discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] AM receiver?


From: Eric Blossom
Subject: Re: [Discuss-gnuradio] AM receiver?
Date: Mon, 2 Feb 2004 17:33:18 -0800
User-agent: Mutt/1.4.1i

On Mon, Feb 02, 2004 at 11:27:50AM -0700, Jung Ko wrote:
> Hi there,
> 
> I'm new to GNURadio mailing list, so apologies if this question has been 
> asked/discussed before.

Welcome!

> I'm interested in receiving AM radio signals. Here I have some 
> questions that hopefully someone will be able to answer me:
> 
> 1. Since the frequency is not that high, I suppose it is enough to hook up
> the antenna directly to the ADC?

You'll need at least some minimal bandpass filtering to restrict what
the A/D sees.  If you've ever grabbed a scope probe with your fingers
you've probably noticed a huge 60 Hz signal!

> 2. If 1) is true, then is it still necessary to write software frequency
> mixer to shift down the frequency to IF (around 455Khz)?

Assuming you're using a fast enough A/D (approx 4 MS/s for broadcast
AM [540 to 1700 kHz]), no mixer will be required.  You'll need the
bandpass filtering and probably some kind of amplifier.  If you're
looking for a minimalist solution you can probably get by with a
simple R/C or R/L/C filter and a single transistor amplifier.  A FET
input op-amp would probably work too.  If you use a long enough piece
of wire for your antenna and you're close to a station, and your A/D
has a high input impedance, you might not need the amp.

With this configuration your ability to listen to weak stations in the
presence of strong stations will be lousy.   Strong station reception
should be OK.

> 3. Which parts of the AM demodulation are already implemented in GNURadio? 
> Does it use envelope detection or synchronous detection? 

I don't know that anyone has pieced it all together, but most of it
should be there.  Use GrFreqXlatingFIRFilterFCF to extract the station
of interest and translate it to baseband.  Envelope detection would be
absolute value (GrMagnitude<VrComplex,float>) followed by a low pass
filter (GrFIRfilterFFF).

Sounds like a great example.  Let us know what it takes to make it
work.  Nothing like building a crystal radio with a couple of thousand
dollars worth of computer hardware ;-)

Eric




reply via email to

[Prev in Thread] Current Thread [Next in Thread]