Modeling communication cost and hardware alternatives in PSO based HW/SW partitioning

In this paper we investigate the modeling of the hardware implementations alternatives as well as the modeling of the communication cost in the particle swarm optimization (PSO) based hardware/software partitioning. The communications modeling accounts for the extra area and delay costs due to the realization of the hardware/software interface. The hardware alternatives are represented by two extreme points that define the area and delay ranges for different hardware implementations. A fine tuning heuristic is proposed to enhance the quality of the solution and guarantees that it meets certain constraints if possible.