Title: Flow Sensitive Slicing for MATLAB/Simulink Models
Abstract: MATLAB/Simulink is a widespread tool for modelbased software development within the automotive domain. Industrial sized models developed with Simulink often contain more than 20000 blocks connected by complex dependency relations. Those relations are mostly concealed by architectural pattern within the model. Common tools to discover dependencies during model development/maintenance are static analyses and slicing algorithms. In this paper we present a flow sensitive definition of data dependence for Simulink models for the inclusion within such analyses. It is tailored to describe dependencies hidden by the architecture of the model. This includes the distinction of data dependencies of virtual from nonvirtual blocks, the impact of buscapable blocks and bus signals. When integrated into a slicing algorithm, the relation enables accurate tracing of the atomic and composite signal flow of a model via its program dependence graph. We evaluate the created slicing algorithm with models from industrial case studies against another approach from the literature. During the evaluation of the slicing algorithms we could observe a reduction of the average slice sizes by up to 66%, due to the inclusion of the proposed data dependency relation.
Publication Year: 2016
Publication Date: 2016-04-01
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
Cited By Count: 8
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot