一种基于深度学习的移动视频QoE跨层优化方法及系统
By predicting QoE quantization metrics using deep learning-based methods, the application or network layers in mobile video transmission are dynamically optimized, solving the problems of limited optimization gain and conflicts in existing technologies, and achieving efficient QoE optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INST OF COMPUTING TECH CHINESE ACAD OF SCI
- Filing Date
- 2023-02-21
- Publication Date
- 2026-07-17
AI Technical Summary
The lack of awareness of different network layers in current mobile video transmission leads to limited optimization gains, optimization conflicts and wasted computing resources, and the inability to accurately determine the network layers that need optimization results in optimization lag.
A deep learning-based approach is adopted to predict the QoE quantization index through a pre-trained neural network, dynamically determine the optimization strategy of the application layer or network layer, selectively optimize the network layer or application layer to improve QoE, and utilize wireless bandwidth resource allocation algorithm and code rate adaptive algorithm for optimization.
It improves QoE optimization efficiency, avoids optimization limitations and conflicts, saves computing resources, and ensures the accuracy and efficiency of QoE.
Smart Images
Figure CN116208984B_ABST