2017 Fiscal Year Annual Research Report
Oligopoly Mobile Data Offloading Market Analysis and Intelligent Network Selection System
Project/Area Number |
16K18109
|
Research Institution | Waseda University |
Principal Investigator |
張 成 早稲田大学, 理工学術院, 助教 (40755089)
|
Project Period (FY) |
2016-04-01 – 2018-03-31
|
Keywords | mobile data offloading / reinforcement learning / wireless LAN / deep Q-network, DQN |
Outline of Annual Research Achievements |
In this project, we study mobile data offloading problem from both from network side perspective and mobile users’ perspective. For the second year, we study the mobile users' offloading policy to minimize his monetary cost and energy consumption with unknown mobile user mobility pattern. We propose to use a kind of reinforcement learning technique called deep Q-network (DQN) for mobile user to learn the optimal offloading policy from past experiences. Extensive simulations are conducted to validate our proposed offloading algorithms. We have published two journal papers and several international conference papers like International Conference on Network and Service Management and domestic conferences papers.
|
Research Products
(7 results)