一种基于边界扫描的FPGA测试优化生成算法

在FPGA(fieldprogrammablegatearray)的故障诊断中,经常需要限定测试向量集的紧凑性指标.针对在限定紧凑性指标的前提下提高测试向量集的故障诊断能力问题,在分析了传统最小权值算法的基础上,提出了一种基于边界扫描的FPGA测试优化生成算法———最小权值优化算法.该算法要求所生成的测试向量集具备最小的权值,而且对权值最大的向量子集中的各向量施加了必须的约束条件.实验表明,最小权值优化算法能够减少征兆误判现象和征兆混淆现象,而且经过编程仿真模拟,优化后的算法的征兆混淆率要优于原有的最小权值算法,是一种实用的测试生成优化算法.

An optimal FPGA test algorithm based on boundary scan architecture

In the fault diagnosis of FPGA(field programmable gate array), the number of test vectors often needs to be limited. Aiming to improve the fault diagnosis capacity under the precondition of limited test vectors, a traditional minimum_weight algorithm was analyzed to present a new optimal FPGA generation method of boundary scan test. The algorithm, also called the optimal minimum_weight algorithm, requires that the test vector set generated has the minimum weight and the vectors in the vector subset of the largest weight have the maximum difference. Simulation experiments show that this algorithm can greatly reduce aliasing syndrome and confounding syndrome, and has a better level of confounding syndrome than the traditional minimum-weight algorithm after simulating by programming. It is a practical test algorithm with good performance.