A software reliability time series growth model with Kalman filter

Junhong Guo; Hongwei Liu; Xiaozong Yang; De Cheng Zuo

A software reliability time series growth model with Kalman filter

Junhong Guo^*
, Hongwei Liu
, Xiaozong Yang
, De Cheng Zuo

^*Corresponding author for this work

School of Computer Science and Technology, Harbin Institute of Technology
Heilongjiang University

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

Some of limitations of conventional software reliability models are discussed. Assumptions condition of software reliability growth models are difficult to be satisfied in actual projects which restricted the universality of models. And classical models neglect observation noise. An approach to overcoming these inadequacies is proposed. To evaluate software more accurately, this paper transforms new time series software reliability growth model into state space model and Kalman filter is used to reduce noise. Testing data of filtered noise can show the essential rule of data better and improve goodness of fit. Finally, numerical examples of software reliability analysis based on the actual testing data are presented.

Original language	English
Pages (from-to)	1-7
Number of pages	7
Journal	WSEAS Transactions on Computers
Volume	5
Issue number	1
State	Published - Jan 2006
Externally published	Yes

Keywords

Kalman filter
Observation noise
Software reliability growth model
State space
Time series

Cite this

@article{37bbc0cd1c2f40509697034d96f9d7fb,

title = "A software reliability time series growth model with Kalman filter",

abstract = "Some of limitations of conventional software reliability models are discussed. Assumptions condition of software reliability growth models are difficult to be satisfied in actual projects which restricted the universality of models. And classical models neglect observation noise. An approach to overcoming these inadequacies is proposed. To evaluate software more accurately, this paper transforms new time series software reliability growth model into state space model and Kalman filter is used to reduce noise. Testing data of filtered noise can show the essential rule of data better and improve goodness of fit. Finally, numerical examples of software reliability analysis based on the actual testing data are presented.",

keywords = "Kalman filter, Observation noise, Software reliability growth model, State space, Time series",

author = "Junhong Guo and Hongwei Liu and Xiaozong Yang and Zuo, \{De Cheng\}",

year = "2006",

month = jan,

language = "英语",

volume = "5",

pages = "1--7",

journal = "WSEAS Transactions on Computers",

issn = "1109-2750",

publisher = "World Scientific and Engineering Academy and Society",

number = "1",

}

TY - JOUR

T1 - A software reliability time series growth model with Kalman filter

AU - Guo, Junhong

AU - Liu, Hongwei

AU - Yang, Xiaozong

AU - Zuo, De Cheng

PY - 2006/1

Y1 - 2006/1

N2 - Some of limitations of conventional software reliability models are discussed. Assumptions condition of software reliability growth models are difficult to be satisfied in actual projects which restricted the universality of models. And classical models neglect observation noise. An approach to overcoming these inadequacies is proposed. To evaluate software more accurately, this paper transforms new time series software reliability growth model into state space model and Kalman filter is used to reduce noise. Testing data of filtered noise can show the essential rule of data better and improve goodness of fit. Finally, numerical examples of software reliability analysis based on the actual testing data are presented.

AB - Some of limitations of conventional software reliability models are discussed. Assumptions condition of software reliability growth models are difficult to be satisfied in actual projects which restricted the universality of models. And classical models neglect observation noise. An approach to overcoming these inadequacies is proposed. To evaluate software more accurately, this paper transforms new time series software reliability growth model into state space model and Kalman filter is used to reduce noise. Testing data of filtered noise can show the essential rule of data better and improve goodness of fit. Finally, numerical examples of software reliability analysis based on the actual testing data are presented.

KW - Kalman filter

KW - Observation noise

KW - Software reliability growth model

KW - State space

KW - Time series

UR - https://www.scopus.com/pages/publications/30644461482

M3 - 文章

AN - SCOPUS:30644461482

SN - 1109-2750

VL - 5

SP - 1

EP - 7

JO - WSEAS Transactions on Computers

JF - WSEAS Transactions on Computers

IS - 1

ER -

A software reliability time series growth model with Kalman filter

Abstract

Keywords

Other files and links

Fingerprint

Cite this