作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (19): 9-11,16. doi: 10.3969/j.issn.1000-3428.2011.19.003

• 专栏 • 上一篇    下一篇

嵌入式系统门级在线自修复技术研究

满梦华a,原 亮b,巨政权a,解双建b,褚 杰a   

  1. (军械工程学院 a. 静电与电磁防护研究所;b. 计算机工程系,石家庄 050003)
  • 收稿日期:2011-03-28 出版日期:2011-10-05 发布日期:2011-10-05
  • 作者简介:满梦华(1984-),男,博士研究生,主研方向:电磁防护仿生,智能故障检测与诊断,容错技术;原 亮,教授;巨政权,博士研究生;解双建,硕士研究生;褚 杰,讲师

Research on Gate-level Online Self-repairing Technology for Embedded System

MAN Meng-hua  a, YUAN Liang  b, JU Zheng-quan  a, XIE Shuang-jian  b, CHU Jie  a   

  1. (a. Electrostatic & Electromagnetic Protection Institute; b. Department of Computer Engineering, Ordnance Engineering College, Shijiazhuang 050003, China)
  • Received:2011-03-28 Online:2011-10-05 Published:2011-10-05

摘要:

为保证复杂电磁环境下嵌入式系统的安全运行,提出一种通用嵌入式系统在线故障自修复模型。基于ARM处理器和商业FPGA芯片实现原型试验系统,借鉴生物系统在遭受一定损伤时能自行康复的特点,结合仿生硬件领域中的内进化试验方法,在原型系统上设计重配置执行单元的门级冗余拓扑结构和自律控制单元的演化修复算法。利用马尔可夫模型评价系统的可靠性与稳态可用度。结果表明,在随机电磁损伤事件频繁发生并导致电路功能故障时,该系统能保持较高的可用度和安全性。

关键词: 嵌入式系统, 自修复, 电磁仿生, 马尔可夫模型, 仿生硬件

Abstract:

The reliability of embedded system has become an integral part of the designing process, especially for those systems applied in the complex electromagnetic environment. In this paper, the online self-repairing model based on general embedded systems is proposed. The prototype system is realized with the ARM processor and commercial FPGA chip. Based on the self-healing characteristic of injured biological systems and the experiment methods of intrinsic Bio-inspired Hardware(BHW), the reconfigured executing unit and self-discipline controlling unit are implemented. The redundant gate array is in each reconfigurable executing unit, while the self-discipline controlling unit implements the genetic algorithm for self-repairing. Markov model is used to analyze the steady-state availability of the prototype. The result improves that it can keep higher available degree and security when the damaging event in the complex electromagnetic environment happened frequently.

Key words: embedded system, self-repairing, electromagnetic bionics, Markov model, Bio-inspired Hardware(BHW)

中图分类号: