2017 Fiscal Year Annual Research Report
Study of Joint Optimization of Quality of Service/Experience and Security for Differentiated Services in 5G Heterogeneous Networks
| Project/Area Number |
16H05858
|
|---|
| Research Institution | Tohoku University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2016-04-01 – 2019-03-31
|
| Keywords | QoS / QoE / Security / 5G / IoT / ネットワーク / wireless network / security |
|---|
| Outline of Annual Research Achievements |
I used game theory to address the tradeoff of QoS and security in the next generation wireless networks. Particularly for the 4G and beyond 4G networks, the tradeoff issue between QoS and security in terms of throughput, delay, encryption/decryption, authentication, and so forth were taken into consideration. The problem was formally constructed using a game-based model by arguing that a model couldn't be constructed to trivially apply conventional optimization methods. Both mobility and fixed scenarios were considered and simulations were conducted showing the effectiveness of the game based method. We then extended our research to other next generation network such as IoT where QoS and security optimization is also critical.
In this fiscal year, we examined the algorithm from multiple approaches and improved the proposed method.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
As with last year, I have used game theory to integrate security and QoS. However, from the recent breakthrough in deep learning, I am studying how to improve the network QoS aspect more by training using large network datasets with the aid of deep learning techniques.
|
| Strategy for Future Research Activity |
Now I am studying the 5G network architecture. In the future, particular focus will be given on ultra dense network or UDN where beamforming and massive MIMO technologies are used. Using traditional FDD or TDD, such technologies are prone to resource allocation problems. I am planning to using deep learning, which is the state-of-the-art machine learning / Artificial Intelligence (AI) technique, to plan for intelligent resource control for mobile users in 5G UDNs to alleviate potential congestion and support next generation services and applications. After the resource allocation based QoS and QoE implementation, security integration focus with deep learning will also be taken into account.
|
