研究生: |
黃彥誠 Yen-Cheng Huang |
---|---|
論文名稱: |
基於機器學習的毫米波多用戶波束合成 Learning based multiuser millimeter wave beam selection |
指導教授: |
林士駿
Shih-Chun Lin |
口試委員: |
陳俊仰
Chun-Yang Chen 謝松年 Sung-Nien Hsieh |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 電子工程系 Department of Electronic and Computer Engineering |
論文出版年: | 2021 |
畢業學年度: | 109 |
語文別: | 中文 |
論文頁數: | 40 |
中文關鍵詞: | 毫米波 、波束對準 、機器學習 、位置補助 |
外文關鍵詞: | Millimeter wave, Beam alignment, Machine learning, Position aided |
相關次數: | 點閱:256 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
在毫米波通訊中,準確和快速的波束對準(beam alignment)是必要的。而數據驅動(data-driven)的學習方法被認為是一個很有潛力的解決方案,可以用來解決波束對準的問題,此方法利用了補助訊息(side information)和現場量測(on the field measurement)。在本論文中我們提出基於支持向量機(support vector machine)的學習方法,當給定一小部份用戶位置的波束方向時,此方法利用用戶的位置補助資訊來預測其餘用戶位置可能的波束方向。我們使用 DeepMIMO 數據庫提供模擬用的通道參數。實驗結果顯示我們提出的方法和基於神經網路(neural network)的學習方法在能量損失機率上有差不多的錯誤率,不過在運行時間方面,我們所提出的方法完全勝過基於神經網路的學習方法。
Accurate and fast beam alignment is important in millimeter wave communications. A data-driven learning method is a promising solution to solve this problem by leveraging side information and on the field measurement. In this thesis, a learning method based on support vector machine is proposed to predict the beam direction on a set of possible users’ position, given beam direction measurements on a small subset of users’ position; based on these predictions, a small subset of beams is recommended, based on position side information. The proposed method is evaluated with DeepMIMO dataset, which provides parameterized channels for the experiment. The simulation result show that our method outperform the learning method based on neural network in computation complexity and also has same power loss probability.
[1] Va, Vutha, et al. "Inverse multipath fingerprinting for millimeter wave V2I beam alignment." IEEE Transactions on Vehicular Technology 67.5 (2017): 4042-4058.
[2] Hur, Sooyoung, et al. "Millimeter wave beamforming for wireless backhaul and access in small cell networks." IEEE transactions on communications 61.10 (2013): 4391-4403.
[3] Akdeniz, Mustafa Riza, et al. "Millimeter wave channel modeling and cellular capacity evaluation." IEEE journal on selected areas in communications 32.6 (2014): 1164-1179.
[4] Chou, Tzu-Hsuan et al. “Millimeter Wave Channel Recommendation System Aided by Tensor Completion.” arXiv: Signal Processing (2019)
[5] Rezaie, Sajad, Carles Navarro Manchón, and Elisabeth De Carvalho. "Location-and orientationaided millimeter wave beam selection using deep learning." ICC 2020-2020 IEEE International Conference on Communications (ICC). IEEE, 2020.
[6] Heath, Robert W., et al. "An overview of signal processing techniques for millimeter wave MIMO systems." IEEE journal of selected topics in signal processing 10.3 (2016): 436-453.
[7] Alkhateeb, Ahmed. "DeepMIMO: A generic deep learning dataset for millimeter wave and massive MIMO applications." arXiv preprint arXiv:1902.06435 (2019).
[8] Cai, Jian-Feng, Emmanuel J. Candès, and Zuowei Shen. "A singular value thresholding algorithm for matrix completion." SIAM Journal on optimization 20.4 (2010): 1956-1982.
[9] G. H. Golub and C. F. Van Loan, Matrix Computations, 4th ed. The Johns Hopkins University Press, 2013.
[10] I. Goodfellow, Y. Bengio, and A. Courville, “Deep Feedforward Network,” Deep Learning, MIT press, 2016, pp. 167–227
[11] Hsu, Chih-Wei, and Chih-Jen Lin. "A comparison of methods for multiclass support vector machines." IEEE transactions on Neural Networks 13.2 (2002): 415-425.
[12] F. Chollet. (2015). Keras. [Online]. Available: https://keras.io.
[13] Pedregosa, Fabian, et al. "Scikit-learn: Machine learning in Python." the Journal of machine Learning research 12 (2011): 2825-2830.
[14] Yoo, Taesang, Nihar Jindal, and Andrea Goldsmith. "Multi-antenna downlink channels with limited feedback and user selection." IEEE Journal on Selected Areas in Communications 25.7 (2007): 1478-1491.