Re: [Discuss-gnuradio] Received Signal Power ?

[bjoernm]

Hi all,

Thanks for the answer! But isn't there a functionality in the XCVR2450  
daughterboard, which already measures the RSSI value?

Isn't there like a function which could output this value?

Secondly. Capturing the samples right at the output of the USRP,  
meaning within "gr.fft_filter_ccc" or maybe within the  
"gr.probe_avg_mag_sqrd_c"?

It's strange, the mailing list archive is full of RSSI related posts,  
but none of which are able to deliver answers to this issue, or am I  
missing something here?

best regards, and thanks again for helping

Quoting "Kresimir Dabcevic" <address@hidden>:

> > -----Original Message-----
> > From: address@hidden on
> > behalf of address@hidden
> > Sent: Fri 9/2/2011 10:26 AM
> > To: address@hidden
> > Subject: [Discuss-gnuradio]  Received Signal Power ?
> > Hi,
> >
> > I'm using the UCLA ZigBee PHY by Thomas Schmid, with the benchmark
> > example from gnuradio.
> >
> > Now I need the signal power of each received packet (at the receiver).
> > For the signal
> > power computation:
> >
> > P_S = sum ( all samples squared ) divided by number of samples
> >
> > but the problem is where to fetch these samples, and how to make sure,
> > that only samples
> > of the received packet are considered, and no noise...
>
> Distinguishing "useful" samples from noise was a problem when I was  
> playing around with UCLA as well. It wasn't an issue when RSS was  
> good, because with power_squelch block, you could filter out the  
> noise. Thus, try setting different TX amplitude / RX gain values in  
> order to get more distinguishable samples.
>
> > 1. is there a gnuradio block/function which would already perform this
> > task of received
> > signal power computation, or RSSI determination, which I could
> > use/edit for my task?
>
> There isn't (or at least wasn't until few months ago, I haven't  
> looked into gnu radio since), so you'll have to extract that  
> yourself. Try searching forum archives for "RSSI measurement",  
> someone (Josh Blum?) explained it to me quite nicely there...
>
> > 2. If not, where would be the best spot to capture the received
> > samples? As the
> > clock_recovery looks to be too late, and capturing the packets earlier
> > makes it hard to
> > distinguish between noise and received signal, as no clock recovery
> > nor packet detection
> > is performed, yet.
>
> It seems that you'll have to capture the samples right at the exit  
> from USRP, no way around that...
>
> Regards,
> Kresimir
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio