[Namaste-dev] Re: System Block Diagram - Version 1, and other matters
rwmcgwier at gmail.com
Thu Jun 26 09:56:27 PDT 2008
This is as good "big picture" as we can get. Thank you for this.
I want to give you all some data points. What follows is information
unrelated to my ego. Please take it as such.
In my work on the REAL dark side of the force as opposed to the "obscure"
signature below ;-), I have some experience of what can be accomplished in
the real world and am completely compared to help begin implementation of
this along with the corresponding half on the other side.
With a USRP2 and a modern PC, I can prototype this very quickly indeed if
we do not do the FH-CDMA. Matt tells me "my" USRP2 are close to "in the
I will be meeting with Rick on Sunday and I will brow beat him about getting
an svn repository up so we can begin this.
Let me give you a data point. On a single Dell 2950 with two Quad core
Xeon at 3.16 GHz, I can do all I/O and process down to bits communications
channels of all sorts of signal channels in parallel. I do, day in and day
out, processing of 70 channels, each one MHz wide, where I process for
energy alarms, classify the content, decide to send it down to a narrower
band processor and run all of those. There are literally thousands of
threads in this system and some of the demodulation and decoding (forward
error correction and removal of interleaver) is intense. It uses the sdr
core you can blame AB2KT and I for having started.
This is not bragging, please believe me. If I wanted to brag, I would have
told you about it ages ago when it was first done (and told you I can
process more data than Vanu can on his servers ;-) ). I bring it up to try
and move us to something concrete as quickly as possible and to demonstrate
my strongest possible belief we can, so we have stuff to show in October.
We have, as demonstrated in Paul's diagram, an almost completely software
defined system. I have a novel idea. Let's write some software. We have
the necessary tools available to this project in the USRP2 and Gnu Radio to
do rapid prototyping of this system. We should. It will cost only a few
thousand dollars to do the entire ground station software prototype out to
baseband and even build a spacecraft simulator in the same space.
We can use existing plug in cards for the USRP2, again inexpensive, to move
it to RF. These are NOT in our bands but we are not trying to do the
initial experiment in the amateur satellite service. We want to operate
this for our own education and to build towards presenting an experimental
R&D system to first responders in Southern California. I have done P25
implementations for work. I won't be sharing those and I won't be writing
them personally but I will be happy to "lead" others to the successful
implementation when we get ready to provide some kind of interoperability.
So, I am going to lead this off. It being open source and built using open
source, I need no one's permission to start and I hope you will follow. It
will be open to any and all who want to develop code for download and to a
more limited set of developers for upload. I can provide the shell of this
block diagram in Gnu Radio blocks pretty quickly so we can begin adding the
pieces for experimentation.
ARRL SDR Working Group Chair, AMSAT VP Engineering.
Member: ARRL, AMSAT, AMSAT-DL, TAPR, Packrats,
NJQRP, QRP ARCI, QCWA, FRC.
"Trample the slow .... Hurdle the dead"
From: namaste-dev-bounces at amsat.org [mailto:namaste-dev-bounces at amsat.org]
On Behalf Of Michelle
Sent: Tuesday, June 24, 2008 1:30 PM
Subject: [Namaste-dev] System Block Diagram - Version 1
Here is the current block diagram. It's in the feed and on the web. We
expect it to change, of course, but we wanted to post it for review and
Potestatem obscuri lateris nescis.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Namaste-dev