System identification: theory for the user
System identification: theory for the user
An Optimal Approach for Random Signals Classification
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
The aim of our study is to develop a computer aided diagnosis tool, allowing characterization of blood tissues from laser Doppler signals of microcirculation. A learning population of four classes corresponding to various local treatments inducing vasoconstriction or vasodilation is available. Signals are modelled by ARMA processes estimated by maximum likelihood identification. Several distances between signals are presented, and we show the first results obtained with the cepstral distance on a set of real data.