研究生: |
柯承佑 Cheng-Yu Ko |
---|---|
論文名稱: |
應用機器學習演算法模擬室內無線訊號強度之研究 Study of Applying Machine Learning Algorithm to Simulate Wireless Signal Strength in an Indoor Environment |
指導教授: |
呂政修
Jenq-Shiou Leu |
口試委員: |
呂政修
Jenq-Shiou Leu 陳省隆 Hsing-Lung Chen 陳永耀 Yung-Yaw Chen 王瑞堂 Jui-Tang Wang |
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: | 2020 |
畢業學年度: | 108 |
語文別: | 中文 |
論文頁數: | 30 |
中文關鍵詞: | 建築資訊模型 、三維空間 、數位孿生 、機器學習 |
外文關鍵詞: | Building Information Modeling(BIM), Three-dimensional space, digital twin, machine learning |
相關次數: | 點閱:355 下載:1 |
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 |
在建築資訊模型的世界中,數位孿生
(Digital Twin)是非常重要的核心概念,
已連續三年被資訊科技研究機構 Gartner評為全球十大科技趨勢之一,簡單來說,
數位孿生代表 透過虛擬世界中的數位物件 ,顯示現實世界中的物體可能的反應、
狀況與效能等。 在建築資訊模型的產業領域中, 無線基地台的布建是重大且具風
險的策略,若無線基地台擺設不好,輕則導致室內空間收訊不佳,重則須重新 施
工,造成施工成本大幅增加,因此許多業者期望透由數位孿生的 概念與機器學習
演算法的結合,在建築物尚未施工時期,利用無線訊號分布模型掌握場域中的訊
號分布作為無線基地台的布建策略參考依據。
在這篇論文中, 我們提出收集實際建築物場域中的無線訊號分布,導入至機
器學習演算法中進行模擬預測空間中的訊號分布模型,以三維空間的型式進行視
覺化呈現與評估該呈現方法是否能無痛導入至建築資訊模型中。為了評估此實驗
假設是否成立,我們 在實際建築物場域中 實地 擺設無線基地台 進行大量的無線訊
號強度值收集,並撰寫 KNN與 DNN機器學習演算法建立無線訊號模型,依此
模型數據比對實際空間的真實無線訊號強度值,作為評估本實驗目的的可行性。
In the of building information modeling, digital twin is a very important core concept. It has been rated as one of the top ten global technology trends by Gartner, an information technology research organization for three consecutive years. In simple terms, digital twin represents Digital objects in the world show the possible reactions, conditions and performance of objects in the real world. In the industrial field, the deployment of wireless base stations is a significant and risky strategy. If the wireless base station is not well-equipped, it will result in poor reception of indoor space, and heavy construction will require re-construction, resulting in significant construction costs. Increase, so many companies expect to combine the concept of digital twins with machine learning algorithms. Before the building is under construction, use the wireless signal distribution model to grasp the signal distribution in the field as a reference for the deployment strategy of wireless base stations.
In this paper, we propose to collect the wireless signal distribution in the actual building field, import it into a machine learning algorithm to simulate and predict the signal distribution model in the space, and visualize and evaluate the representation in a three-dimensional space. Whether the method can be painlessly imported into the building information model. In order to evaluate whether this experimental hypothesis is true, we set up wireless base stations in the field of actual buildings to collect a large number of wireless signal strength values, and wrote KNN and DNN machine learning algorithms to establish wireless signal models, based on this model data comparison The actual wireless signal strength value in the actual space is used to evaluate the feasibility of this experiment.
[1] S. Azhar, “Building Information Modeling (BIM): Trends, Benefits, Risks, and Challenges for the AEC Industry”, Leadership and Management in Engineering, Vol. 11, Issue 3, 2011.
[2] R. Sacks, L. Koskela, B. A. Dave and R. Owen, “Interaction of Lean and Building Information Modeling in Construction”, Journal of Construction Engineering and Management, Vol. 136, Issue 9, 2010.
[3] W. Liu, Z. Wang, X. Liu, N. Zeng, Y. Liu and F. E. Alsaadi, “A survey of deep neural network architectures and their applications”, Neurocomputing, Volume 234, 2017, Pages 11-26.
[4] R. Volk, J. Stengel and F. Schultmann, “Building Information Modeling (BIM) for existing buildings - Literature review and future needs”, Automation in Construction, Volume 38, 2014, Pages 109-127.
[5] A. Porwal and K. N. Hewage, “Building Information Modeling (BIM) partnering framework for public construction projects”, Automation in Construction, Volume 31, 2013, Pages 204-214.
[6] V. Garcia, E. Debreuve and M. Barlaud, “Fast k nearest neighbor search using GPU”, IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, 2008.
[7] Y. C. Gültekin, W. J. van Houtum, A. G. C. Koppelaar and F. M. J. Willems, "Enumerative Sphere Shaping for Wireless Communications With Short Packets," in IEEE Transactions on Wireless Communications, vol. 19, no. 2, pp. 1098-1112, Feb. 2020, doi: 10.1109/TWC.2019.2951139.
[8] M. Tucci, S. Barmada, L. Sani, D. Thomopulos and N. Fontana, "Deep Neural Networks Based Surrogate Model for Topology Optimization of Electromagnetic Devices," 2019 International Applied Computational Electromagnetics Society Symposium, Miami, FL, USA, 2019, pp. 1-2.
[9] J. Huang, Y. Wei, J. Yi and M. Liu, "An Improved kNN Based on Class Contribution and Feature Weighting," 2018 10th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA), Changsha, 2018, pp. 313-316, doi: 10.1109/ICMTMA.2018.00083.
[10] R. V and G. R, "AI based Semantic Extensibility and Querying Techniques for Building Information Model," 2019 International Conference on Intelligent Computing and Control Systems (ICCS), Madurai, India, 2019, pp. 1497-1501, doi: 10.1109/ICCS45141.2019.9065840.