Routing in 3D NoCs Using Genetic Algorithm and Particle Swarm Optimization

Networks-on-chip are a new concept in System-on-chip interconnections, facilitating and optimizing complex components integration. However, as it is a new technology, it still requires some research, especially concerning the acceleration and simplification of design phases. Networks-on-chip can be arranged in different topologies, such as hypercube, mesh and torus. Since several data packets can be transmitted simultaneously through the network, an efficient routing strategy must be used in order to avoid congestion delays. In this paper, we propose and evaluate the performance of two routing methods, based on genetic algorithm and particle swarm optimization, for Networks-on-chip with 3D mesh topology. The routing is driven by the minimization of total latency in packets transmission between tasks. The simulation results show that the routing based on genetic algorithm and particle swarm optimization outperforms other routing algorithms in terms of latency.