Table of Contents

Class: LanczosFilter LanczosFilter.py

LinearFilter derived class for Lanczos filtering

For notation and mathematical details, see Duchon, 1979, Journal of Applied Meteorology, Lanczos filtering in one and two dimensions, volume 18, pages 1016-1022.

Base Classes   
pyclimate.LinearFilter.LinearFilter
Methods   
__init__
getcoefs
prettyprintcoefs
  __init__ 
__init__ (
        self,
        filtertype='bp',
        fc1=0.0,
        fc2=0.5,
        n=None,
        )

Constructor for the class LanczosFilter

Initialize the variables to hold the needed data The filter works as follows, first, it gets the coefficients to be used to weight each record performing a recursive filtering on the impulse response function Next, filter the records by means of a running buffer which is weighted using the weights... and that's all

Optional Arguments:

filtertype
String identifying the filter type. lp for low-pass filter, hp for high-pass filter and bp for band-pass filter.
fc1
First cutoff frequency (in inverse time steps). Defaults to 0.0
fc2
Second cutoff frequency (in inverse time steps). Defaults to 0.5
n
Number of points for the filter. This number of points will be missed at the beginning and at the end of the raw record. Defaults to a number a 30% higher than that recomended for secure band-pass filtering by Duchon (1979).

NOTE: With the default arguments the filter does NOT filter at all. It performs a band-pass filtering for all frequencies.

NOTE 2: If the first and second cutoff frequencies are different band-pass filtering is assumed even though lp or hp filtertype were selected.

  getcoefs 
getcoefs ( self )

Filter coefficients

  prettyprintcoefs 
prettyprintcoefs ( self )

Pretty-print the coefficients


Table of Contents

This document was automatically generated on Mon Dec 9 15:58:37 2002 by HappyDoc version 2.1