Hallo!
adaptive is a PD-external library for adaptive systems and filters.
We (Gerda Stroble + me) implemented various versions of the LMS (least mean square) and NLMS algorithm and also added a lot of examples (see PD-patches in example folder): system identification, interference cancelation, adaptive equalization, decision-directed equalization, ...
A RLS (recursive least square) algorithm and more example patches (e.g. echo cancelation, ...) will follow when I have to learn for the exam ... :)
For more documentation and tarballs look here: http://grh.mur.at/software/adaptive.html
LG Georg
Wow, this is wonderful. Great work. I think this is just what was missing for a long time, and I never wanted to get into the nitty gritty myself! I think maybe another good example would be an 'adaptive line enhancer' for something like automatic removal of mains noise from broadband signals, but its not needed, it's already nice as it is. Regards,
Matt
-=-=-=-=-=-=-=-=-=-=-=- http://www.loopit.org -=-=-=-=-=-=-=-=-=-=-=- ----- Original Message ----- From: "Georg Holzmann" grhPD@gmx.at To: pd-announce@iem.at; pd-list@iem.at Sent: Monday, July 11, 2005 1:28 PM Subject: [PD] library for adaptive systems/filters
Hallo!
adaptive is a PD-external library for adaptive systems and filters.
We (Gerda Stroble + me) implemented various versions of the LMS (least mean square) and NLMS algorithm and also added a lot of examples (see PD-patches in example folder): system identification, interference cancelation, adaptive equalization, decision-directed equalization, ...
A RLS (recursive least square) algorithm and more example patches (e.g. echo cancelation, ...) will follow when I have to learn for the exam ... :)
For more documentation and tarballs look here: http://grh.mur.at/software/adaptive.html
LG Georg
Hallo!
I think maybe another good example would be an 'adaptive line enhancer' for something like automatic removal of mains noise from broadband signals, but its not needed, it's already nice as it is.
:) - yes, of course, there could (and maybe will) be much more ... Also the code of the LMS and NLMS is not really optimized, but maybe I'll do it when I need it ...
LG Georg