.png){kind=icon}
返回
An Online Buffer-Aware Resource Allocation Algorithm for Multiuser Mobile Video Streaming
.png){kind=icon}
PRE
AI
DOI：10.1109/TVT.2020.2966701.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Mobile video traffic has experienced exponential growth in recent years due to increasing prevalence of mobile devices and continuous advancement of wireless communications. However, network fluctuations often occur when transmitting video data, which greatly impact the quality of experience (QoE) of mobile users. Buffering technique has been widely adopted to mitigate the effect of network fluctuations on QoE for mobile users by prefetching certain video data to mobile devices in advance. However, mobile users often abort video watching before the videos are fully watched and such early departure behavior can result in considerable wastage of buffered video data. In this paper, we study buffer-aware resource allocation for multiuser mobile video streaming system by jointly considering video users early departure behavior and wireless channel fluctuation property. We formulate a stochastic optimization problem in which the objective is to reduce both the amount of wasted buffered data and the video re-buffering time. The problem is then transformed into a series of deterministic per-slot problems by using Lyapunov optimization theory. We propose an online buffer-aware resource allocation algorithm to solve the per-slot optimization problems. After proving the convexity of the per-slot optimization problem, we further design a fast-convergent iterative algorithm based on Alternating Direction Method of Multipliers (ADMM) to perform each of the per-slot resource allocations. Simulation results show that our proposed algorithm has very low running time, near optimal performance, and is able to achieve a good trade-off between QoE improvement and data wastage reduction.
Keyword:
Mobile video streaming
data wastage
resource allocation
Lyapunov optimization
alternating direction method of multipliers (ADMM)
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
7.1
论文数:
1.8W
被引数:
6.6W
机构
引用论文
Smart Resource Allocation for Mobile Edge Computing: A Deep Reinforcement Learning Approach移动边缘计算的智能资源分配: 一种深度强化学习方法
.png){kind=icon}
PREAI
分享收藏Buffer-Aware Streaming in Small-Scale Wireless Networks: A Deep Reinforcement Learning Approach小规模无线网络中的缓冲区感知流: 一种深度强化学习方法
PREAI
分享收藏分享收藏分享收藏分享收藏
查看更多内容