一种基于信道建模驱动和几何整形的端到端损伤抑制光子太赫兹通信系统
The end-to-end damage suppression photonic terahertz communication system driven by channel modeling and geometric shaping utilizes a conditional denoising diffusion model and an autoencoder neural network to achieve high-precision modeling and suppression of channel damage. This solves the problem of difficult compensation for signal damage coupling in terahertz communication systems and improves signal transmission quality and stability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING UNIV OF POSTS & TELECOMM
- Filing Date
- 2026-01-16
- Publication Date
- 2026-07-17
AI Technical Summary
In existing terahertz communication systems, linear and nonlinear signal impairments are coupled together and are difficult to compensate accurately on their own. Traditional methods are highly complex, and impairment suppression techniques based on end-to-end learning have limitations in generalization ability.
An end-to-end impairment suppression photonic terahertz communication system based on channel modeling and geometric shaping is adopted. A high-fidelity, highly generalizable symbol-level differentiable channel model is performed using a conditional denoising diffusion model, and an autoencoder neural network is embedded for overall training to achieve joint sensing, modeling and suppression of channel impairment.
It significantly improves signal transmission quality and stability, optimizes the performance of photonic terahertz communication systems, and enhances system bit error rate performance and generalized mutual information.
Smart Images
Figure CN121750111B_ABST