The Journal of China Universities of Posts and Telecommunications ›› 2021, Vol. 28 ›› Issue (4): 29-38.doi: 10.19682/j.cnki.1005-8885.2021.2003

• Artificial intelligence • Previous Articles     Next Articles

QoE-based video segments caching strategy in urban public transportation system

Wang Hang, Li Xi, Ji Hong, Zhang Heli   

  1. School of Telecommunication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2020-10-29 Revised:2021-06-06 Accepted:2021-07-29 Online:2021-08-31 Published:2021-10-11
  • Contact: Corresponding author: Li Xi, E-mail:
  • Supported by:
    This work was supported by the National Natural Science Foundation of China (61771070).

Abstract: With the rapid development of vehicle-based applications, entertainment videos have gained popularity for passengers on public vehicles. Therefore, how to provide high quality video service for passengers in typical public transportation scenarios is an essential problem. This paper proposes a quality of experience (QoE)-based video segments caching (QoE-VSC) strategy to guarantee the smooth watching experience of passengers. Consequently, this paper considers a jointly caching scenario where the bus provides the beginning segments of a video, and the road side unit (RSU) offers the remaining for passengers. To evaluate the effectiveness, QoE hit ratio is defined to represent the probability that the bus and RSUs jointly provide passengers with desirable video segments successfully. Furthermore, since passenger volume change will lead to different video preferences, a deep reinforcement learning (DRL) network is trained to generate the segment replacing policy on the video segments cached by the bus server. And the training target of DRL is to maximize the QoE hit ratio, thus enabling more passengers to get the required video. The simulation results prove that the proposed method has a better performance than baseline methods in terms of QoE hit ratio and cache costs.

Key words: caching strategy, quality of experience, deep reinforcement learning, urban public transportation system

CLC Number: