Title :
Matrix Fisher inequalities for non-invertible linear systems
Author :
Vignat, Christophe ; Bercher, Jean
Abstract :
We show how Fisher inequalities first derived by Zamir (1998) can be proved and extended using a simplified approach. Cases of equality are detailed to reveal interesting links between notions of Gaussianity and invertibility.
Keywords :
Gaussian distribution; information theory; linear systems; matrix algebra; Gaussian distribution; MMSE estimate; invertibility; matrix Fisher inequalities; minimum mean square error; minimum norm solution; noninvertible linear systems; Computer aided software engineering; Cramer-Rao bounds; Data processing; Gaussian distribution; Gaussian processes; Information analysis; Linear matrix inequalities; Linear systems; Mean square error methods; Vectors;
Conference_Titel :
Information Theory, 2002. Proceedings. 2002 IEEE International Symposium on
Print_ISBN :
0-7803-7501-7
DOI :
10.1109/ISIT.2002.1023509
