一种隐藏终端影响下蜂窝车联网频谱资源分配方法
By employing multi-agent deep reinforcement learning and binary tree state space decomposition algorithms, the problem of data packet collisions caused by hidden terminal interference in cellular vehicle-to-everything (V2V) networks was solved, achieving highly reliable spectrum resource allocation and improving the reliability and scalability of V2V communication.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV OF SCI & TECH
- Filing Date
- 2025-12-25
- Publication Date
- 2026-07-17
AI Technical Summary
In existing cellular vehicle-to-everything (C-V2X) communication, the probability of data packet collisions caused by hidden terminal interference increases, affecting the scalability of the system. Furthermore, traditional channel resource allocation methods cannot effectively handle the collision probability analysis under different vehicle environments.
A spectrum resource allocation method based on multi-agent deep reinforcement learning is adopted, combined with a binary tree state space decomposition algorithm, to construct a discrete-time Markov chain model. By using a data packet transmission model and a PIR probability model, the training complexity is reduced, and highly reliable V2V communication is achieved.
It significantly reduces packet collisions caused by hidden terminal interference, improves the reliability and scalability of V2V communication, and meets millisecond-level communication latency requirements.
Smart Images
Figure CN121547867B_ABST
Abstract
Citation Information
Patent Citations
Congestion control scheme for jointly regulating physical carrier sense range and minimum competition window in VANET
CN103686843A
Crossing ad hoc node assisted urban vehicle routing protocol method
CN105592138A