次世代異質性網路中以人工智慧方法分配無線資源之研究

Heterogeneous Networks (HeNets) are an important feature in the 5th generation (5G) communication system. For HeNets consisting of long term evolution (LTE) Femtocell and WiFi Access Points (APs), Licensed Assisted Access LTE (LTE-LAA) is a promising approach to provide efficiency of spectrum utilization. In this project, we investigate the resource allocation problem of LTE-LAA Femtocells in 5G HeNets. We provide a resource allocation approach based on artificial intelligence (AI) for LTE Femtocell to improve the spectrum utilization and throughput fairness of HeNets. We provide an analytical model of HeNets consisting of Femtocell and WiFi APs in which Listen-Before-Talk (LBT) is adopted for accessing unlicensed spectrum. In particular, we investigate the impact of Transmission Opportunity (TXOP) to the channel access efficiency and fairness. Based on the analytical model, we therefore develop a resource allocation scheme using Reinforcement Learning (RL), which adjusts TXOP of Femtocell to improve the spectrum utilization and throughput fairness of HeNets. By both the theoretical analyses and solutions expected to be investigated in this project, we hope to thoroughly provide an in depth view about the design and implementation of the 5G communication system.