在未來的5G通訊系統中，將會有一項至為關鍵的設備，為了改善通訊設備越來越多，而導致大型基地台的頻譜覆蓋不足，這項設備也就是微小型基地台(Small Cell)，微小型基地台不僅在佈建方面靈活，體積也比目前家用型Wi-Fi大上一倍而已，能夠隨意地擺放於室內，並且能提升整體的網路容量。
本論文將提出一種於次世代異質性網路中基於人工智慧方法分配無線資源之研究，這項技術將安裝在Small Cell上面，希望能藉由此人工智慧的方式來減少微小型基地台在分配資源時，因為封包的碰撞而浪費的時間或是預留的資源塊沒有被預期的使用，並且透過微小型基地台因為新技術而增加的切換功能，以改善有執照頻譜的短缺、以及未來可能會產生有執照頻譜與免執照頻譜的使用者因為使用共同的網路系統而發生干擾，希望能想出一個公平的機制提升整體網路效能和保持每個用戶的服務品質。
本研究將做數個流量的實驗來驗證深度學習是否能套用於流量，並且模擬了兩台微小型基地台以及數位使用者在不同情境時的吞吐量與延遲，並使用Python以人工智慧方法學習真實的影音流量，再透過深度學習進行預測結果，最後我們就透過本研究所採用的新技術與預測結果做結合進行資源分配。

In the future, there will be a critical device in the 5G communication system. In order to improve communications devices more and more, As a result, the spectrum coverage of base stations is insufficient. This equipment called Small cell. It's not only in terms of flexibility, volume than the current home-based Wi-Fi only twice as large, it can be placed freely in the room, and can enhance the overall network.
This paper will propose a Resource Allocation Using Artificial Intelligence in Next Generation Heterogeneous Networks. This technology will be installed on the Small Cell. Hoping by artificial intelligence to reduce the distribution of micro-base station means of when the resource. Because the time wasted by the collision of the packet or the reserved resources were not intended use. Through micro-small cell because of new technology increase to the switching function in order to improve the shortage of licensed spectrum, as well as possible future generate licensed. Users who spectrum and unlicensed spectrum will interfere with the use of a common network system, and in the future, users with licensed spectrum and unlicensed spectrum may experience interference due to the use of a common network system. Hope to come up with a fair mechanism to improve overall network performance and maintain the quality of service for each user.

This study will do several flow experiments to verify whether deep learning can be applied to traffic, to simulate the throughput and latency of two micro-base stations and digital users in different situations, and use Python to learn by artificial intelligence. Then through deep learning to predict results. Finally we combine the new technology and prediction results used in this study to allocate resources.

[1]J. Bergstra, O. Breuleux, F. Bastien, P. Lamblin, R. Pascanu, G. Desjardins, J. Turian, D. Warde-Farley, Y. Bengio, "Theano: A cpu and gpu math compiler in python", Proc. 9th Python in Science Conf, vol. 1, 2010.
[2]K. Cho, B. van Merrienboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, Y. Bengio, "Learning phrase representations using rnn encoder-decoder for statistical machine translation", Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1724-1734, 2014.
[3]J. Wang, J. Tang, Z. Xu, Y. Wang, G. Xue, X. Zhang, D. Yang, "Spatiotemporal modeling and prediction in cellular networks: A big data enabled deep learning approach", INFOCOM 2017-IEEE Conference on Computer Communications, pp. 1-9, 2017.
[4]V. Mushunuri, B. Panigrahi, H. K. Rath and A. Simha, "Fair and Efficient Listen Before Talk (LBT) Technique for LTE Licensed Assisted Access (LAA) Networks," 2017 IEEE 31st International Conference on Advanced Information Networking and Applications (AINA), Taipei, 2017, pp. 39-45.
[5]W. Wang, P. Xu, Y. Zhang and H. Chu, "Network-sensitive adaptive LAA LBT strategy for downlink LAA-WiFi coexistence," 2017 9th International Conference on Wireless Communications and Signal Processing (WCSP), Nanjing, 2017, pp. 1-6.
[6]J. Xiao, J. Zheng, L. Chu and Q. Ren, "Performance Modeling of LAA LBT with Random Backoff and a Variable Contention Window," 2018 10th International Conference on Wireless Communications and Signal Processing (WCSP), Hangzhou, 2018, pp. 1-7.
[7]U. Challita, L. Dong and W. Saad, "Proactive Resource Management for LTE in Unlicensed Spectrum: A Deep Learning Perspective," in IEEE Transactions on Wireless Communications, vol. 17, no. 7, pp. 4674-4689, July 2018.
[8]M. Haider and M. Erol-Kantarci, "Enhanced LBT Mechanism for LTE-Unlicensed Using Reinforcement Learning," 2018 IEEE Canadian Conference on Electrical & Computer Engineering (CCECE), Quebec City, QC, 2018, pp. 1-4.
[9]B. Chen, J. Chen, Y. Gao and J. Zhang, "Coexistence of LTE-LAA and Wi-Fi on 5 GHz With Corresponding Deployment Scenarios: A Survey," in IEEE Communications Surveys & Tutorials, vol. 19, no. 1, pp. 7-32, Firstquarter 2017.
[10]Z. Ali, L. Giupponi, J. Mangues-Bafalluy and B. Bojovic, "Machine learning based scheme for contention window size adaptation in LTE-LAA," 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), Montreal, QC, 2017, pp. 1-7.
[11]H. D. Trinh, L. Giupponi and P. Dini, "Mobile Traffic Prediction from Raw Data Using LSTM Networks," 2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), Bologna, 2018, pp. 1827-1832.
[12]M. Yazid, A. Ksentini, L. Bouallouche-Medjkoune and D. Aïssani, "Performance Analysis of the TXOP Sharing Mechanism in the VHT IEEE 802.11ac WLANs," in IEEE Communications Letters, vol. 18, no. 9, pp. 1599-1602, Sept. 2014.
[13]B. Wei, W. Kawakami, K. Kanai, J. Katto and S. Wang, "TRUST: A TCP Throughput Prediction Method in Mobile Networks," 2018 IEEE Global Communications Conference (GLOBECOM), Abu Dhabi, United Arab Emirates, 2018, pp. 1-6.