Equations for the location of Paynter Filter poles

[Kral]

Hi,

I'm interested in equations for the pole locations of the Paynter
filter. The publication "The Lightning Empericist" has the equations
for the even order Paynter filter pole locations, and for the third
order pole locations, but no equations for higher odd order filters.


The Paynter filter is an approximation to a linear phase filter. The
linear phase approximation is exact at integer multiples of w0/4,
where
w0 is the cutoff frequency. It is not an equi-ripple filter. The
values of the successive phase error ripple peaks increase with
frequency.

 

[flatulent]

Re: Paynter Filter Poles

http://biosig.sourceforge.net/help/biosig/t300/ is a matlab program for this. Even if you do not have mattlab, you can examine this file and back out the calculation method.

 

[Kral]

Re: Paynter Filter Poles

f,

I visited the biosig... site. The file describes the calling sequence for a Matlab function. I'm not familiar with Matlab, but there doesn't seem to be a way to get at the code.

Regards,
K

 

[flatulent]

Re: Paynter Filter Poles

https://jap.physiology.org/cgi/content/full/84/1/378#B13 is a discussion of this filter. It appears to be an ad hoc type.

From the graphs and the use, a parabolic filter would also work. It has the fastest rise time without overshoot for a given noise bandwidth. There is a graphical method (in the S plane) to obtain the pole locations which can be reduced to algebra and trigonometry.

This is discussed in M. S. Ghausi, Principles and Design of Linear Active Circuits, McGraw Hill 1965 page 90 This gives references to

S. K. Mullick, Pulse Networks With Parabolic Distribution of Poles,

IRE Transactions on Circuit Theory, CT-9, pages 302-305, September 1963