A method for embedding a service function chain of fusion features

By combining feature aggregation and PPO model, the virtual network embedding of satellite nodes is optimized, which solves the problems of transmission latency and service blockage in the integrated air-space-ground network and achieves more efficient network resource utilization and service delivery.

CN120050676BActive Publication Date: 2026-06-16BEIJING INST OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
BEIJING INST OF TECH
Filing Date
2025-01-15
Publication Date
2026-06-16

AI Technical Summary

⚠Technical Problem

Existing methods for embedding business function chains cannot adapt to dynamic network environments, resulting in the inability to fully utilize network resources and limiting network flexibility and service delivery efficiency. In particular, transmission latency and service congestion are prominent issues in integrated air-space-ground networks.

⚗Method used

A method combining feature aggregation and PPO model is adopted to encode the neighbor link information of satellite nodes into feature vectors. The optimal embedding method is learned step by step through PPO model, and the routing is selected by combining Dijkstra's algorithm to optimize the embedding of virtual nodes in physical network. The embedding process is optimized by using policy gradient update and gradient backpropagation of PPO model.

🎯Benefits of technology

In dynamic network environments, it significantly reduces end-to-end transmission latency, improves the success rate of embedding service function chains and the efficiency of network resource utilization, and achieves more efficient service delivery.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120050676B_ABST
    Figure CN120050676B_ABST
Patent Text Reader

Abstract

The application discloses a service function chain embedding method fusing features, and comprises the following steps: step S1, feature aggregation; step S2, embedding a virtual node into a physical node to be embedded; step S3, route selection; and step S4, PPO model updating. Through the application, the end-to-end transmission delay in the service function chain embedding process caused by satellite movement can be optimized and reduced in a dynamic network environment.
Need to check novelty before this filing date? Find Prior Art