基于改进Markov模型软件测试策略  

Software Testing Strategy Based on Improved Markov Model

作　　者：李泽雪 李相民[1] 薛亮[2] 

机构地区：[1]海军航空工程学院,山东烟台264001 [2]海军装备研究院,北京100161

出　　处：《指挥控制与仿真》2016年第3期108-112,共5页Command Control & Simulation

摘　　要：针对已有软件测试Markov模型与工程实践不符的情况,通过引入软件需求覆盖率改进Markov模型。在改进的Markov模型基础上,本文以软件测试过程中测试总代价最小为控制目标,采用交叉熵方法修正测试剖面,由优化测试剖面生成测试用例序列。仿真结果表明这种方法能够有效地降低软件测试总代价,是一种有效的软件测试方法。A software testing method of optimizing testing strategy based on improved Markov model is presented in this paper. This method uses a new stochastic optimization method called Cross Entropy method. By iterative correcting of the testing profile in Markov model,we consider the minimization of testing cost and try to generate testing data automatically. The experimental results show that this optimization method is a promising option for tackling this problem. It can reduce the software testing cost effectively.

关 键 词：软件测试 交叉熵法 覆盖率 

分 类 号：TP311.53[自动化与计算机技术—计算机软件与理论]