Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach

Jiaying Yin; Lixin Li; Yang Xu; Wei Liang; Huisheng Zhang; Zhu Han

doi:10.1109/GlobalSIP.2018.8646427

Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach

Jiaying Yin, Lixin Li, Yang Xu, Wei Liang, Huisheng Zhang, Zhu Han

School of Electronics and Information

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

15 Scopus citations

Abstract

Compared with traditional device-to-device (D2D) communication networks, the users in the cache-enabled D2D communication networks can easily obtain their requested contents from the nearby users, and reduce the backhaul costs. In this paper, we investigate the caching strategy for the cache-enabled D2D communication networks, with the consideration of caching placement and caching delivery. The content popularity and user mobility are predicted by a machine learning approach of echo state networks (ESNs) in order to determine which content to cache and where to cache. Furthermore, a deep Q-learning network (DQN) algorithm is proposed to optimize the content delivery problem, with taking the delays and energy consumption into consideration. Simulation results show that the content hit rate and the traffic offloading can be remarkably improved with the proposed approach, compared to the random caching strategy.

Original language	English
Title of host publication	2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	609-613
Number of pages	5
ISBN (Electronic)	9781728112954
DOIs	https://doi.org/10.1109/GlobalSIP.2018.8646427
State	Published - 2 Jul 2018
Event	2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Anaheim, United States Duration: 26 Nov 2018 → 29 Nov 2018

Publication series

Name	2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

Conference

Conference	2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018
Country/Territory	United States
City	Anaheim
Period	26/11/18 → 29/11/18

Keywords

Caching
Deep Q-learning networks
Echo State Networks
Popularity prediction

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/GlobalSIP.2018.8646427

Cite this

Yin, J., Li, L., Xu, Y., Liang, W., Zhang, H., & Han, Z. (2018). Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach. In 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings (pp. 609-613). Article 8646427 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2018.8646427

Yin, Jiaying ; Li, Lixin ; Xu, Yang et al. / Joint content popularity prediction and content delivery policy for cache-enabled D2D networks : A deep reinforcement learning approach. 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 609-613 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings).

@inproceedings{2e39725e4db04b168fb635e7e41dec7f,

title = "Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach",

abstract = "Compared with traditional device-to-device (D2D) communication networks, the users in the cache-enabled D2D communication networks can easily obtain their requested contents from the nearby users, and reduce the backhaul costs. In this paper, we investigate the caching strategy for the cache-enabled D2D communication networks, with the consideration of caching placement and caching delivery. The content popularity and user mobility are predicted by a machine learning approach of echo state networks (ESNs) in order to determine which content to cache and where to cache. Furthermore, a deep Q-learning network (DQN) algorithm is proposed to optimize the content delivery problem, with taking the delays and energy consumption into consideration. Simulation results show that the content hit rate and the traffic offloading can be remarkably improved with the proposed approach, compared to the random caching strategy.",

keywords = "Caching, Deep Q-learning networks, Echo State Networks, Popularity prediction",

author = "Jiaying Yin and Lixin Li and Yang Xu and Wei Liang and Huisheng Zhang and Zhu Han",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 ; Conference date: 26-11-2018 Through 29-11-2018",

year = "2018",

month = jul,

day = "2",

doi = "10.1109/GlobalSIP.2018.8646427",

language = "英语",

series = "2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "609--613",

booktitle = "2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings",

}

Yin, J, Li, L, Xu, Y, Liang, W, Zhang, H & Han, Z 2018, Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach. in 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings., 8646427, 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 609-613, 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018, Anaheim, United States, 26/11/18. https://doi.org/10.1109/GlobalSIP.2018.8646427

Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach. / Yin, Jiaying; Li, Lixin; Xu, Yang et al.
2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. p. 609-613 8646427 (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Joint content popularity prediction and content delivery policy for cache-enabled D2D networks

T2 - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018

AU - Yin, Jiaying

AU - Li, Lixin

AU - Xu, Yang

AU - Liang, Wei

AU - Zhang, Huisheng

AU - Han, Zhu

PY - 2018/7/2

Y1 - 2018/7/2

N2 - Compared with traditional device-to-device (D2D) communication networks, the users in the cache-enabled D2D communication networks can easily obtain their requested contents from the nearby users, and reduce the backhaul costs. In this paper, we investigate the caching strategy for the cache-enabled D2D communication networks, with the consideration of caching placement and caching delivery. The content popularity and user mobility are predicted by a machine learning approach of echo state networks (ESNs) in order to determine which content to cache and where to cache. Furthermore, a deep Q-learning network (DQN) algorithm is proposed to optimize the content delivery problem, with taking the delays and energy consumption into consideration. Simulation results show that the content hit rate and the traffic offloading can be remarkably improved with the proposed approach, compared to the random caching strategy.

AB - Compared with traditional device-to-device (D2D) communication networks, the users in the cache-enabled D2D communication networks can easily obtain their requested contents from the nearby users, and reduce the backhaul costs. In this paper, we investigate the caching strategy for the cache-enabled D2D communication networks, with the consideration of caching placement and caching delivery. The content popularity and user mobility are predicted by a machine learning approach of echo state networks (ESNs) in order to determine which content to cache and where to cache. Furthermore, a deep Q-learning network (DQN) algorithm is proposed to optimize the content delivery problem, with taking the delays and energy consumption into consideration. Simulation results show that the content hit rate and the traffic offloading can be remarkably improved with the proposed approach, compared to the random caching strategy.

KW - Caching

KW - Deep Q-learning networks

KW - Echo State Networks

KW - Popularity prediction

UR - http://www.scopus.com/inward/record.url?scp=85063079624&partnerID=8YFLogxK

U2 - 10.1109/GlobalSIP.2018.8646427

DO - 10.1109/GlobalSIP.2018.8646427

M3 - 会议稿件

AN - SCOPUS:85063079624

T3 - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

SP - 609

EP - 613

BT - 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 26 November 2018 through 29 November 2018

ER -

Yin J, Li L, Xu Y, Liang W, Zhang H, Han Z. Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach. In 2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2018. p. 609-613. 8646427. (2018 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2018 - Proceedings). doi: 10.1109/GlobalSIP.2018.8646427

Joint content popularity prediction and content delivery policy for cache-enabled D2D networks: A deep reinforcement learning approach

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this