[Top][All Lists]

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

Re: [Discuss-gnuradio] GRC filter tap evaluations - difference between c

From: Josh Blum
Subject: Re: [Discuss-gnuradio] GRC filter tap evaluations - difference between current version and v 0.7
Date: Tue, 21 Apr 2009 15:58:04 -0700
User-agent: Thunderbird (X11/20090318)

It has to be evaluatable by the python parser.

map(lambda x: x/32768., [-23, 4, 245, 1235, 245, 4, -23])
numpy.array([-23, 4, 245, 1235, 245, 4, -23])/32768.

If you want to use the numpy, then make sure to include an import block with the line: import numpy


Richard Clarke wrote:
Hi All,

With version 0.7 of GRC I could enter the taps of a filter in the following
format [-23, 4, 245, 1235, 245, 4, -23]/32768 in the relevant GRC block
window  (taps field) and GRC would do the division calculation for each tap,
i.e the tap vector automatically became [-7.019e-4,1.22e-4, 7.47e-3,
3.76e-2, 7.47e-3, 1.22e-4, -7.019e-4].

Trying to do the same thing in the current version of GRC that is included
with GNU Radio 3.2rc2 results in a warning message that the tap vector
cannot be evaluated. Does anyone know how to achieve the same thing as
demonstrated above for version 0.7?
A small thing I know but most coefficients of filters that I have come as
fixed point 16 bit integers (-32768<=coeff<=32767) and doing a copy/paste is
more convenient than doing the extra calculation and then formatting the
resultant floating point vector into a copy and pasteable form for entry
into the GRC blocks.

Thanks very much.



Discuss-gnuradio mailing list

reply via email to

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