APP下载

基于改进蚁群算法的嵌入式系统软硬件划分

2017-03-04潘颖阮文惠

现代电子技术 2017年3期

潘颖 阮文惠

摘 要: 针对标准蚁群算法的软硬件划分问题求解难题,提出改进蚁群算法的系统软硬件划分方法。首先分析了当前嵌入式系统软硬件划分研究的现状,并构建软硬件划分的数学模型;然后采用蚁群算法模拟蚂蚁觅食行为搜索数学模型的最优解,并引入逆反馈机制提高蚁群算法的搜索性能;最后通过实验证明软硬件划分问题求解的有效性。实验结果表明,改进蚁群算法提高了问题求解的效率,获得了合理的软硬件划分结果,且结果优于标准蚁群算法。

关键词: 硬件系统; 蚁群算法; 软件系统; 收敛速度; 仿真测试

中图分类号: TN919?34; TP181 文献标识码: A 文章编号: 1004?373X(2017)03?0164?03

Embedded system hardware and software partition

based on improved ant colony algorithm

PAN Ying, RUAN Wenhui

(School of Digital Media, Lanzhou University of Arts and Science, Lanzhou 730030, China)

Abstract: Since it is hard for the standard ant colony algorithm to solve the software and hardware partition problem, the hardware and software partition method based on the improved ant colony algorithm is proposed. The research status of the embedded system hardware and software partition is analyzed. The mathematical model of the hardware and software partition was constructed. The ant colony algorithm is used to simulate the foraging behavior of ants to search the optimal solution of the mathematical model. The inverse feedback mechanism is introduced to improve the search performance of the ant colony algorithm. The solving validity of the hardware and software partition problem was proved by means of experiments. The results show that the improved ant colony algorithm has improved the efficiency of problem solving, and obtained the reasonable hardware and software partition result, which is better than the result of the standard ant colony algorithm.

Keywords: hardware system; ant colony algorithm; software system; convergence speed; simulation test

0 引 言

随着计算机技术和微电子技术的不断成熟,嵌入式系统的应用范围更加广泛[1]。嵌入式系统包括硬件系统和软件系统两部分,以前由于硬件成本高,价格高,软件系统占的比例大。近几年,随着集成技术的发展,硬件价格急剧下降,出现了硬件系统和软件系统平分天下的格局[2]。硬件系统和软件系统均有各自的优势,对于一个嵌入式系统,如何对硬件系统和软件系统进行准确划分,降低系统的成本,同时保证系统满足實际要求,成为人们关注的焦点[3?4]。

自嵌入式系统出现以后,很多机构和大学专门成立了嵌入式软硬件系统划分研究团队,提出了许多有效的嵌入式软硬件划分算法[5]。最原始的方法为整数规划算法,工作灵活,可以适应不同类型的嵌入式系统,但整数规划算法的嵌入式软硬件系统划分问题求解时间长、速度慢,不能满足嵌入式系统软硬件划分的效率[6]。……

登录APP查看全文