discuss-gnuradio
[Top][All Lists]
Advanced

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

Re: [Discuss-gnuradio] Where to find the "Decimating FIR Filter" for gr-


From: Anderson, Douglas J.
Subject: Re: [Discuss-gnuradio] Where to find the "Decimating FIR Filter" for gr-lte
Date: Fri, 24 Jul 2015 23:46:55 +0000

Thanks Kevin, I did try to fft_filter and it's working great.

I should have thought to grep through the grc source. Thanks...

-Doug

________________________________________
From: Kevin Reid address@hidden on behalf of Kevin Reid address@hidden
Sent: Friday, July 24, 2015 5:42 PM
To: Anderson, Douglas J.
Cc: GNURadio Discussion List
Subject: Re: [Discuss-gnuradio] Where to find the "Decimating FIR Filter" for 
gr-lte

On Jul 24, 2015, at 14:47, Anderson, Douglas J. <address@hidden> wrote:

> Hi all,
>
> I'm taking a read through Kristian Maier's thesis document on the gr-lte repo 
> (and by read I mean looking at the pictures because I don't know German).
>
> Looks like he was piping his Coarse PSS Sync block into a "Decimating FIR 
> Filter" block, but I can't seem to find that (I don't use GRC or even have it 
> built, so I'm just googling and looking through gnuradio.filter for the name 
> I would use via Python).

The decimating filter is the “normal” FIR filter. The corresponding Python name 
is gnuradio.filter.fir_filter_xxx. (Replace xxx with input, output, taps types, 
of course.)

I knew this already, but confirmed it by searching my copy of the GNU Radio 
sources, which turned up "Decimating FIR Filter" in 
gnuradio/gr-filter/grc/filter_fir_filter_xxx.xml.

> Alternately, I read through Tom's blob post on using fft_filter in place of 
> FIR filters: 
> http://www.trondeau.com/blog/2014/2/27/to-use-or-not-to-use-fft-filters.html, 
> and it seems like there may be some benefit to using it instead of the FIR 
> for the ~50 taps in Kristians filter.

fft_filter is a drop-in replacement for fir_filter, so feel free to try it out 
and see if it improves performance. The last machine I tested on, it was 
potentially helpful for anything over 10 taps.

--
Kevin Reid                                  <http://switchb.org/kpreid/>




reply via email to

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