物联网低轨卫星通信数据交互生成方法及系统
By constructing a state feature field and a collaborative deep reinforcement learning framework, the routing configuration of the low-Earth orbit satellite network is dynamically adjusted, solving the data transmission problem of the low-Earth orbit satellite network under dynamic topology changes, and realizing the stability and real-time performance of marine environmental monitoring data.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUAXIN ZHENGNENG GRP CO LTD
- Filing Date
- 2026-03-31
- Publication Date
- 2026-07-17
AI Technical Summary
The routing methods of low-Earth orbit satellite networks are difficult to adjust quickly when the dynamic topology changes, which leads to data transmission interruptions, increased latency, or decreased transmission efficiency. Furthermore, centralized routing decisions increase signaling overhead and are difficult to meet the real-time requirements of marine environmental monitoring data.
By collecting network state information from satellite nodes, a state feature field is constructed, feature subdomains are divided, aggregated information blocks are generated, dynamic adjustment coefficients are extracted, a collaborative deep reinforcement learning framework is built, dynamic group partitioning and routing decisions are performed, dynamic topology awareness and quantitative representation are achieved, and routing configurations are quickly adjusted.
It enables precise perception and quantitative characterization of dynamic topology changes in low-Earth orbit satellite networks, ensuring the continuity and stability of marine environmental monitoring data, reducing signaling overhead, and improving the real-time performance and efficiency of data transmission.
Smart Images
Figure CN121966693B_ABST