2results about How to "Resist attack" patented technology

This invention relates to the fields of federated learning and cryptography, specifically to a secure three-party aggregation method for federated learning in the Industrial Internet of Things (IIoT). The method includes an initialization phase, a training phase, an online phase, and a verification phase. By setting up an architecture with one honest server and two aggregation servers, each server receives a portion of the secret share of local model gradient parameters sent by various IoT devices, thus protecting the privacy of local data. The honest server assists the two aggregation servers in interactively executing a three-party weight calculation protocol to calculate the secret share of the aggregation weights. They also interactively execute a three-party multi-weight aggregation protocol to calculate the secret share of the gradient parameters of the current aggregation model. Furthermore, a linear homomorphic hashing method is used to verify the correctness of the gradient parameters of the current aggregation model. This method solves the problems of low efficiency, weak security, and low robustness in federated learning during model training and aggregation, and can promote data flow and utilization in IoT scenarios.

This invention discloses a secure computation method, apparatus, and chip for information. The method includes: during secure computation, obtaining the dividend and modulus of a modulo operation; determining a transformation constant for converting the dividend to the Montgomery domain based on the modulus, and secretly calculating a Montgomery modular multiplication factor based on the transformation constant and the modulus; dividing the dividend into multiple local dividends based on the transformation constant, and performing a matching Montgomery modular multiplication on each local dividend according to its position within the dividend and the Montgomery modular multiplication factor, obtaining multiple local modulo results; and calculating the modulo operation result between the dividend and the modulus based on the multiple local modulo results. The technical solution of this embodiment is based on an improved Montgomery modular multiplication algorithm, achieving fast and secure modulo computation during secure computation of information, improving the speed of modulo computation while resisting side-channel attacks.