DeepCC: Multi-Agent Deep Reinforcement Learning Congestion Control for Multi-Path TCP Based on Self-Attention.
Bo HeJing-Yu WangQi QiHaifeng SunJianxin LiaoChunning DuXiang-Yang Alex LiuZhu HanPublished in: IEEE Trans. Netw. Serv. Manag. (2021)
Keyphrases
- congestion control
- multipath
- end to end
- reinforcement learning
- multi agent
- wireless ad hoc networks
- mobile ad hoc networks
- fading channels
- ad hoc networks
- wireless channels
- window size
- multi hop
- bit error rate
- real time
- multiple description coding
- network bandwidth
- routing algorithm
- application layer
- motion estimation
- scalable video
- high resolution