基于组合算法的嵌入式系统软硬件划分方法

摘要

Hardware-software partitioning of embedded system design is a multiple constraint conditions,multi-objective combinatorial op-timisation problem,single algorithm is difficult to find the optimal design scheme.Therefore,we proposed a novel hardware-software partitio-ning algorithm for embedded system,which combines genetic algorithm with particle swarm optimisation.First,we built the mathematical model of hardware-software partitioning problem for embedded system,and then employed the genetic algorithm to obtain the feasible problem solution,finally used the particle swarm optimisation to find the optimal partitioning scheme,and carried out the simulation experiment to test the performance of the algorithm.Simulation results showed that the proposed method improved the solving efficiency on the problem,and could rapidly find more optimal scheme for hardware-software partitioning.%嵌入式系统软硬件划分是一个多约束条件、多目标的组合优化问题，单一算法难以找到最优设计方案，为此，提出一种遗传算法和粒子群算法组合的嵌入式系统软硬件划分方法。首先建立嵌入式系统软硬件划分问题的数学模型，然后利用遗传算法找到问题的可行解，最后采用粒子群算法找到最优方案，并采用仿真实验测试算法的性能。仿真结果表明，该方法提高了嵌入式系统软硬件划分问题的求解效率，可以快速找到更优的软硬件划分方案。