Redundancy by Coding Versus Redundancy by Replication for Failure-Tolerant Sequential Circuits

Authors:
Ronald W. Larsen;Irving S. Reed
Affiliations:
U. S. Naval Undersea Research and Development Center, Pasadena, Calif.;Department of Electrical Engineering, University of Southern California, Los Angeles, Calif. 90037.
Venue:
IEEE Transactions on Computers
Year:
1972

Citing 0
Cited 6

Probability of State Transition Errors in a Finite State Machine Containing Soft Failures

IEEE Transactions on Computers
Realization of Fault-Tolerant Machines Linear Code Application

IEEE Transactions on Computers
An Algebraic Model of Fault-Masking Logic Circuits

IEEE Transactions on Computers
Fault-Tolerant Asynchronous Networks

IEEE Transactions on Computers
Realization of fault-tolerant and Fail-Safe sequential machines

IEEE Transactions on Computers
Periodic and non-concurrent error detection and identification in one-hot encoded FSMs

Automatica (Journal of IFAC)

Quantified Score

Hi-index	15.00

Visualization

Abstract

A synthesis procedure for failure-tolerant sequential circuits using error-correcting codes is presented. Coding redundancy is then compared with replication as to circuit complexity and reliability improvement. Using appropriate assumptions, it is shown that for a specified ability to tolerate failures, replication yields better circuit reliability than coding redundancy. When circuit complexity as well as reliability are taken into consideration, it is shown that schemes based on orthogonizable codes generally provide a greater improvement in reliability for a given complexity than replication. As in any redundant scheme, these results presuppose reasonably good reliabilities for irredundant circuits.