基于部分覆盖表的错误交互定位方法

Locating Error Interactions Based on Partial Covering Array

在组合测试定位模型的基础上提出了部分覆盖表的错误交互定位方法,该方法在错误交互个数已知的条件下,通过生成部分覆盖表,利用测试用例运行结果提供的信息来对软件错误交互定位.从理论上证明了部分覆盖表等价于一类特殊的错误定位表,进而研究了部分覆盖表行数的上界,提出生成部分覆盖表的贪心算法,从而给出了定位引发软件故障的错误交互的非自适应算法,并通过实验验证了该方法的有效性.

Combinatorial testing is practical and effective method to detect the faults triggered by the interactions among parameters or components in the system. It is a key problem that how to locate the faulty interactions after some faults were detected in the test stages. Error locating arrays （ELAs） were defined by Martinez C et al for detecting and locating the faulty interactions in a system. In this paper, the authors prove that a special partial covering array （PCA） is also a special error locating array and study the upper bound of PCA＇s size. It is obtained that the number of tests given by PCAs is polynomial in logk and d, where k is the number of parameters and d is the number of faulty interactions in the system. Furthermore the authors propose the greedy algorithms to generate the special PCAs, and these algorithms are also a non-adaptive algorithm of producing ELAs. The experimental results show the above approach for detecting and locating the faulty interactions using PCAs is effective.