A Novel Approach for Hardware Software Partitioning in Embedded Systems

One of the most important elements in the Codesign of modern Embedded Systems (ES) is the process of Hardware Software Partitioning (HSP), which has the objective of mapping the best partition to hardware (HW) part and the best partition to software (SW) part. Most of previous works deal with the HSP problem with the objective of optimizing two metrics, particularly, the execution time and the hardware cost. Another important metric to take in consideration while solving the HSP problem, is the power consumption. In this paper, we propose a heuristic approach to deal with the HSP problem with three metrics. The proposed approach aims to optimize the system in term of one metric while respecting constraints on the two other metrics. The algorithm consists of using the LR method with the combination of the 0-1 Knapsack algorithm (KP) and the Genetic Algorithm (GA). To validate the efficiency of the proposed approach, a comparisons with the Simulated Annealing (SA) and the Genetic GA algorithms has been preformed.