Title :
Karhunen-Loeve transform using neural networks
Author :
Zhu, Xianing ; Zhang, Shengwei ; Constantinides, A.G.
Author_Institution :
Ocron Inc., Santa Clara, CA, USA
fDate :
27 Jun-2 Jul 1994
Abstract :
The optimality of Karhunen-Loeve transform (KLT) over other transforms has been well known, together with the difficulty in implementing practical KLT systems. The wide applications of the transform deserve a new investigation on realizing such systems by using artificial neural networks. In this paper the KLT is known to be equivalent to a constrained optimization problem by maximizing covariance of output signals with the constraint of orthonormality. A neural network is then developed which can converge to the basis vectors of the transform
Keywords :
Artificial neural networks; Computer networks; Constraint optimization; Covariance matrix; Educational institutions; Eigenvalues and eigenfunctions; Equations; Karhunen-Loeve transforms; Matrix converters; Neural networks;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.375007
