Fully homomorphic data packaging method for lightweight single-user multi-data

Abstract

The invention discloses a fully homomorphic data packaging method for lightweight single-user multi-data, which is characterized by comprising the following five stages of system initialization, key generation, data encryption, encryption data outsourcing calculation and decryption. Compared with the prior art, the method is simple and easy to implement, the efficiency is high, and efficient privacy protection outsourcing calculation is realized in a scene of a single user and a plurality of input data; according to the method, a cloud server and an unauthorized receiver are combined, additionand multiplication operations on a ciphertext domain are supported at the same time, input privacy can achieve unconditional security (information theory security) under collusion attacks of the cloud server and the unauthorized receiver, calculation result privacy can achieve adaptive ciphertext security selection (CCA2 security) for the unauthorized receiver, and the requirements for high efficiency and privacy are met.

Description

technical field [0001] The invention relates to the technical field of data encapsulation, in particular to a lightweight single-user multi-data fully homomorphic data encapsulation method based on an arbitrary one-way trapdoor permutation structure. Background technique [0002] With the development of computers and the Internet and the increasing popularity of big data and cloud computing services, massive data is generated. Data owners with limited storage and computing resources often need to store large quantities of data locally and perform highly complex computing tasks. Outsourced to resource-rich cloud servers to complete, therefore, efficient secure outsourced computing has become one of the research hotspots in the field of cryptography and security. At the same time, cloud servers usually work in a semi-trusted or malicious environment. Although a semi-trusted cloud server can honestly implement the agreement to obtain correct calculation results, it still attemp...

AI Technical Summary

Problems solved by technology

However, although public-key fully homomorphic encryption can functionally realize secure outsourcing calculations in the ciphertext domain, its algorithm complexity is too high. The complexity of the user's local public-key encryption algorithm is O(n), where n is number of input data

Therefore, its huge computing overhead and communication overhead cannot meet the objective performance requirements of local users with limited resources.

Images