TY - GEN
T1 - QoE-Based Server Selection for Mobile Video Streaming
AU - Tapang, Daniel Kanba
AU - Huang, Siqi
AU - Huang, Xueqing
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/11
Y1 - 2020/11
N2 - Mobile devices make up the bulk of clients that stream video content over the internet. Improving one of the most popular services, i.e., mobile video streaming, has the potential to make the most market impact. Video streaming giants like YouTube, Netflix, Hulu, and Amazon video aim to provide the best quality service and expand market share. The problem of selecting the best server is critical for ensuring the qualified experience for video streaming on a mobile device. Traditional server selection strategies use proximity as a server selection rule. Improved strategies select servers by considering more factors that also impact the quality of experience (QoE). Currently, reinforcement learning is being used to maximize QoE when selecting servers. This paper seeks to further develop an RL agent that performs better on mobile devices. The result is an RL agent that quickly learns to select servers that offer the best QoE.
AB - Mobile devices make up the bulk of clients that stream video content over the internet. Improving one of the most popular services, i.e., mobile video streaming, has the potential to make the most market impact. Video streaming giants like YouTube, Netflix, Hulu, and Amazon video aim to provide the best quality service and expand market share. The problem of selecting the best server is critical for ensuring the qualified experience for video streaming on a mobile device. Traditional server selection strategies use proximity as a server selection rule. Improved strategies select servers by considering more factors that also impact the quality of experience (QoE). Currently, reinforcement learning is being used to maximize QoE when selecting servers. This paper seeks to further develop an RL agent that performs better on mobile devices. The result is an RL agent that quickly learns to select servers that offer the best QoE.
KW - Q-Learning
KW - QoE
KW - Reinforcement Learning
KW - Server Selection
KW - Video Streaming
UR - http://www.scopus.com/inward/record.url?scp=85102190490&partnerID=8YFLogxK
U2 - 10.1109/SEC50012.2020.00066
DO - 10.1109/SEC50012.2020.00066
M3 - Conference Proceeding
AN - SCOPUS:85102190490
T3 - Proceedings - 2020 IEEE/ACM Symposium on Edge Computing, SEC 2020
SP - 435
EP - 439
BT - Proceedings - 2020 IEEE/ACM Symposium on Edge Computing, SEC 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 5th IEEE/ACM Symposium on Edge Computing, SEC 2020
Y2 - 11 November 2020 through 13 November 2020
ER -