From 43529f3e92d06ea3560322ba012faa3b2c3db33a Mon Sep 17 00:00:00 2001
From: Eric Kooistra <kooistra@astron.nl>
Date: Mon, 19 Aug 2024 08:24:59 +0200
Subject: [PATCH] Improved description.
---
applications/lofar2/model/rtdsp/firfilter.py | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/applications/lofar2/model/rtdsp/firfilter.py b/applications/lofar2/model/rtdsp/firfilter.py
index 8ae1a6c419..b7bab7bb45 100644
--- a/applications/lofar2/model/rtdsp/firfilter.py
+++ b/applications/lofar2/model/rtdsp/firfilter.py
@@ -179,7 +179,7 @@ def design_fir_low_pass_filter(method,
. Method 'firls' does not support zero BW at fcutoff. With single fcutoff point error occurs: 'bands must be
monotonically nondecreasing and have width > 0'. Work around: use a very narrow band cutoffBW at fcutoff.
The cutoffBW band has to be much smaller than the transition band transistionBW, to avoid a flattened gain
- around fcutoff, but also not too small, to because then the cutoffGain at fcutoff is not achieved.
+ around fcutoff, but also not too small, because then the cutoffGain at fcutoff is not achieved.
Default force cutoffBW = transistionBW / 10 for 'firls' when input cutoffBW=0. For 'remez' method the
cutoffBW=0 effectively, without work around, because 'remez' does support single fcutoff value specification.
@@ -300,8 +300,8 @@ def design_fir_low_pass_filter_adjust(method,
kaiserBeta=0, fs=1.0, resolution=1e-4, verbosity=1):
"""Derive FIR coefficients for low pass filter for exact fcutoff
- Uses design_fir_low_pass_filter() but achieves cutoffGain at fcutoff by adjusting fpass or fstop, and
- fpass < fcutoff < fstop.
+ Uses design_fir_low_pass_filter() but achieves cutoffGain at fcutoff by iteratively adjusting
+ fpass or fstop, and fpass < fcutoff < fstop.
Input:
. resolution: Resolution measure for frequency and for gain. It appears that at some fine resolution
--
GitLab