Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


34243.pdf (2.65 MB)

ETD Abstract Container

Abstract Header

Software Development Process and Reliability Quantification for Safety Critical Embedded Systems Design

Lockhart, Jonathan A
http://orcid.org/0000-0002-1298-791X
http://rave.ohiolink.edu/etdc/view?acc_num=ucin1562673285477425

Abstract Details

2019, PhD, University of Cincinnati, Engineering and Applied Science: Electrical Engineering.
Embedded systems are at the forefront of everyday life, being utilized in smart devices, such as cell phones and internet of things (IoT), devices around the home, as well as the latest components in aerospace and automobile technology. Reliance on these devices is critical to the current day to day operations of society, and these devices are required to be secure and reliable to maintain the safety of those who depend on them. Among these devices, trusted safety critical embedded systems are rigorously designed with security and reliability so they can be counted on to perform their assigned responsibilities with a low probability of failure, as such a failure could cost people their lives in the worst case scenario. Trusted embedded systems are often developed with hardware, using the latest in field programmable gate arrays (FPGA), and integrated circuits (ICs), as hardware development has a long established process for producing high quality, fault tolerant systems and reporting performance in a standardized way. The processes utilized are mature and repeatable, and this shows the reliability of these systems is consistent. These systems have increased in complexity, performing more and more tasks with each incremental increase in hardware performance. Unfortunately the end to Moore’s Law is coming, and though the development of new architectures and techniques has allowed for the end to be delayed, a shift in design is required to continue increasing the complexity of trusted embedded systems. Software is being looked at to continue the trend of complex, high performance systems, but suffers from its utilization in modern, agile development environments and the use of unreliable metrics for reliability. Therefore, software is not currently always suitable for integration into safety critical systems, and requires a new, encompassing development procedure that utilizes techniques and metrics to allow it to be used in hardware/software solutions. This dissertation discusses the current state of research and industry practices for software, and proposes an initial solution to developing more trustworthy hardware/software systems. The use of formal methods, new error analysis techniques, and statistical analysis culminates in a complete, comprehensive development procedure for generating reliable software for safety critical systems with metrics similar to hardware. It also establishes a baseline benchmark for programs which future techniques and design improvements can be compared against, allowing for the future adoption of software as a more mainstream component of safety critical embedded systems.
Carla Purdy, Ph.D. (Committee Chair)
Wen-Ben Jone, Ph.D. (Committee Member)
Daniel M. Peairs, Ph.D. (Committee Member)
Ranganadha Vemuri, Ph.D. (Committee Member)
Philip Wilsey, Ph.D. (Committee Member)
320 p.
Computer Engineering
Embedded Systems Software; Reliability Quantification; Software Reliability; Safety Critical Systems; Software Error Modeling; Reliable Software Design

Recommended Citations

Citations

  • Lockhart, J. A. (2019). Software Development Process and Reliability Quantification for Safety Critical Embedded Systems Design [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1562673285477425

    APA Style (7th edition)

  • Lockhart, Jonathan. Software Development Process and Reliability Quantification for Safety Critical Embedded Systems Design. 2019. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin1562673285477425.

    MLA Style (8th edition)

  • Lockhart, Jonathan. "Software Development Process and Reliability Quantification for Safety Critical Embedded Systems Design." Doctoral dissertation, University of Cincinnati, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1562673285477425

    Chicago Manual of Style (17th edition)

ucin1562673285477425
469
© 2019, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.