A novel neural network search for energy-efficient hardware-software partitioning

One of the most crucial steps in the design of embedded systems is hardware-software partitioning, that is, deciding which components of the system should be implemented in hardware and which ones are in software. The trends towards low power design of distributed embedded systems indicate the need for energy-efficient hardware-software partitioning algorithms, which is not enough emphasized so far. In this paper, a new formal model of energy-efficient hardware-software partitioning problem is proposed, and moreover, tabu search on a neural network, which is a novel heuristic algorithm, is constructed to solve the problem. Extensive experiments are conducted, including a realistic GPS encoder example, which demonstrate the effectiveness of the approach. Reductions in power consumption of up to 42.87% are reported, compared with genetic algorithm.