Template design methods for binary cellular neural networks

The problem of template design for cellular neural networks (CNNs) with binary outputs is addressed. A theorem is provided, that yields some rigorous conditions for the correct behavior of a CNN and allows us to develop exact and simple design rules.