连续执行软件可靠性验证测试方法  

Reliability Demonstration Testing Method for Continuous Execution Software

作　　者：覃志东[1] 雷航[1] 桑楠[1] 熊光泽[1] 

机构地区：[1]电子科技大学计算机学院实时系统研究室,成都610054

出　　处：《计算机科学》2005年第6期202-205,共4页Computer Science

基　　金：国防预研基金(41315040106);国家高技术研究发展863计划专项经费(2003AA1Z2210);电子科技大学优秀青年教师基金(L08010601YF020806)

摘　　要：针对基于经典统计假设检验的固定期测试方法测试持续期太长,难以满足现代连续执行软件可靠性验证测试的需要,提出了一种基于经验贝叶斯统计推断的连续执行软件可靠性验证测试方法。该方法通过可靠性增长测试阶段的测试记录获得软件失效强度的先验分布,从而得到所需要的验证测试持续期,并结合检证测试过程中的具体情况,提出了先验动态整合的方法。实验表明,该方法在不降低验证测试结果可信性的条件下,能有效地减少可靠性验证测试的持续期。To the question that fixed duration testing method, which bases on the classical statistical hypothesis test- ing, can't fulfill the requirements of reliability demonstration testing for modern continuous execution software due to the long testing duration, a new method to demonstrate the reliability for continuous execution software is presented basing on the empirical Bayesian statistical reference. Firstly, the prior distribution of software failure intensity is de- rived on the basis of analyzing the testing records of software reliability growth testing. Then, the reliability demon- stration testing duration can be calculated. Finally, the prior knowledge dynandc integration method taking the specif- ic testing information into account is provided. Experiment shows that the method provided above is effective to re- duce the testing duration without decreasing the confidence level in the testing results.

关 键 词：可靠性验证 测试方法 执行 验证测试 统计假设检验 软件可靠性 可靠性增长 持续期 统计推断 先验分布 软件失效 测试阶段 测试过程 动态整合 贝叶斯 可信性 

分 类 号：O212.1[理学—概率论与数理统计] V243[理学—数学]