A note on RNS architectures for the implementation of the diagonal function

This paper focuses on some considerations on the diagonal function and its applications to implement non-modular operations like magnitude comparison and sign detection in residue number system (RNS), recently proposed in the literature. According to our results, any implementation involving the diagonal function proposed to date results in excessive hardware overhead and delay, which make it impractical from the application point of view, so that it cannot compete with more traditional approaches. We consider magnitude comparison implementation in residue number system (RNS).Performance of hardware implementations using diagonal function is evaluated.More realistic delay parameters of the basic building blocks is assumed.Comparator based on the Chinese remainder theorem (CRT) is still the fastest.