DySPAN 2018 Paper Accepted
I just got notified that our submission to DySPAN 2018, Opportunistic Channel Access Using Reinforcement Learning in Tiered CBRS Networks, was accepted. Matthew Tonnemacher from SMU and Samsung Research America led this paper which focuses on using machine learning to help overcome the hidden terminal problem in the emerging CBRS band.
Machine learning has been getting extensive attention throughout the world over the last few years. Much of the buzz surrounds the achievements in classification tasks such as image recognition or the ability to outperform humans in complicated games such Go. However, it has not been widely adopted in wireless communications. In this paper, we use reinforcement learning to improve network performance by learning the ideal energy detection threshold for users to decide if the channel is available for transmission or not.
The IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN) is in Seoul, Korea this year from October 22nd through the 24th.
I’ll write more details about the work after the conference.
Update: Paper available:
- Enabling a “Use-or-Share” Framework for PAL–GAA Sharing in CBRS Networks via Reinforcement Learning
- Method and apparatus for improving coexistence performance by measurements in wireless communication systems
- Opportunistic channel access using reinforcement learning in tiered CBRS networks
- Machine Learning Enhanced Channel Selection for Unlicensed LTE
- Neural-Network DPD via Backpropagation through a Neural-Network Model of the PA