本論文的研究動機在於如何使用深度學習應用於室內定位的技術上，並探討模型結構對於定位準確度的影響。
在本研究的第一部分（詳見第二章），主要介紹為了蒐集數據所開發的手機應用程式。由於在室內定位的主題中，現今大多所使用的訊號為藍芽或WiFi等訊號，藉由分析、處理這些訊號達到室內定位的應用。因為無線網路基地台的取得容易，因此在本論文中所蒐集與使用的數據為WiFi訊號，並進一步的進行分析，此電磁訊號稱為「接收訊號強度指標」（Received Signal Strength Indications, RSSI）。
在本研究的第二部分（詳見第三章），主要探討利用深度學習應用於室內定位之方法，以實現「利用電磁訊號辨識法實現無線區域網路定位」。在本論文中所使用深度學習的方法為「卷積神經網絡」（Convolutional Neural Network, CNN），利用將蒐集到的RSSI依照特定的方式排列成圖像，利用CNN良好的圖形分類特性實現室內定位之應用。
此外，在本論文的第二部分中亦討論了利用奇異值分解（Singular Value Decomposition, SVD）搭配Hankel法來分離訓練資料集當中的雜訊，透過分離雜訊並降低雜訊來提升深度學習模型的穩健性與準確度。在經過大量的實驗與調適CNN模型合適的參數，最後證明了在訊號特徵較少的情況下亦能有優於現有方法的表現。
本論文共分四章。第一章為緒論，介紹研究動機與目的、文獻回顧以及論文架構。第二章介紹WiFi訊號量測手機APP之研製。第三章探討以電磁訊號辨識法實現無線區域網路定位，包含卷積神經網絡（CNN）、奇異值分解（SVD）與Hankel法。第四章為結論。

In this paper, we applied to technology of indoor localization by deep learning, and discussed structure of model effect upon accuracy of positioning.
In the first part of this study, we introduced the application which is used to collect signal data from WAP(Wireless Access Point). Choosing WAP to be signal data is due to that WIFI and Bluetooth signal is easy to measure in the topic of indoor localization. Therefore, in this paper, we took WIFI’s signal as electromagnetic signal, which called RSSI (Received Signal Strength Indications).
In the second part of this study, we explored the method of using deep learning to apply indoor positioning to realize the "Localization of Wireless Local Area Network by Identification of Electromagnetic Signals". In this research, we used CNN (Convolutional Neural Network) to realize indoor localization. We arranged signals which are collected RSSIs into feature images, and used these feature images to achieve indoor localization.
In addition, SVD (Singular Value Decomposition) with Hankel method was adopt as the method of noise reduction.By extensive experimental results and adjusting parameters of CNN model, we proved that there are better performances in the case of fewer features.

[1] Marc Ciurana, Francisco Barceló and Sebastiano Cugno, “Indoor tracking in WLAN location with TOA measurements”, Proceedings of the 4th ACM international workshop on Mobility management and wireless access, pp. 121-125, Oct. 02, 2006.
[2] Carl Wong, Richard Klukas and Geoffrey G. Messier, “Using WLAN Infrastructure for Angle-of-Arrival Indoor User Location”, IEEE 68th Vehicular Technology Conference, pp. 1090-3038, Oct. 24, 2006.
[3] Muzaiyanah Hidayab, Abdul Halim Ali and Khairul Bariah Abas Azmi, “Wifi signal propagation at 2.4 GHz”, IEEE Asia Pacific Microwave Conference, pp. 2165-4743, Singapore, Jan. 6, 2010.
[4] K. Kaemarungsi and P. Krishnamurthy, “Properties of indoor received signal strength for WLAN location fingerprinting”, The First Annual International Conference on Mobile and Ubiquitous Systems: Networking and Services, pp. 14-23, Boston, Aug. 26, 2004.
[5] Yifei Jiang, Xin Pan, Kun Li, Qin Lv, Robert P. Dick, Michael Hannigan and Li Shang, “ARIEL: automatic wi-fi based room fingerprinting for indoor localization”, Proceedings of the 2012 ACM Conference on Ubiquitous Computing, pp. 441-450, Pittsburgh, Pennsylvania, Sep. 05, 2012.
[6] Shih-Hau Fang and Tsung-Nan Lin, “Indoor Location System Based on Discriminant-Adaptive Neural Network in IEEE 802.11 Environments”, IEEE Transactions on Neural Networks, pp. 1973-1978, Nov. 07, 2008.
[7] Zhang Wei, Liu Kan, Zhang Weidong, Zhang Youmei and Gu Jason, “Deep Neural Networks for wireless localization in indoor and outdoor environments”, Journal Neurocomputing, pp. 279-287, Feb. 18, 2016.
[8] Hao Chen, Yifan Zhang, Wei Li, Xiaofeng Tao and Ping Zhang, “ConFi: Convolutional Neural Networks Based Indoor Wi-Fi Localization Using Channel State Information”, IEEE Access, vol. 5, pp. 18066 – 18074, Sep. 07, 2017.
[9] T. K. Moon and W. C. Stirling WC, Mathematical Methods and Algorithms for Signal Processing, NJ: Prentice Hall, 2000.
[10] K. C. Lee KC, J. S. Ou, and M. C. Fang, “Application of SVD noise-reduction technique to PCA based radar target recognition”, Prog. Electromagn. Res., vol. 81, pp. 447-459, 2008.
[11] Hermus K, Dologlou I, Wambacq P, and Compernolle DV, “Fully adaptive svd-based noise removal for robust speech recognition”, Sixth European Conference on Speech Communication and Technology, pp. 1951-1954, Budapest, Hungary, Sep. 5-9, 1999.
[12] K. C. Lee and T. N. Lin, “Application of neural networks to analyses of nonlinearly loaded antenna arrays including mutual coupling effects”, IEEE Trans. Antennas Propagat., vol. 53, no. 3, pp. 1126-1132, Mar. 14, 2005.
[13] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, “Gradient-based learning applied to document recognition”, Proceedings of the IEEE, vol. 86, pp. 2278–2324, Nov. 1998.
[14] C. Christodoulou and M. Georgiopoulos, Applications of Neural Networks in Electromagnetics, MA: Artech House, 2000.
[15] Matthew D. Zeiler, “ADADELTA: An Adaptive Learning Rate Method”, Cornell University Library, arXiv.org, cs, arXiv., pp. 1212.5701, 2012.
[16] Pascal Vincent, Hugo Larochelle, Yoshua Bengio and Pierre-Antoine Manzagol, “Extracting and composing robust features with denoising autoencoders”, Proceedings of the Twenty-fifth International Conference on Machine Learning, pp.1096-1103, Jul. 05-09, 2008.