2011/10/1 Rudy Eschauzier <reschauz...@yahoo.com>: >> Plus, h1_deriv takes ts as argument but doesn't do anything with it. >> Is something missing one the code? > > This is probably a leftover from earlier experimentation and can be removed. > > As additional background information, I tracked down the paper I used to > write the code: > > http://www.ingelec.uns.edu.ar/pds2803/Materiales/Articulos/Cavicchi--ImpulseInvarianceAndMultipleOrderPoles.pdf > > I also found some feedback I got from Paul Kienzleback back in 2007. It was > in a private email and gives suggestions for optimizing the Octave code: > > 1) Vectorize. Octave has a slow interpreter, but the vector operations > are fast. E.g., functions such as polyrev can be written > > function y=polyrev(p), y=p(end:-1:1); end > > 2) Octave has prepad which is equivalent to pad_poly. > > 3) Octave/matlab have poly(p) which build a polynomial given its roots. > This is equivalent to: > > a=1; for i=1:length(p),a=conv(a,[1,-p(i)]); end > > 4) I suspect without having looked too closely at the code that > inv_residue could be simplified with the use of the 'roots' function. > > 5) Test cases are always good. E.g., > > %!assert(impinvar(b,a),expected_output,tolerance)
Thank you very much for digging this up. I made most of those changes and added that paper as reference on the help text. I just missed the last 2 suggestions from Paul Kienzleback if someone wants to do it. Carnë ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2dcopy2 _______________________________________________ Octave-dev mailing list Octave-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/octave-dev
