A privacy protection method, system and terminal based on a centerless streaming federated learning

CN114417420BActive Publication Date: 2026-05-29HANGZHOU ROLL CUMULUS TECH CO LTD
2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HANGZHOU ROLL CUMULUS TECH CO LTD
Filing Date
2022-01-25
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing technologies cannot effectively address the communication overhead and privacy protection issues in federated learning of streaming data in decentralized scenarios, especially in real-time data interaction scenarios such as vehicle-to-everything (V2X) networks, where there are risks of communication delays and privacy leaks.

Method used

We employ a decentralized streaming federated learning approach, which randomly initializes model parameters at edge nodes, adds differential privacy noise protection, performs local model updates, and shares and adaptively adjusts parameters in an intermittent interactive manner, thereby reducing communication frequency and privacy budget consumption.

Benefits of technology

It enables privacy-preserving collaborative federated learning for large-scale nodes in a decentralized scenario, reducing the frequency of communication between devices and the consumption of privacy budget, and improving the efficiency and security of data interaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114417420B_ABST
    Figure CN114417420B_ABST
Patent Text Reader

Abstract

The application discloses a privacy protection method and system based on a centerless flow federated learning and a terminal, wherein online learning is performed on local real-time data flow by an edge node, then a communication interaction opportunity between nodes is adaptively determined based on changes of local model parameters, and a privacy protection based on a Laplace mechanism is performed on the model parameters during the communication interaction, and then the model parameters are broadcasted and shared with adjacent nodes, and no parameter transmission is performed during a non-communication interaction time, so as to reduce communication overhead and a privacy budget. Finally, dynamic model training and updating of global data flow are cooperatively performed by the edge node under the premise of privacy protection. The application has good application effects in a privacy protection scene of actual large-scale distributed node cooperative online machine learning, and can be applied to data privacy protection scenes in application scenes such as vehicle networking driving intelligence, mobile socialization and online recommendation.
Need to check novelty before this filing date? Find Prior Art