AN ENHANCED METHOD OF LMS PARAMETER ESTIMATION FOR SOFTWARE REALIABILITY MODEL

摘要

A software reliability model specifies the form of a random process that describes the behavior of software failures with respect to time. In this work the author has taken a Non Homogenous Poisson software reliability model called Gompertz model to predict the reliability of the software. It is shown that the proposed model can be derived from the well-known statistical theory of extreme value and has the quite similar sympatric property to the classical Gompertz curve. We have applied the Gompertz software reliability model to assess the software reliability and to predict the number of initial fault contents. The parameters used in this model are unknown, estimating this model parameter using an alternative approach of Least Mean Square estimation method. This new parameter estimation approach may function better than the existing estimation methods and is attractive in terms of goodness of fit, test based on information criteria and mean squared error. Software undergoes several stages of testing before it is put into operation. In every stage of testing, modification and correction are made with the hope of increasing reliability. All existing software reliability models are developed for the software products that are statically constructed normally by a company or institution that has the full control of the development process. The evolutional shift from the product-oriented software architecture to the Service Oriented Architecture (SOA) and Web Services (WS) invalids many techniques developed for traditional software. Hence in this work we have considered web based application with alternative approach for parameter estimation in Gompertz software reliability model.