On a class of arithmetic codes and a decoding algorithm (Corresp.)

DAVID M. MANDELBAUM, MEMBER, IEEE termc and M RlCmnd M factor the product of n + r t distinct moduli. The smallest product of Abstract-Multiple-error-correcting arithmetic codes which are min (n + r t) = 12 distinct moduli is ny=r mi = M,. Therenonlinear are constructed by residue encoding. A simple algorithm is fore, M, I M,B/C mod M, < M,. Hence M,.B/C mod M, can given for correcting multiple errors which makes use of continued be written as M,B’IC, where B’ q B mod C. Then max (X’) = fractions. M,, 1 + M,B’/C = M,.(I/R + B//C) 1. Now 1 B’/C 2 l/C 2 l/R so that max (X’) = M,(l/R + B//C) 1 5 M, 1.