Cellular neural network design using a learning algorithm

Abstract

A learning algorithm for cellular neural networks (CNN) is proposed. The cloning templates can be obtained by using this algorithm, which is based on the relaxation method for solving sets of linear inequalities. The symmetry of templates can be forced through additional equality constraints. Simulation examples show that some useful templates with the smallest neighborhood N/sub 1/(i, j) are generated by the application of the training rule.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>