基于二阶Markov模型的改进相对熵测试充分性准则  

Improved relative entropy test adequacy criterion based on second-order Markov model

作　　者：张凡 李福川 陈丽容 吕中凯 ZHANG Fan;LI Fu-chuan;CHEN Li-rong;LYU Zhong-kai(Institute 706,Second Academy of China Aerospace Science and Industry Corporation,Beijing 100854,China;93160 Unit of the Chinese People’s Liberation Army,Beijing 100166,China)

机构地区：[1]中国航天科工集团第二研究院七〇六所,北京100854 [2]中国人民解放军93160部队,北京100166

出　　处：《计算机工程与设计》2022年第12期3428-3434,共7页Computer Engineering and Design

基　　金：国家重大工程支持基金项目(2020D204007)。

摘　　要：为解决基于一阶Markov模型的软件测试充分性准则不精确、充分性判定过程容易产生误判的问题,在分析现有的软件测试充分性准则的基础上,提出一种基于二阶Markov模型的改进相对熵测试充分性判别准则。消除单纯Markov模型的“一阶无后效性”影响并将原始相对熵进行边界化处理和对称性设计,使测试执行过程更加充分和稳定。数值实验结果表明,在不降低可信性的前提下,该方法能有效解决测试用例生成过程过早收敛的问题、增强测试充分性判定的稳定性,对单元测试的测试用例生成策略的优化设计有一定的指导意义。To solve the problem that the software testing adequacy criterion based on the first-order Markov model is inaccurate and the adequacy determination process is prone to misjudgment,an improved relative entropy test adequacy criterion based on the second-order Markov model was proposed on the basis of analyzing the existing software reliability test adequacy criteria.The first-order nil posteriority effect of the simple Markov model was eliminated,and the original relative entropy was bounded and symmetrically designed,which made the test execution process more stable and sufficient.The numerical experiment verifies that the proposed method can effectively solve the problem of premature convergence of the test case generation process and enhance the stability of test adequacy determination without reducing the credibility.At the same time,it has certain guiding significance for the optimization design of test case generation strategy of unit testing.

关 键 词：马尔可夫模型 软件测试 测试充分性准则 二阶马尔可夫 改进相对熵 测试用例生成 单元测试 

分 类 号：TP302.7[自动化与计算机技术—计算机系统结构]