[Top][All Lists]

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

Re: [Discuss-gnuradio] Sub-Band Tuner / Carrier Detect / AGC

From: Eric Blossom
Subject: Re: [Discuss-gnuradio] Sub-Band Tuner / Carrier Detect / AGC
Date: Tue, 18 Jan 2005 00:58:02 -0800
User-agent: Mutt/1.5.6i

On Mon, Jan 17, 2005 at 10:00:37PM -0700, address@hidden wrote:
> I've set the USRP IF to the center channel in the desired band.  Using the 
> freq_xlating_fir_filter_ccf method, I should be able to go up and down in 
> 25KHz step to listen to each channel.  This works great, except I hear the 
> image channel, as well as the desired channel at the same time.  So I hear 
> +/- 50KHz in the above example.  Should the freq_xlating_fir_filter_ccf 
> method just select the upper or lower side band signal based on the sign 
> of center_freq.

This should work.  It could be that the filter you designed wasn't
good enough.  You might want to compute the freq response and check it
it, or use the remez code (which needs a simple interface).

> -----------------------------------
> Is there a way to access the magnitude of the channel to setup a 
> squelch function?  How would that work with this chain of blocks?  Is 
> there another block that would go in the chain to zero the output if the 
> I*I + Q*Q of the signal is below a threshold?

You can get the magnitude with gr.complex_to_mag ().
Then we probably need a comparator block that take the mag in and a
controllable knob (the squelch setting) and produces a one or zero
out.  Feed that as one of the inputs to gr.multiply_ff.

> ---------------------------
> Is there a block that impliments AGC for AM signals?


> ------------------------
> With the latest CVS code block, there is now a sequence of aU characters 
> printing out while the flow graph runs.  Do these denote under runs in one 
> or more subsystems?

Here's the secret decoder ring:

    aU   audio underun
    uU   USRP underun
    aO   audio overrun
    uO   USRP overrun

The aU's are generally harmless and occur because there are two
non-locked time bases in use: the USRP clock and the sound card clock.

A better solution than "just ignore the problem" is welcome.
There are other ways that ALSA could be setup.  It can simulate an
interrupt callback.  In the callback we could fill or drop samples as required.

> ----------------------
> I have a number of off air sample signals now captured.  I'll post them to 
> my web site in the next day or two.  These include:
> Multiple narrow band FM signals (In the same IF pass band)
> Airband AM ATIS recording
> Multiple NDB (Aircraft) siganls (In the same IF pass band)
> Upper and lower SSB signals


> Still working on getting NTSC and GPS capture
> Sorry for so many questions in the same message.  Should I break them down
> to one question per e-mail to make it easier to follow threads?  
> Hopefully these are of use to others on the board.

Either is fine.  One Q per msg might be a little easier, as long as
you don't have to recapitulate the setup.


reply via email to

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