Abstract
An increasing mass market for cryptographic products leads to greater pressure on companies to fabricate chips which will recover from, and correct, sporadic errors resulting from design and fabrication faults, inadequate testing, smaller technology, ionising radiation, random noise, and so on. Where encryption is subject to such errors, large quanities of data can become totally corrupted or inaccessible unless fault detection is an integral part of the hardware arithmetic. Here realistially cheap methods are examined for checking the correctness of the arithmetic computations which are the basis of the RSA cryptosystem and Die-Hellman key exchange. As in ordinary integer multiplication, a modular residue checker function is used to detect errors and triggerre-computation when necessary. The mechanism will also detect most permanent faults. Some suggestions are made on how to correct infrequent errors without using additional hardware. © Springer-Verlag Berlin Heidelberg 2000.
Author supplied keywords
Cite
CITATION STYLE
Walter, C. D. (2000). Data integrity in hardware for modular arithmetic. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1965 LNCS, pp. 204–215). Springer Verlag. https://doi.org/10.1007/3-540-44499-8_15
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.