From Euclid's GCD to Montgomery Multiplication to the Great Divide

  • Authors:
  • Sheueling Chang Shantz

  • Affiliations:
  • -

  • Venue:
  • From Euclid's GCD to Montgomery Multiplication to the Great Divide
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

Euclid's method for finding the greatest common divisor(GCD) of two integers was first described around the year 300 B.C.This simple iterative method is often regarded as the grandfatherof all algorithms in Number Theory today. Many advances have beenmade since then--for example, Berlekamp's algorithm formultiplicative inverse and Montgomery's technique formodular multiplication. These binary add-and-shiftalgorithms for efficient finite field arithmetic operations haveplayed important roles in today s public-key cryptographic systems.Yet, two thousand three hundred years after Euclid's GCD, onealgorithm remained missing--division. For many decades we did nottackle modular division problems directly. Instead, we relied onthe Extended Euclidean algorithm for calculating inversion and wecomputed division in a two-step process--inversion followed bymultiplication. This practice is so deeply rooted in our teachingsand doings today that we have neglected to ask whether the ideaunderlying the binary Extended Euclidean algorithm can also beapplied to finding a general solution for field division. Thispaper describes such a solution: a binary add-and-shift algorithmfor modular division in a residue class. This technique forfast computation of divisions in GF(2m) is thekey to a highly efficient implementation of elliptic curvecryptosystems.