> Well, that might be an idea to use both accelerometers. Currently > (in more is usually better than less ;-) I haven't yet looked at the machine learning part of the recognition you do (I know that you described it in your MS thesis I think), but doubling the number of features for classification/detection should have only positive effects here -- those features are not bogus and well correlated with each other, thus I would expect significant boost in performance. And taking the fact that recognition works quite well already with just a single accelerometer is great news -- it means that with 2 it should be just better ;-)
-- .-. =------------------------------ /v\ ----------------------------= Keep in touch // \\ (yoh@|www.)onerussian.com Yaroslav Halchenko /( )\ ICQ#: 60653192 Linux User ^^-^^ [175555] _______________________________________________ Openmoko community mailing list community@lists.openmoko.org http://lists.openmoko.org/mailman/listinfo/community