From: Hendrik-Jan van Veen <veen@mpik-tueb.mpg.de>
Subject: Re: TECH: Filter on the output signal of a reciever
Date: Mon, 04 Nov 1996 09:00:43 +0100
Message-ID: <327DA2AB.41C6@mpik-tueb.mpg.de>
Organization: Max-Planck-Institute for Biological Cybernetics


Leon Heller wrote:
> 
> In article <32774E4B.2F1C@nlr.nl> Hilbrink@nlr.nl "Niels Hilbrink" writes:
> 
> > I was wondering if anybody has ever tried to put a noise filter on the
> > output signal of tracker (Polhemus Fastrak) ?
> <...>
> > * Has anybody ever used a filter?
> > * What kind of filtre was used, Kalman, Wavelet Transform, Fast Fourier
> >   Transform, etc. ?
> > * What is the effect of the filter on system latency ?

Have you considered using the internal adaptive filters of the
fastrak?  Appendix C of the (my?) manual deals with this and does
explain latency side-effects.

With a previous polhemus system I did use my own filtering for the
orientation quaternions: low-pass third order filter with a cut-off
frequency of 12Hz using off-line FFT. For my application this seemed
to be optimal. In still another case I used Lagrange's Classical
Formula to predict sinusoidal headmovements. There exist better
methods but a combination of prediction and filtering might do the job
for you.


Dr. Hendrik-Jan van Veen
Max-Planck-Institute for Biological Cybernetics
Spemannstrasse 38 -- 72076 Tuebingen -- GERMANY
E-MAIL: veen@mpik-tueb.mpg.de 
EUROPE: http://www.mpik-tueb.mpg.de/people/personal/veen/veen.html
USA: http://www.cog.brown.edu/mpik-mirror/people/personal/veen/veen.html