A Multi-layer Dependency Model for Analysis of Safety-critical Embedded Systems

Safety-critical embedded-system designs are typically both complex and expensive. Domains, such as medical devices, however, require safety but also demand affordability. However, conventional safety and reliability engineering methods, including redundancy or conventional dependency analysis, often lead to expensive and complex system designs. In this work, we propose a multi-layer dependency framework to analyze safety-critical systems. This framework captures fine-grained dependencies in safety-critical systems compared with traditional dependency graph analysis. Due to this new approach, we are able to verify the safety of systems with a reduced degree of redundancy, compared with conventional reliability engineering methods. To show the effectiveness of the multi-layer dependency framework, we apply it to four applications in the medical and control domains. These studies show a reduction in the complexity of the associated safety subsystems, which translates to both a reduction in cost and a reliability improvement for the safety subsystem. We specifically discuss the the applicability of our dependency framework to distributed medical systems where the conventional two-layer dependency model is unable to analyze the safety of complicated supervisory frameworks for such systems.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2011

Author dc:creator

• Rahmaniheris, Maryam

Contributors dc:contributor

• Sha, Lui R.

• Safety-critical
• Embedded Systems
• Dependency Model
• Reliability

Statement dc:rights

• Copyright 2011 Maryam Rahmaniheris

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/24480

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/24480