NOVEL ELLIPTIC CURVE SCALAR MULTIPLICATION ALGORITHMS FOR FASTER AND SAFER PUBLIC-KEY CRYPTOSYSTEMS
NOVEL ELLIPTIC CURVE SCALAR
MULTIPLICATION ALGORITHMS FOR FASTER AND SAFER PUBLIC-KEY CRYPTOSYSTEMS
Vinay S. Iyengar
Oregon Episcopal School,
Portland, Oregon, USA
ABSTRACT
Efficient and secure public-key
cryptosystems are essential in today’s age of rapidly growing Internet
communications. Elliptic curve scalar multiplication in particular, which
refers to the operation of multiplying a large integer by a point on an elliptic
curve, is crucial for both data encryption technology as well as testing the
security of cryptographic systems. The purpose of this project was to design
and implement an elliptic curve scalar multiplication algorithm 10% faster than
one of the best algorithms currently used: the binary double-add algorithm. The
algorithm designed was based off of an operation that allowed for the tripling
of a point and was derived from point doubling and addition operations. The
algorithm was further optimized using an integer representation in both ternary
and balanced ternary form. Finally, a novel modification was made that allowed
for the pre-computation and storage of multiples of a point. The algorithms
were written and tested in a C++ program that selected random points over
elliptic curves in both Weierstrass and Edwards form, and multiplied them by a
large range of integers using each of the various algorithms. Timings for these
operations were taken in terms of processor cycles and finally averaged out.
The final averaged results showed that the most optimized algorithm was 66.63%
more efficient that the binary double-add algorithm. In conclusion, the
algorithm designed in this research showed a large improvement in efficiency
compared to one of the best algorithms in commercial use today, and has
significant implications for the field of cryptography.
KEYWORDS
Public-key cryptography, elliptic
curves, scalar multiplication, point tripling, storage ladder.
SOURCE URL
VOLUME URL
Comments
Post a Comment