Abstract:

A genetic algorithm based hierarchical topology generation approach (GATG) is proposed to solve the problem of low energy design of applicationspecific network on chip (NoC). The aim is to reduce the network communication energy. Under the constraints of the bandwidth and latency, GATG automatically maps IP cores onto the selected routers according to the communication requirements of given application and the characteristics of router architectures. In addition, a recursion based routing path construction algorithm embedded in GATG is proposed to construct links between routers, and finally the application specific irregular NoC topology is formed. Experimental results show that GATG achieves a reduction by an average of 46.5% in energy consumption and 52.7% in router resource overhead compared with regular 2D Mesh topology.