Application Mapping Onto Network on Chip Using Particul Swarm Optimisation With Genetic Algorithm “GAPSO”

Abstract

Network-on-chip is a new concept of interconnection in single-chip systems. This architecture is associated with the traditional methods of design of the interconnections which aims to carry out several functionalities and to stage the limits of that of the traditional methods. However, like any new technology, it requires research efforts, in particular to speed up and simplify the design phases. The mapping phase is a main step in the network-on-chip design process, it has a direct impact on system performance. This phase makes it possible to assign each task of an application to a physical resource while respecting the imposed constraints. This work aims to evaluate the performance for single-objective placement based on particle swarm optimization PSO and hybridization of this algorithm with the genetic algorithm.