Semester

Spring

Date of Graduation

1999

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Hany H. Ammar.

Abstract

One of the major challenges facing the software industry, in general and IV&V (Independent Verification and Validation) analysts in particular, is to find ways for analyzing dynamic behavior of requirement specifications of large software systems early in the development lifecycle. Such analysis can significantly improve the performance and reliability of the developed systems. This dissertation addresses the problem of developing an IV&V framework for extracting semantics of dynamic behavior from requirement specifications based on: (1) SART (Structured Analysis with Realtime) models, and (2) UML (Unified Modeling Language) models.;For SART, the framework presented here shows a direct mapping from SART specification models to CPN (Colored Petrinets) models. The semantics of the SART hierarchy at the individual levels are preserved in the mapping. This makes it easy for the analyst to perform the analysis and trace back to the corresponding SART model. CPN was selected because it supports rigorous dynamic analysis. A large scale case study based on a component of NASA EOS system was performed for a proof of the concept.;For UML specifications, an approach based on metamodels is presented. A special type of metamodel, called dynamic metamodel (DMM), is introduced. This approach holds several advantages over the direct mapping of UML to CPN. The mapping rules for generating DMM are not CPN specific, hence they would not change if a language other than CPN is used. Also it makes it more flexible to develop DMM because other types of models can be added to the existing UML models. A simple example of a pacemaker is used to illustrate the concepts of DMM.

Download

Share

COinS