TY - JOUR
T1 - LTE-U and Wi-Fi Coexistence Algorithm Based on Q-Learning in Multi-Channel
AU - Su, Yuhan
AU - Du, Xiaojiang
AU - Huang, Lianfen
AU - Gao, Zhibin
AU - Guizani, Mohsen
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/2/9
Y1 - 2018/2/9
N2 - Due to the lack of resources in the low spectrum, long term evolution (LTE) in unlicensed spectrum (LTE-U) technology has been proposed to extend LTE to unlicensed spectrum. LTE-U undertakes the task of streaming data traffic for licensed spectrum, which can greatly enhance the capacity of the system. However, the introduction of LTE-U technology also gives rise to the problem of coexistence with Wi-Fi systems. In this paper, an LTE-U and Wi-Fi coexistence algorithm is proposed in multi-channel scenarios based on Q-learning. By taking the idea of alternately transferring data in LTE-U and Wi-Fi, the algorithm takes into account both the fairness and the performance of the system and optimizes the duty cycle. The simulation results show that the proposed algorithm can effectively improve the throughput of the system in the premise of ensuring fairness.
AB - Due to the lack of resources in the low spectrum, long term evolution (LTE) in unlicensed spectrum (LTE-U) technology has been proposed to extend LTE to unlicensed spectrum. LTE-U undertakes the task of streaming data traffic for licensed spectrum, which can greatly enhance the capacity of the system. However, the introduction of LTE-U technology also gives rise to the problem of coexistence with Wi-Fi systems. In this paper, an LTE-U and Wi-Fi coexistence algorithm is proposed in multi-channel scenarios based on Q-learning. By taking the idea of alternately transferring data in LTE-U and Wi-Fi, the algorithm takes into account both the fairness and the performance of the system and optimizes the duty cycle. The simulation results show that the proposed algorithm can effectively improve the throughput of the system in the premise of ensuring fairness.
KW - Long term evolution (LTE)-U
KW - Q-Learning
KW - Wi-Fi
KW - coexistence algorithm
KW - reinforcement learning
UR - http://www.scopus.com/inward/record.url?scp=85041830886&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85041830886&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2018.2803258
DO - 10.1109/ACCESS.2018.2803258
M3 - Article
AN - SCOPUS:85041830886
VL - 6
SP - 13644
EP - 13652
JO - IEEE Access
JF - IEEE Access
ER -