Concealed information transmission method and reception method based on autoencoder
By using an autoencoder encoding, modulation, and decoding method, and training the encoder and decoder with multiple sets of sample data, the problem of high bit error rate in traditional covert communication is solved, achieving covert communication with high confidentiality and low bit error rate.
CN122179060APending Publication Date: 2026-06-09GUANGZHOU HAIGE COMMUNICATION GROUP INCORPORATED COMPANY
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGZHOU HAIGE COMMUNICATION GROUP INCORPORATED COMPANY
- Filing Date
- 2026-03-30
- Publication Date
- 2026-06-09
AI Technical Summary
Technical Problem
Traditional covert communication methods sacrifice demodulation performance for the sake of covertness, resulting in a high bit error rate.
Method used
An autoencoder is used for encoding, modulation, and decoding. The encoder and decoder are trained with multiple sets of sample data, and the first and second losses are used for joint optimization to achieve covert communication with a high degree of confidentiality.
Benefits of technology
It reduces the bit error rate of covert communication and improves the covertness and demodulation performance of communication.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122179060A_ABST
Abstract
This invention provides a covert information transmission and reception method based on an autoencoder. The transmission method includes: acquiring information to be transmitted; inputting the information to be transmitted into an encoder for encoding and modulation; acquiring the output signal output by the encoder; and transmitting the output signal to a receiver. The reception method includes: receiving a channel transmission signal; inputting the channel transmission signal into a decoder; and acquiring the decoded information output by the decoder. The encoder and decoder are jointly trained based on multiple sets of sample data. Each set of sample data includes sample information to be transmitted and corresponding sample decoded information. The training process includes a first loss and a second loss. The first loss reflects the confidentiality of the sample output signal obtained after encoding and modulating the sample information to be transmitted by the encoder, and the second loss reflects the difference between the sample information to be transmitted and the sample decoded information. This application can reduce the bit error rate of covert communication.
Need to check novelty before this filing date? Find Prior Art