在5G通訊協定中，由於使用mmWave的基地台的覆蓋區域變小，因此常在一個區域內密集布置5G基地台，使得使用者可能在這些區域中頻繁換手，而導致使用者跟基地台間的斷線率提高，也讓使用者較有機會發生乒乓效應。根據3GPP的規定，換手時機會影響到 UE 的通訊服務品質，近年來也有一些研究以不同的方法來決定換手的時機。由於此問題的某些特性，使得這類問題適合以機器學習的方式來解決。在此篇論文中，我們設計了一個類神經網路來決定使用者是否要換手，我們的目標是降低斷線率以及乒乓效應發生的次數。實驗的結果顯示所設計的方法能成功使使用者不斷線，並有效減少乒乓效應的發生。

In recent years, 5G mmWave base stations have been deployed more and more dense. Since an mmWave base station only covers a small region, UE usually experiences more handover in 5G networks than in 4G networks. According to 3GPP specification, the handover timing significantly affect the communication quality, and thus, many methods determining the UE's handover timing are proposed in the literature. In this thesis, a learning-based method is proposed to determine the UE's handover timing. Simulation results show that the proposed algorithm has good performance in terms of the number of disconnections and the number of handovers, as compared to the state-of-the-art methods.

[1] K. D. C. Silva, Z. Becvar, and C. R. L. Frances, "Adaptive hysteresis margin
based on fuzzy logic for handover in mobile networks with dense small cells,"
IEEE Access, vol. 6, pp. 17 178-17 189, 2018.
[2] K. C. Silva, Z. Becvar, E. H. Cardoso, and C. R. Franc^es, "Self-tuning handover
algorithm based on fuzzy logic in mobile networks with dense small cells," in 2018
IEEE Wireless Communications and Networking Conference (WCNC). IEEE,
2018, pp. 1-6.
[3] L. Sun, J. Hou, and T. Shu, "Optimal handover policy for mmwave cellular net-
works: A multi-armed bandit approach," in 2019 IEEE Global Communications
Conference (GLOBECOM). IEEE, 2019, pp. 1-6.
[4] M. Sana, A. De Domenico, and E. C. Strinati, "Multi-agent deep reinforcement
learning based user association for dense mmwave networks," in 2019 IEEE Global
Communications Conference (GLOBECOM). IEEE, 2019, pp. 1-6.
[5] Z. Wang, L. Li, Y. Xu, H. Tian, and S. Cui, "Handover optimization via asyn-
chronous multi-user deep reinforcement learning," in 2018 IEEE International
Conference on Communications (ICC). IEEE, 2018, pp. 1-6.
[6] Zhi-Hong Huang, Yi-Lin Hsu, Pu-Kang Chang, and Ming-Jer Tsai, "Ecient
Handover Algorithm in 5G Networks using Deep Learning ," Technical report,
2020.
[7] J. Bai, S.-p. Yeh, F. Xue, and S. Talwar, "Route-aware handover enhancement for
drones in cellular networks," in 2019 IEEE Global Communications Conference
(GLOBECOM). IEEE, 2019, pp. 1-6.