Title :
On preprocessing for mismatched classification of Gaussian signals
Author :
Ephraim, Yariv ; Roberts, William J.J.
Author_Institution :
Dept. of Electr. & Comput. Eng., George Mason Univ., Fairfax, VA, USA
fDate :
3/1/2001 12:00:00 AM
Abstract :
The optimal linear preprocessor for classifying two zero-mean Gaussian discrete-time signals which have been corrupted by additive zero-mean Gaussian noise is studied. Conditions for existence of the optimal linear preprocessor that achieves the performance of the likelihood ratio test for the noisy signals are given and the preprocessor is explicitly derived
Keywords :
Gaussian noise; optimisation; signal classification; additive zero-mean Gaussian noise; i.i.d sources; likelihood ratio test; mismatched classification; noisy signals; optimal linear preprocessor; performance; preprocessing; zero-mean Gaussian discrete-time signals; Additive noise; Australia; Covariance matrix; Degradation; Gaussian noise; Information technology; Signal design; Signal processing; Signal to noise ratio; Testing;
Journal_Title :
Information Theory, IEEE Transactions on
