A new generalized cellular automata approach to optimization of fast packet switching

Abstract

The optimization of fast packet switching (FPS) in computer networks is of great significance for improving the network performance. This paper presents a new generalized cellular automata (GCA) approach to effectively solve the FPS optimization problem. In contrast to the Hopfield-type neural network (HNN) and cellular neural network (CNN), the proposed GCA approach is featured by the pyramid architecture that is composed of multi-granularity macro-cells, and by the evolutionary dynamics that involves the dynamical feedbacks among macro-cells. The GCA architecture, dynamics, algorithm and properties are discussed in the context of the FPS optimization. The analysis and simulations on the FPS optimization have shown that the GCA approach has advantages over the HNN and CNN methods in terms of the solution quality, optimal ratio, convergence speed, real-time performance, interconnection complexity, and parameter selection.