|
| From: | Terry Duell |
| Subject: | Re: Any interest in a Human Response to Vibration package? |
| Date: | Thu, 07 Aug 2014 11:06:16 +1000 |
| User-agent: | Opera Mail/12.16 (Linux) |
On Wed, 06 Aug 2014 23:41:19 +1000, Damian Harty <address@hidden> wrote:
Anyway, Terry sharpened his comments up to suggest a dedicated ISO2631 package, which gets round all of my cautions. So now I'm all for it, and will help however I can.I'm currently looking at my code to see how it can be 'cleaned up'.
This is proving to be a bit of a revelation...it is interesting to see what one has forgotten when digging up history! I first implemented functions to do ISO2631 HRV analysis in Rlab2, back in about 2000. Those functions were adapted from some Matlab functions provided to me by a colleague. The Matlab functions used the Matlab bilinear transform with prewarp, and I implemented an Rlab2 bilinear function to do that. I have since migrated to Linux and Octave and implemented all those functions in Octave. The original Matlab functions are open, but I don't know what was the basis of my Rlab2 bilinear function. I can't remember doing it from scratch, and can't find anything that would have served as my prototype. As a result it does seem that I shouldn't release this bilinear function until I can be sure it is OK to do so. An alternate is if anyone has an open bilinear transform function that is compatible with the Matlab function (i.e has prewarp) that I can test. I should add that I have noted some comments and a bug report to the effect that Octave's bilinear is not compatible with Matlab's bilinear, but thus far there hasn't been any sign that anyone has taken up the challenge.
Terry -- Regards, Terry Duell
| [Prev in Thread] | Current Thread | [Next in Thread] |