Covert communication method and device

Abstract

The invention provides a covert communication method and device, and the method comprises the steps: obtaining a to-be-exchanged model parameter between a client participating in federated learning and a server, embedding covert data into the to-be-exchanged model parameter between the client participating in federated learning and the server, and obtaining a target model parameter, so that hidden data are hidden in to-be-interacted model parameters, it is guaranteed that the hidden data are not likely to be perceived; on this basis, sending the target model parameters to the client or the server, so the effective hidden communication between the client and the server is guaranteed while it is guaranteed that the source model is not affected.

Description

technical field [0001] The present application relates to the technical field of communication, and in particular to a covert communication method and device. Background technique [0002] With the advent of the era of big data, data has become an important resource, and the exchange, sharing and full utilization of data is the core goal of realizing the value of big data and improving the efficiency of big data. Therefore, the demand for data exchange and sharing between different domains is becoming more and more urgent. However, there are data barriers that are difficult to break between different domains. With the introduction of the Internet of Things, it is difficult for data to be directly shared and exchanged across domains, resulting in the current big data becoming more and more "data islands" in a sense, unable to truly play the value of big data analysis and mining. [0003] In order to solve the above problems, federated learning, as a new generation of machine...

AI Technical Summary

Problems solved by technology

[0004] However, in the federated learning scenario that only allows the exchange of model parameter information, how to achieve covert communication becomes a problem

Images