A heterogeneous network multi-spectrum coordination method, device and medium
Through a multi-agent multi-spectrum collaborative approach, real-time reinforcement learning and neural networks are used to optimize frequency band resource allocation, solving the user experience issues of different service types under multi-band networking in heterogeneous networks, and achieving efficient multi-type service integration and resource scheduling.
Patent Information
- Application Number
- CN202510109943.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2024-02-04
- Filing Date
- 2025-01-23
- Publication Date
- 2025-10-21
- Estimated Expiration
- 2045-01-23
AI Technical Summary
Existing technologies fail to effectively address the latency, rate, reliability, and power consumption requirements of different service types in multi-band networking in heterogeneous networks, especially the inherent network intelligence requirements of Internet of Vehicles services and immersive communications, resulting in poor user experience.
It adopts a multi-agent multi-spectrum collaborative method, obtains user information and uses neural network units of real-time reinforcement learning, experience replay and target reinforcement learning to optimize frequency band resource allocation and user scheduling, thereby achieving seamless integration of multiple types of services.
It improves the user experience of different types of services in heterogeneous networks, meets various business needs through a lightweight multi-spectrum collaborative approach, reduces algorithm complexity and improves reinforcement learning efficiency.
Smart Images

Figure CN119946884B_ABST
Abstract
Citation Information
Patent Citations
D2D communication network slice allocation method based on deep reinforcement learning
CN113163451A
Wireless positioning device and method based on multi-band CSI cooperation
CN113938823A