Mobile user credible authentication method and device, electronic device and storage medium

Abstract

The invention discloses a mobile user credibility authentication method and device, an electronic device and a storage medium, and the method comprises the steps: constructing a private block chain based on a smart contract, the private block chain being composed of a requester and a mobile device willing to participate in model training, sensing data during man-machine interaction between a user and the mobile equipment being stored in the mobile device; sending a global model to the mobile device, so that after the mobile device receives the global model, training the global model by using the sensing data to obtain a local model, and sending the local model to the requester; receiving the local model, verifying the local model by using a quality proof-based consensus mechanism, and providing rewards for the mobile device generating the effective local model; updating the global model by utilizing a federated learning aggregation algorithm according to the effective local model; and carrying out credible authentication on the user on the mobile device by utilizing the updated global model.

Description

technical field [0001] The present application relates to the technical field of information security, in particular to a mobile user trusted authentication method and device, electronic equipment, and a storage medium. Background technique [0002] With the development of the Industry 4.0 era, artificial intelligence is widely used in the field of energy security. At the level of production equipment, machine learning methods are used to detect abnormal behaviors in the process of energy safety production; at the level of human-computer interaction, machine learning methods are used for credible authentication of employees. [0003] When user authentication is performed in human-computer interaction, most machine learning techniques need to aggregate private data representing authenticated users to a central server for model training and analysis. This will not only lead to high communication and storage costs, but users also face serious risks of privacy leakage. In orde...

AI Technical Summary

Problems solved by technology

[0004] However, in real tasks (i.e., security-related user authentication), the above methods have the following drawbacks: First, the data collection efficiency is low and there is no effective incentive mechanism

Most of the data collection steps in the existing machine learning tasks have been completed in advance, and the number of participating employees is limited, and employees are required to collect new data under each behavior. less consideration, and the rate at which data is collected in this study is largely dependent on how often employees use the devices

Second, the traditional data collection method requires employees to collect data under each behavior and manage it centrally. However, with the federated learning method, it is uncertain whether the data collected by edge mobile devices is true and effective, and there may be mobile The case where the device uses fake data to train the model and returns wrong gradients to achieve a poisoning attack on the model

Images