[Top][All Lists]

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

Re: [Discuss-gnuradio] gnuradio on ubuntu touch

From: Marcus Müller
Subject: Re: [Discuss-gnuradio] gnuradio on ubuntu touch
Date: Wed, 06 Aug 2014 17:11:28 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0

doing your computations on a reduced sampling rate always makes a lot of sense.

Also, if you do complex to mag² prior to filtering, you get the power in the overall nyquist band ignoring power fluctuations that do not pass the low pass, and not the power in the pass band of the filter.

Example: Assume we have a complex sine at frequency $f>f_\text{cutoff}$  $s(t) =
      e^{2\pi j f t}$. It has a constant mag² of $|s|^2 = 1$.
Now, the low pass of a constant 1 is still 1.
Now, let's low pass first with a filter $h(t)$ and then mag². Assume filter stop band attenuation is infinite, then $r(t) = h(t) *
      s(t) = 0$ because $f>f_\text{cutoff}$. It follows that $|r|^2 = 0 \neq

Yet another side note: have a look at the polyphase filter banks (PFBs) that are relatively new in GNU Radio, if you want to process multiple sub-bands at once.


On 06.08.2014 16:47, address@hidden wrote:

Try decimating after the filter, before you start computing log10 at the
input rate. There's no need to do those calculations at the original 

 sample rate once you've computing complex-to-mag**2 and filtered. 

On 2014-08-06 10:30, rejunte wrote: 

Tried that, but still ocurring overflow.

Another thing I had was that on my computer I was using a 

complex to mag^2 -> single pole IIR filter -> log10 -> multiply const (10)
to measure the signal strength

that is also causing overflow on the single board computer. If I put a RMS
block instead, no overflow occurs.

View this message in context: http://gnuradio.4.n7.nabble.com/gnuradio-on-ubuntu-touch-tp49315p49807.html [1]
Sent from the GnuRadio mailing list archive at Nabble.com.

Discuss-gnuradio mailing list
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio [2]

[2] https://lists.gnu.org/mailman/listinfo/discuss-gnuradio

Discuss-gnuradio mailing list

reply via email to

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