Study Of Software Reliability Testing

To begin with, this paper introduces the art of the state and the development in the research field of software reliability. The basic concepts related to software reliability and software reliability models are discussed. Then the mathematical knowledge used in the research field of software reliability is enumerated. New method and analysis are explored and investigated for software testing and software reliability models. The main results are as follows:1. A new method to compare the efficacy of partition testing and random testing is put forward. In this paper, we investigate the expected number of failures detected by partition testing and random testing without test cases replacement and compare the efficacy of partition testing and random testing strategies using the program reliability after testing.2. A model is proposed. In this model, imperfect debugging is considered in the sense that new faults can be introduced into the software during debugging and the detected faults may not be removed completely. A model is proposed to integrate fault removal rate, and fault introduction rate into software reliability assessment. Most SRGMs assume that faults detected during tests will eventually be removed. Consideration of fault removal efficiency in the existing models is limited.3. A new model is proposed,this model is the extension of NHPP software reliability models. In this model, the number of introduced fault and debugged fault is considered to be random variable,and the new model is used to estimate software reliability measures. In this article, we apply EM algorithm to estimating the values of the parameters of NHPP models. The estimating accuracy may be increased by this way. So the result software reliability analysis may be improved by this way too.