首页> 外文会议>International Symposium on VLSI Design, automation, and Test >Improve transition fault diagnosability via observation point insertion
【24h】

Improve transition fault diagnosability via observation point insertion

机译:通过插入观察点提高过渡故障的可诊断性

获取原文

摘要

In this work, a design for diagnosability (DFD) method based on observation point (OP) insertion is proposed to improve the diagnosis resolution of transition faults in a circuit. The main objective is to minimize the number of observation points since this number will directly affect the area overhead of the circuit. We develop a novel algorithm to generate a set of OP candidates and then select a minimal number of OPs from this set which can distinguish all targeted fault pairs. An observation point insertion logic is also proposed that can efficiently reuse the output pins in the original circuit so as to reduce the number of extra output pins. In addition, a novel structural distance calculation method for synthesized circuits is proposed that considers the mixed structure of primitive gates and complicated gates, including AOI or OAI gates. Experimental results show that after applying the OP insertion method, all aborted fault pairs can be distinguished and the number of required observation points is quite small. We also use the observation points to distinguish those indistinguished far-away fault pairs. Experimental results show that all targeted fault pairs can be distinguished with a few observation points and a set of diagnosis patterns for ISCAS89 and ITC99 circuits.
机译:在这项工作中,提出了一种基于观察点(OP)插入的可诊断性(DFD)方法设计,以提高电路中过渡故障的诊断分辨率。主要目的是减少观察点的数量,因为该数量将直接影响电路的面积开销。我们开发了一种新颖的算法来生成一组OP候选项,然后从该组中选择最小数量的OP,以区分所有目标故障对。还提出了一种观察点插入逻辑,该逻辑可以有效地重用原始电路中的输出引脚,从而减少额外的输出引脚的数量。此外,提出了一种新颖的合成电路结构距离计算方法,该方法考虑了原始门和复杂门(包括AOI或OAI门)的混合结构。实验结果表明,采用OP插入方法后,可以区分所有异常中止的故障对,所需的观察点数量也很少。我们还使用观察点来区分那些无法区分的远距离断层对。实验结果表明,可以通过几个观察点和一组针对ISCAS89和ITC99电路的诊断模式来区分所有目标故障对。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号