A method for geometric representation of spatial terahertz communication network link space

By using terahertz channel models and manifold learning dimensionality reduction techniques, a geometric representation method for the link space of spatial terahertz communication networks is constructed, which solves the problems of insufficient network representation and optimization, and improves network performance and reliability.

CN122419632APending Publication Date: 2026-07-17THE 32008TH UNIT OF THE PEOPLES LIBERATION ARMY OF CHINA
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
THE 32008TH UNIT OF THE PEOPLES LIBERATION ARMY OF CHINA
Filing Date
2026-03-16
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing satellite communication systems are unable to effectively represent and optimize the space of space terahertz communication network links when facing emerging applications such as large-scale Internet of Things, high-definition video transmission, and real-time telemedicine, resulting in insufficient network capacity, response latency, and transmission rate.

Method used

By introducing a channel model unique to terahertz and manifold learning dimensionality reduction techniques, a geometric representation method for the link space is constructed, including defining the link space, extracting geometric feature vectors, performing manifold learning dimensionality reduction and iterative updates, thereby optimizing network resource scheduling and interference coordination.

Benefits of technology

It achieves efficient representation and optimization of link space, simplifies network monitoring and fault diagnosis, improves network capacity, transmission rate and link reliability, and provides a high-fidelity data foundation for dynamically adjusting network topology.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122419632A_ABST
    Figure CN122419632A_ABST
Patent Text Reader

Abstract

This invention discloses a geometric representation method for the link space of a space-based terahertz communication network. A multi-layered, three-dimensional heterogeneous network model is constructed, including nodes in space, near space, air, and ground, and terahertz frequency band links are configured. Potential communication links in the network are defined as the link space and categorized into a metric space. The link length, spatial angle, and terahertz link quality factor incorporating atmospheric attenuation, molecular absorption, and rain attenuation effects are extracted to construct a high-dimensional feature vector. Based on a composite distance metric, a manifold learning algorithm is used to perform nonlinear dimensionality reduction and topological embedding on the high-dimensional feature space to obtain a low-dimensional geometric representation of the link space. Based on this geometric representation, routing and resource scheduling problems are transformed into geometric computation problems to perform network optimization. The geometric representation is iteratively updated according to changes in node mobility or channel environment. This invention achieves geometric representation of the link space of complex three-dimensional heterogeneous terahertz networks, improving network planning efficiency, link reliability, and resource utilization.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Intranet information cross-network transfer method and device and electronic equipment

    CN113381908A

  • DMRS configuration method, electronic equipment and storage medium

    CN114726491A