Robust Wavelet Shrinkage Estimation without Data Transform for Software Reliability Assessment

Abstract

Since software failure occurrence process is well-modeled by a non-homogeneous Poisson process, it is of great interest to estimate accurately the software intensity function from observed software-fault count data. In the existing work the same authors introduced the wavelet-based techniques for this problem and found that the Haar wavelet transform provided a very powerful performance in estimating software intensity function. In this paper, we also study the Haar-wavelet-transform-based approach, but without using approximate transformations. In numerical study with real software-fault count data, we compare the proposed robust estimation with the existing wavelet-based estimation as well as the conventional maximum likelihood estimation and least squares estimation methods.