protocols that
> > are supported in every operating system available. Using a modified
> > TCP requires more software support that creates more problems when
> > you are building and deploying operational control centers and
> > level-zero processing systems. We really prefer to use standard
> > protocols at layer 4 and below and do any space specific tweaks
> > in the application layer.
>
> coupled with the context of the discussion that prompted your message -
the
> use of TCP by spacecraft elements - lead me (and perhaps a few others) to
> totally misinterpret the actual significance of the ESA press release (and
> product information) links that you provided.
>
> Sometimes we read what we want to see :o)
>
> The unfortunate result of this is that:
>
> The inclusion of Envisat as an example of a space based sensor
> platform using of IP is not quite an accurate portrayal of how
> the Envisat data is actually being disseminated...
>
> As the press release (and associated product links) so clearly state,
> the mission data *is* indeed disseminated to subscribers via multicast
> UDP/IP/DVB but this is only *after* the Envisat delivers the data to the
> ground non-IP related mechanisms;
>
> The Envisat spacecraft itself does not appear to be using DVB or IP,
> Envisat seems to be only using CCSDS link protocols (probably nothing
> more at layers 3 or above):
>
> http://ccsds.gst.com/implementations/missions.html (mission 98)
> and
> http://esapub.esrin.esa.it/pff/pffv6n4/polv6n4.htm
>
> Envisat's DDS (Data Dissemination System) is what so successfully exploits
> commercial communication satellite(s) and DVB encapsulation to
redistribute
> the data.
>
> http://esapub.esrin.esa.it/bulletin/bullet109/badessi.pdf
>
> This is more than slightly different from Envisat itself using
> multicast IP. Again, I misread the significance of the press release
> you were providing.
>
> What Envisat does illustrate so well to this community is an extremely
> pragmatic way of doing business:
>
> (1) Delivering science products efficiently, reliability to
> the ground
>
> (2) Disseminating that data to the user community via IP-based
> networks
>
> (*) Envisat is *not* the only/first mission to do this, but their
> data volume is truly impressive!
>
> The end-result is that this approach allows for the subscriber community
> to receive their data using "vanilla" (non-space tweaked) IP-based
networking
> technology and the method of communication to/from the spacecraft is of no
> consequence to them.
>
> Kudos to ESA for an excellent (high profile) illustration of how to
exploit
> IP technology for the distribution of space science data.
>
> Space missions have been doing this sort of thing for some time - and
should
> continue to do so.
>
> Eric
>
This archive was generated by hypermail 2b29 : Sat Jul 06 2002 - 16:06:08 EDT