Time-frequency representations (TFRs) are efficient tools for nonstationary
signal classification. However, the choice of the TFR and of the distance
measure employed is critical when no prior information other than a learnin
g set of limited size is available. In this letter we propose to jointly op
timize the TFR and distance measure by minimizing the (estimated) probabili
ty of classification error The resulting optimized classification method is
applied to multicomponent chirp signals and real speech records (speaker r
ecognition). Extensive simulations show the substantial improvement of clas
sification performance obtained with our optimization method.