Research On Software Reliability Model Based On The Process Of Fault Detection And Correction

With the continuous development of computer technology, software is widely used in every field of national economy and national defense construction, expanding the scale and complexity is gradually increasing. At the same time,how to guarantee the quality of the software has become the focus of attention and research. Software reliability index is an important attribute of the software quality. It is quantitatively described the failure in the software life cycle. Software reliability growth model is an important means of software reliability prediction using statistical.Software testing mainly includes two processes: fault detection and fault correction. The effection of test effort on the influence of the test effort on the reliability of software is is researched.Software reliability growth model based on the process of fault detection and correction is established. The main research content is as follows:Firstly, there are ten kinds of existing test effort function. Due to poor applicability, the researchers have very big workload to modeling and the result is not satisfactory. This paper proposes a new test effort function. The function form is more flexible, more resilient. Then,software reliability growth model based on the new test effort function is established. Using published data to validate the feasibility of new test effort function and the modeling method.Secondly, Aim at the problem of the existing software reliability growth model based on the new test effort function ignores the fault correction process, this paper puts forward the concepts of fault detection effort and fault correction effort and establish software reliability growth model based on detection effort and correction effort.Then, using two sets of published data to validate the fitting and forecast capability of the new model.Thirdly, the problem of the change points is researched and the traditional method of change points estimationis analyzed.Then,it proposes a progressive adjustment method to calculate locations of change points accurately. The feasibility and effectiveness of the new method are examined by using two public software failure data sets. The results show that the proposed model with new change points fits the failure data better and can predict two sets of data more accurately.Lastly, this paper introduces the electronic chart system and applies three software reliability growth models to predict the reliability of the electronic chart system.