Re: [Discuss-gnuradio] Rational resampler misunderstanding

[Marcus Müller]

Hi Jason,

so talking about the GNU Radio rational resampler:
> Under the hood, is GNURadio basically creating a FIR
> decimator?
Yeah! Multirate processing!
So what happens in the block that's called "rational resampler" in GRC
is that if you leave "taps" empty, the Python block [1] designs a FIR
that suppresses images (on interpolation) or aliases (on decimation),
whatever is harder (ie. anti-image filter if interp>decim, anti-alias
otherwise). Basically, a "little less than Nyquist max" low pass is built.
These taps are then used in the C++ rational_resampler_base_XXX, which
does a bit of polyphase magic. Which, when interp=1, leaves you with a
f_nyquist*0.8/decim LPF with decim decimation.

Cheers,
Marcus

[1] gr-filter/python/filter/rational_resampler.py
On 26.10.2015 13:27, Jason Matusiak wrote:
> I think I am overthinking something with the rational resampler.  I am
> working on an RFNoC block to incorporate a working gnuradio script we
> have, and am somehow being silly about the rational resampler.
>
> I am decimating by 4 (and not interpolating at all), to reduce the
> sample rate by 4.  Under the hood, is GNURadio basically creating a FIR
> decimator?  Is it that simple, or am I missing some other pieces?
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio