A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions

Hai Lin; Bo Zhang; Yangjie Cao; Zhi Liu; Xianfu Chen

doi:10.1109/VTC2021-Fall52928.2021.9625496

A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions

Hai Lin, Bo Zhang, Yangjie Cao, Zhi Liu, Xianfu Chen

BA6401 Connectivity solutions

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

4 Citations (Scopus)

Abstract

The point cloud videos, thanks to the multi-view and immersive experiences, have recently attracted notable attentions from both academia and industry. Due to the high data volume, a point cloud video also raises the challenge of quality-of-experience (QoE), which is in terms of the balance between playback quality and buffering delay during the transmission under time-varying system conditions. In this paper, we propose a deep reinforcement learning (DRL) approach to optimize the expected long-term QoE for the client. Over the time horizon, the proposed approach learns to select the tiles of the corresponding video for transmissions in an iterative way. Under various settings, numerical experiments based on real throughput data traces are conducted to evaluate the proposed approach. Compared to the baselines, our approach not only enhances the video quality but also reduces the re-buffering time, obtaining an improvement of average QoE for the client by 9%–14%.

Original language	English
Title of host publication	2021 IEEE 94th Vehicular Technology Conference, VTC 2021-Fall - Proceedings
Publisher	IEEE Institute of Electrical and Electronic Engineers
Pages	1-5
ISBN (Electronic)	978-1-6654-1368-8
DOIs	https://doi.org/10.1109/VTC2021-Fall52928.2021.9625496
Publication status	Published - Sept 2021
MoE publication type	A4 Article in a conference publication
Event	IEEE 94th Vehicular Technology Conference, VTC2021-Fall - Norman, OK, USA Duration: 27 Sept 2021 → 30 Sept 2021

Conference

Conference	IEEE 94th Vehicular Technology Conference, VTC2021-Fall
Period	27/09/21 → 30/09/21

Access to Document

10.1109/VTC2021-Fall52928.2021.9625496

Cite this

@inproceedings{83ce739c468440a5a56e3aa0ba81c60b,

title = "A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions",

abstract = "The point cloud videos, thanks to the multi-view and immersive experiences, have recently attracted notable attentions from both academia and industry. Due to the high data volume, a point cloud video also raises the challenge of quality-of-experience (QoE), which is in terms of the balance between playback quality and buffering delay during the transmission under time-varying system conditions. In this paper, we propose a deep reinforcement learning (DRL) approach to optimize the expected long-term QoE for the client. Over the time horizon, the proposed approach learns to select the tiles of the corresponding video for transmissions in an iterative way. Under various settings, numerical experiments based on real throughput data traces are conducted to evaluate the proposed approach. Compared to the baselines, our approach not only enhances the video quality but also reduces the re-buffering time, obtaining an improvement of average QoE for the client by 9%–14%.",

author = "Hai Lin and Bo Zhang and Yangjie Cao and Zhi Liu and Xianfu Chen",

note = "Funding Information: H. Lin, B. Zhang and Y. Cao are with Zhengzhou University, China. Z. Liu is with University of Electro-Communications, Japan. X. Chen is with VTT Technical Research Centre of Finland, Finland. This work was supported in part by the Collaborative Innovation Major Project of Zhengzhou (20XTZX06013) and in part by the National Natural Science Foundation of China (61972092). Publisher Copyright: {\textcopyright} 2021 IEEE.; IEEE 94th Vehicular Technology Conference, VTC2021-Fall ; Conference date: 27-09-2021 Through 30-09-2021",

year = "2021",

month = sep,

doi = "10.1109/VTC2021-Fall52928.2021.9625496",

language = "English",

pages = "1--5",

booktitle = "2021 IEEE 94th Vehicular Technology Conference, VTC 2021-Fall - Proceedings",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

address = "United States",

}

Lin, H, Zhang, B, Cao, Y, Liu, Z & Chen, X 2021, A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions. in 2021 IEEE 94th Vehicular Technology Conference, VTC 2021-Fall - Proceedings. IEEE Institute of Electrical and Electronic Engineers, pp. 1-5, IEEE 94th Vehicular Technology Conference, VTC2021-Fall, 27/09/21. https://doi.org/10.1109/VTC2021-Fall52928.2021.9625496

A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions. / Lin, Hai; Zhang, Bo; Cao, Yangjie et al.
2021 IEEE 94th Vehicular Technology Conference, VTC 2021-Fall - Proceedings. IEEE Institute of Electrical and Electronic Engineers, 2021. p. 1-5.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions

AU - Lin, Hai

AU - Zhang, Bo

AU - Cao, Yangjie

AU - Liu, Zhi

AU - Chen, Xianfu

N1 - Funding Information: H. Lin, B. Zhang and Y. Cao are with Zhengzhou University, China. Z. Liu is with University of Electro-Communications, Japan. X. Chen is with VTT Technical Research Centre of Finland, Finland. This work was supported in part by the Collaborative Innovation Major Project of Zhengzhou (20XTZX06013) and in part by the National Natural Science Foundation of China (61972092). Publisher Copyright: © 2021 IEEE.

PY - 2021/9

Y1 - 2021/9

N2 - The point cloud videos, thanks to the multi-view and immersive experiences, have recently attracted notable attentions from both academia and industry. Due to the high data volume, a point cloud video also raises the challenge of quality-of-experience (QoE), which is in terms of the balance between playback quality and buffering delay during the transmission under time-varying system conditions. In this paper, we propose a deep reinforcement learning (DRL) approach to optimize the expected long-term QoE for the client. Over the time horizon, the proposed approach learns to select the tiles of the corresponding video for transmissions in an iterative way. Under various settings, numerical experiments based on real throughput data traces are conducted to evaluate the proposed approach. Compared to the baselines, our approach not only enhances the video quality but also reduces the re-buffering time, obtaining an improvement of average QoE for the client by 9%–14%.

AB - The point cloud videos, thanks to the multi-view and immersive experiences, have recently attracted notable attentions from both academia and industry. Due to the high data volume, a point cloud video also raises the challenge of quality-of-experience (QoE), which is in terms of the balance between playback quality and buffering delay during the transmission under time-varying system conditions. In this paper, we propose a deep reinforcement learning (DRL) approach to optimize the expected long-term QoE for the client. Over the time horizon, the proposed approach learns to select the tiles of the corresponding video for transmissions in an iterative way. Under various settings, numerical experiments based on real throughput data traces are conducted to evaluate the proposed approach. Compared to the baselines, our approach not only enhances the video quality but also reduces the re-buffering time, obtaining an improvement of average QoE for the client by 9%–14%.

UR - http://www.scopus.com/inward/record.url?scp=85122990531&partnerID=8YFLogxK

U2 - 10.1109/VTC2021-Fall52928.2021.9625496

DO - 10.1109/VTC2021-Fall52928.2021.9625496

M3 - Conference article in proceedings

SP - 1

EP - 5

BT - 2021 IEEE 94th Vehicular Technology Conference, VTC 2021-Fall - Proceedings

PB - IEEE Institute of Electrical and Electronic Engineers

T2 - IEEE 94th Vehicular Technology Conference, VTC2021-Fall

Y2 - 27 September 2021 through 30 September 2021

ER -

A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions

Abstract

Conference

Access to Document

Other files and links

Fingerprint

MISSION: Mission-Critical Internet of Things Applications over Fog Networks

Cite this

A Deep Reinforcement Learning Approach for Point Cloud Video Transmissions

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Projects

MISSION: Mission-Critical Internet of Things Applications over Fog Networks

Cite this