[Discuss-gnuradio] Re: A newbie question

[Vanu Bose]

> I have been fascinated with the ideas soft-radio for quite a few
> years but have been bothered by one detail - saturation.  The 
> dynamic range of any reasonable ADC looks much to small to accommodate
> the range in signal strengths actually seen at the antenna.  In conventional
> analog receivers saturation is prevented in downstream stages by analog
> filtering in upstream stages to remove a powerful station that would otherwise
> block a weak station.  In a soft-radio, however, all stations are present
> in the signal at the time of a->d conversion. The converter must have
> sufficient resolution to encode not only the powerful signals but the weakest
> one of interest with enough bits to provide reasonable fidelity.
> 
> Given the dynamic range of actual signals that I have observed with a 
> spectrum analyzer I don't understand how they can be coded with adequate
> resolution using coders that are affordable to the average hobbyiest or
> that would be practical in consumer electronics.  Are log-coders 
> used - something similar to mu-law coders that are used for audio?
> How wide are the coders?  How much do they cost?
> 
> Would someone say a few words on the point, please?

Bill,

You've identified a key issue for software radio. 5 years ago, the
specs. on the spurious free dynamic range of the A/D converters
were not good enough to handle wideband software radio implementation
for most cellular standards. However, in the past several years there
has been a significant improvement in the underlying technology.

Cellular system employ some sort of power control on the link, which
sets a limit on the difference between the smallest and largest
signals that ill be seen at the basestation receiver. For GSM, you
need about 91 dB of spurious free dynamic range in order to receive
signals within spec. The latest wideband converter from Analog Devices
had 105 dB spurious free dynamic range (I think this is at 65 MSPS,
but you should double check). This give you the capability for
digitizing an entire 15 MHz PCS band and process all of the signals
in software. If you are thinking about digitizing bands larger than 30
MHz at a time, then the converters probably aren't good enough yet.

Ironically, one of the hardest things for a software radio is AM. This
is because the station broadcast powers can be so high, that you would
need a very large dynamic range to digitize the entire band and be
able to pull out the weak signals.

As to your question of affordability, I think analog devices sells
eval. boards for their converters that cost roughly $100 - $200. You
can take a wideband IF (e.g. 70 MHz) directly into these boards.

-Vanu

------------------------------------------------------------------------
Dr. Vanu G. Bose
President & CEO
Vanu, Inc.
One Porter Square
Suite 18
Cambridge, Ma 02142
ph: 617.864.1711x203
fx:617.864.1697
www.vanu.com

Assistant:
Elizabeth Griffin
address@hidden
ph: 617.864.1711x239
------------------------------------------------------------------------