Homomorphic multiplication hardware computing system and computing method based on remainder system

Abstract

The invention provides a homomorphic multiplication hardware computing system based on a remainder system. The system includes: a controller which controls the whole operation process; a read-write address generation module which generates a stored read-write address; a data storage module which is used for storing ciphertext source data and pre-calculation parameters; a ciphertext base expansion module which realizes mutual expansion of polynomial coefficients of a ciphertext between a remainder representation form of a basic base and a remainder representation form of an expansion base; a para-position multiplication module and a ciphertext scaling module which realize para-position multiplication of the ciphertext in the NTT domain, and obtain a three-item homomorphic multiplication result through scaling and base expansion; a relinearization module which is used for reducing three results into two results; and an output module which outputs the homomorphic product ciphertext. The invention aims to overcome the difficulties of high operation complexity, low operation speed and the like caused by overlarge modulus in homomorphic operation, and realizes homomorphic multiplication with wider range, deeper depth and higher speed by utilizing the unweighted property and parallelism of a remainder system.

Description

technical field [0001] The invention relates to the field of homomorphic encryption computing, in particular to a homomorphic multiplication hardware computing system and computing method based on a remainder system. Background technique [0002] In the context of the information society, the issue of privacy protection has attracted much attention. For security reasons, users need to encrypt private data when they outsource data to other services. Fully homomorphic encryption technology refers to the encryption technology that the ciphertext satisfies the property of homomorphic operation, that is, the result of the multiplication and addition operation on the ciphertext after decryption is consistent with the result obtained by directly performing the same operation on the plaintext. Fully homomorphic encryption can guarantee the security of private data and the effectiveness of encrypted data processing, and realize the "usable but invisible" data, so it is called the "h...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problems of high complexity and low operation efficiency of homomorphic multiplication in the prior art, the present invention proposes a method for deploying homomorphic multiplication calculations on hardware, using the remainder system to decompose large integers, and using parallel structures to accelerate operations to improve the speed and efficiency of homomorphic operations

Images