基于SRAM型FPGA的实时容错自修复系统设计方法

SRAM based FPGA system capable of runtime fault tolerance and recovery

为提高辐射环境中电子系统的可靠性,提出了一种基于SRAM型FPGA的实时容错自修复系统结构和设计方法。该设计方法采用粗粒度三模冗余结构和细粒度三模冗余结构对系统功能模块进行容错设计;将一种细粒度的故障检测单元嵌入到各冗余模块中对各冗余模块进行故障检测;结合动态部分重构技术可在不影响系统正常工作的前提下实现故障模块的在线修复。该设计结构于Xilinx Virtex^■-6FPGA中进行了设计实现,实验结果表明系统故障修复时间和可靠性得到显著提高。

In order to enhance the reliability of electronic system working in radiation environment, a reliability enhanced SRAM based FPGA system with runtime fault recovery is presented in this work. Coarse-grained TMR architecture and fine-grained TMR architecture are proposed to achieve the ability of fault tolerance. Fine-grained fault detection units are inserted in each redundancy module to mask the faulty module. The technique of partial dynamic reconfiguration is used to repair the faulty module without interrupting other parts of the system.The scheme is implemented in Xilinx Virtex^■-6 FPGA, and the results show that the improvements in recovery time and reliability are achieved by the proposed technique.