Re: [Discuss-gnuradio] Welcome and brief update

[John Turner]

Hi Folks,

I think I'm what is described as a "lurker". I also design RF hardware,
downconverters and sampled data systems. As background, my interest in SDR
is form the perspective of using digital techniques to improve radio
performance which I think is along the same lines as a comment from Chris
Albertons's comment about Radio Astronomy. I think he was also asking what
performance levels are really appropriate.

I just wanted to put in my two cents on what is possible or reasonable. What
I'm trying to lead up to is a low cost implementation that allows the
greatest flexibility.

Generically the hardware of an SDR platform breaks down to an RF Tuner and a
sampling card. On the RF side everyone would like to have as wide a range RF
input as possible and the best range for the dollar are TV tuners, like the
Microtune evaluation modules.

TV tuners are implemented with a dual conversion topology to achieve a wide
frequency range (50 to 860 MHz) and good RF performance (60dB selectivity
and image rejection). The tuners are linear converters hat have a noise
figure of about 10 dB and a bandwidth of between 6 and 18 MHz. TV tuners
have been used for years as the front ends for hobby level spectrum
analyzers. The issue is maintaining performance to a low frequency output.

The sampling is where things get interesting but a little awkward. There are
a number of great ADC ICs out there that are well suited to SDR
applications. The issue is how do you get this sampled data into a useable
format effectively and cost efficiently. The 20MSPS ADC cards mentioned
translate a reasonably wide bandwidth, ~7MHz,  to the PCI bus but they are
reasonable expensive. At the other end of the spectrum are audio cards which
are inexpensive but function to high end audio frequencies (192ksps is the
highest I've heard of)

I see considerable value in using the USB port, at 12Mbps versus the 480
Mbps of USB 2.0,. I think would present a more practical interface then an
audio port, i.e. low hardware cost but with an effective sampling rate of
~500ksps. Furthermore USB has a transfer rate that must be tolerant of
transmission errors, i.e. USB cameras, which would match well.

I believe the following topology represents a good value performance
tradeoff, between $500 to $1000 unit cost.
TV Tuner offering 50 to 860 MHz range and 60dB dynamic range performance A
channelized output of 40 MHz.
ADC converter operating at 70 Msps, the idea is that the IF is undersampled.
Digital Downconverter as a sample rate converter
Sample rate to USB converter
I believe the Downconverter and the USB interface could be built into an
FPGA.

The issues I have are then how to process the USB data but I think this
should be manageable.

This allows a fully remote device, i.e. no plug in acquisition cards, the
FPGA could take care of programming the synthesizers in the tuner. Alternate
FPGA code or digital downconverter coefficients could allow some initial
channel filtered etc..

I don't know if this could be built as a "kit" i.e. with Mini-circuits
modules, but if there is enough interest I believe we could supply it at a
reasonable cost.

JT

address@hidden
www.valhalla-systems.com





----- Original Message -----
From: Eric Blossom <address@hidden>
To: <address@hidden>
Sent: Friday, January 10, 2003 10:06 AM
Subject: [Discuss-gnuradio] Welcome and brief update


> I'd like to extend a welcome to all the new subscribers.
>
> Although the list has been relatively quiet, there's been a lot going
> on behind the scenes.  Matt and I and the others who've been working
> on GNU Radio have agreed to move the bulk of our discussions to the
> list, so everybody is more included.
>
> We're putting together a Wiki (user editable web site) for discussion,
> FAQs, ideas, brain storming and project coordination that should make
> it easier for folks to contribute and see what others are up to.  Look
> for that in a week or so.
>
> As far as I can tell we've got folks with a wide variety of
> backgrounds on the list.  We've got radio folks without much
> experience in software, we've got software folks who are interested,
> but don't know much about radio, we've got industry lurkers, students
> as well as generalists.
>
> Here are a few areas where we can use help:
>
> (1) RF Hardware design: to make GNU radio more accessable for those
> without high speed A/D hardware, we'd like one of you RF guys to
> design a simple RF front end that could be directly hooked to one of
> the "HiFi" 96 KHz sampling rate sound cards.  Please make it simple
> and buildable out of things that anybody can order from mini-circuits,
> digikey, etc.  The idea is not to build the greatest piece of RF
> hardware ever, but to build something minimalist that actually works.
> Perhaps a basic 2M or 70cm downconverter.
>
> (2) CORBA expert: We're currently investigating what's the right
> remote control paradigm for GNU Radio.  I'd like to talk to somebody
> who's got deep knowledge and experience with CORBA (or some better
> idea) who can help me evaluate our options for glueing CORBA into a
> real time runnning system.
>
> (3) ATSC Equalizer expert: If you happen to be one, or are good on
> comms theory and practice, we'd love to talk to you or have your
> direct involvement.  We can provide gigabytes of off the air samples,
> so you won't need any RF hardware.
>
> (4) Examples and "HOWTOs": We could use some simple applications,
> perhaps some that are complete simulations (no hardware required), and
> instructions on how to use them.  "Getting Started with GNU Radio"
> (Yes, I understand that we're a bit of moving target...)
>
> If you've got other ideas, please jump in.
>
> Again, welcome to all.  With the ground work that's been laid, we have
> a great opportunity to expand the GNU Radio community of users and
> developers this coming year.
>
> Eric
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/discuss-gnuradio