语义通信模型的训练方法和装置
By introducing a channel module into the semantic communication model to simulate a multi-user wireless random access scenario and optimizing encoding and decoding processes, the problem of low communication efficiency in a multi-user environment is solved, achieving more efficient semantic information transmission and adapting to the communication needs of the 6G era.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENZHEN UNIV
- Filing Date
- 2023-12-20
- Publication Date
- 2026-07-17
AI Technical Summary
Existing semantic communication models mainly focus on point-to-point models with a single sender and a single receiver, which do not fully consider communication scenarios with multiple users and random wireless access. This results in low communication efficiency and susceptibility to collision interference in multi-user environments.
A channel module is introduced, including a collision channel simulation layer and a noise simulation layer, to simulate a multi-user wireless random access communication scenario. Encoding and decoding are performed through a semantic encoder and decoder to optimize the basic network model and improve its adaptability and reliability.
It improves the performance of the semantic communication model in multi-user wireless random access scenarios, reduces user data packet collision interference, enhances communication efficiency and reliability, and promotes the practical application of the semantic communication model in the 6G era.
Smart Images
Figure CN117896739B_ABST