基于边界扫描的FPGA等权值抗误判测试算法  

An equal-weight wrong-estimate algorithm for FPGA test based on boundary scan architecture

作　　者：吴继娟[1] 孙媛媛[1] 贾勇 

机构地区：[1]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2004年第4期509-512,共4页Journal of Harbin Engineering University

摘　　要：在FPGA(fieldprogrammablegatearray)的故障诊断中,故障征兆误判和混淆现象是影响故障诊断精度的主要问题.针对测试过程中的征兆误判的特征,给出抗误判定理,并在此基础上提出了一种边界扫描测试向量优化生成方法———等权值抗误判算法.该算法具备抗征兆误判的故障诊断能力,实现了测试向量集的紧凑性与故障分辨率之间合理的折衷,是一种实用的测试向量生成算法.实验表明该算法能消除征兆误判现象,并且在一定程度上能减少征兆混淆现象,经过编程仿真模拟,生成的测试向量集具备最优紧凑性指标.In the fault diagnosis of FPGA, the aliasing syndrome and confounding syndrome is the main problem affecting the accuracy of fault diagnosis. Aiming at the feature of the aliasing syndrome, an anti_liasing theorem was used to present a new optimal test vectors generation method of boundary scan test. The algorithm, also called the equal_weight wrong_estimate algorithm, can eliminate aliasing fault syndrome and can generate test vector set nearly as compact as that of modified counting sequence algorithm with much better diagnosis capacity. It is a test algorithm with good performance. Simulation experiments show that this algorithm can avoid aliasing syndrome and reduce confounding syndrome, and arrive at the least number of test vectors after simulating by programming.

关 键 词：FPGA 测试 抗误判定理 等权值抗误判算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]