Fault tolerance in RNS: an efficient approach
Damu Radhakrishnan, TAEJIN PYON · 2002
Minimization of the error-correction hardware for single fault tolerance in residue number systems (RNSs) is proposed. A novel approach for the design of an error calculator for single fault tolerance in RNS arithmetic is presented. It corrects the error concurrently during normal operation. Furthermore, the computation of each error is achieved in-parallel by a number of smaller look-up tables. This allows a considerable reduction in the total error calculator hardware.>