.png){kind=icon}
.png){kind=icon}
Return
Optimization for Mobile Streaming Media Based on Deep Q-learning
.png){kind=icon}
PRE
AI
DOI：10.1109/CBD.2019.00058.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Mobile streaming application is one of the pervasive mobile applications that billions of mobile users perform on a daily basis, and it consumes more than half of the mobile internet traffic. At the same time, battery life is a primary concern to many mobile users, while the current underlying architecture does not fully understand the characteristics of mobile streaming applications, so that leads to huge energy consumption. This paper proposes a DQN (Deep Q-learning) based energy optimization model for mobile streaming applications on heterogeneous mobile platforms. This model takes the current network state, video buffer status, and device power into consideration and uses the reinforcement learning optimal decision to schedule the data loading process with the optimal CPU configuration. The results show that our approach reduces energy consumption with an average of 14% when compared with the system default scheduler Interactive.
Keywords:
Mobile streaming media
heterogeneous multi-core platform
Deep Q-learning
low energy consumption
resource scheduling
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
S
IF:
0
Papers:
1
Citations:
0