Traffic-Aware Resource Allocation and Spectrum Share for LTE-U and Wi-fi

In order to solve the effective coexistence problem between long-term evolution (LTE) in unlicensed bands (LTE-U) and Wi-fi, a dynamic resource allocation scheme-based traffic-aware and reinforcement learning (RL) is proposed. We proposed a two-layer network composed of prediction and decision which allows LTE-U to make full use of the channel without causing interference with Wi-fi. The first layer can perceive Wi-fi system signal and train a predictive network with perceived data. Then, it can predict the Wi-fi access regular pattern at next period and put the prediction result into the second layer network. The second layer network can interact with the environment and maximize the system rewards with deep Q-network (DQN). Simulation results show the proposed algorithms can maximize the throughput and delay satisfaction degree of LTE-U without causing interference to Wi-fi.

Authors and Affiliations

1. State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing, China

   Pengming Tian

Corresponding author

Correspondence to Pengming Tian .

Editors and Affiliations

1. Department of Computer Science and Engineering, SOA University, Bhubaneswar, Odisha, India

   Srikanta Patnaik

2. Department of Information Management and Business Analytics, Montclair State University, Montclair, NJ, USA

   John Wang

3. School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China

   Zhengtao Yu

4. Department of Information Technology, Techno India College of Technology, Durgapur, Kolkata, India

   Nilanjan Dey

Reprints and permissions

© 2020 Springer Nature Singapore Pte Ltd.

Policies and ethics