Accurate characterization of error propagation in a highly parallel architecture

Venugopal R. Shamapant; Jacob A. Abraham; D. G. Saab

doi:10.1117/12.23505

1 November 1990 Accurate characterization of error propagation in a highly parallel architecture

Venugopal R. Shamapant, Jacob A. Abraham, D. G. Saab

Proceedings Volume 1348, Advanced Signal Processing Algorithms, Architectures, and Implementations; (1990) https://doi.org/10.1117/12.23505
Event: 34th Annual International Technical Symposium on Optical and Optoelectronic Applied Science and Engineering, 1990, San Diego, CA, United States

Abstract

Errors caused by faults in a system manifest themselves in the output results in various patterns. These output patterns of erroneous data the nature of their propagation through the various modules of the system and their impact on the system are collectively referred to as error characerisics. This paper deals with efficient methods of obtaining these error characteristics. Both experimental and analytical approaches to the problem are considered and a probabilistic approach to this problem is proposed. The effectiveness of this approach with respect to the accuracy of results and amount of computation involved is discussed.

Citation Download Citation

Venugopal R. Shamapant, Jacob A. Abraham, and D. G. Saab "Accurate characterization of error propagation in a highly parallel architecture", Proc. SPIE 1348, Advanced Signal Processing Algorithms, Architectures, and Implementations, (1 November 1990); https://doi.org/10.1117/12.23505

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available