Pipelined Montgomery Modular Multiplication Algorithm

Abstract

The invention discloses a pipelined Montgomery modular multiplication operation method and a pipelined Montgomery modular multiplication operation device, relates to the technical field of data cryptography algorithms, adopts a pipelined mode to improve Montgomery modular multiplication algorithm performance, increases throughput of a single Montgomery modular multiplier, and consumes less hardware resources and area than a traditional method using a plurality of Montgomery modular multipliers under the condition of the same throughput. Therefore, the modular multiplication number in unit timeis increased by dozens of times, and the Montgomery modular multiplier has higher performance / resource ratio than a non-pipelined Montgomery modular multiplier. Meanwhile, the asymmetric key algorithm hardware using the pipelined Montgomery modular multiplier can achieve higher performance with fewer hardware resources, and the throughput of Montgomery modular multiplication operation of any length is improved.

Description

technical field [0001] The invention relates to the technical field of data encryption algorithms, in particular to a Montgomery modular multiplication operation method and a calculation device. Background technique [0002] Information security is based on security algorithms, and one type of security algorithms is asymmetric key algorithms. The advantage of the asymmetric key algorithm is high security, but the disadvantage is that the encryption speed is much slower than the block cipher, so people have been studying how to improve the operation speed of the asymmetric key algorithm. At present, there are mainly two types of asymmetric key algorithms, one is RSA, and the other is Elliptic Curve Cryptography (ECC). [0003] Among the modular multiplication algorithms, the most efficient and easy-to-implement algorithm is the Montgomery modular multiplication algorithm. During the use of Montgomery modular multiplication, the ordinary number A needs to be converted into t...

AI Technical Summary

Problems solved by technology

However, most of them save resources and area by sacrificing performance, resulting in a decrease in system cost performance and an increase in energy consumption.

A single Montgomery modular multiplication operation takes too long, so if a single Montgomery modular multiplier does not support pipeline, it is difficult to increase the number of modular multiplication operations per unit time, that is, its throughput

A feasible performance improvement solution is to use multiple Montgomery modular multipliers in a computing device, but this will cause a multiplication of hardware resources, area and energy consumption

Images