隨著著名的全球衛星定位系統（GPS）已經能夠達到百米以內的精準度，戶外定位服務（Location-Based Service）漸漸地趨於成熟。然而由於衛星訊號無法穿透建築物，全球衛星定位系統（GPS）並不適用於室內定位的需求，因而室內適地性的服務 (Indoor Location-Based Service) 儼然成為各家爭相研究的領域。另一方面，卷積神經網路 (Convolutional Neural Network) 在近年的各項電腦視覺領域競賽無往不利收穫了傲人的成績，其利用卷積層 (Convolution Layer) 由點的比對轉換成局部的比對的方式透過將一塊塊的特徵研判並擷取，逐步堆疊綜合比對訓練出更完善辨識率更好的模型架構，擊敗各路競爭對手的同時也引來大量目光聚焦，紛紛想研究了解突破性成果背後的原理及運作。
綜觀以上，本論文提出了基於 Bluetooth Mesh Network 之藍芽燈具的卷積神經網路室內定位模型架構，以 BLE Beacon 作為定位目標、 Bluetooth Mesh 藍芽燈具作為接收端的角色，透過擷取 BLE Beacon 於藍芽燈具之 Beacon 感測器上的接收訊號強度 (RSSI) 作為本論文提出之卷積神經網路室內定位模型的輸入，利用等同實驗場域長寬大小的 RSSI 組成灰階圖像送入神經網路模型中進行訓練，最後儲存訓練完成的網路模型並輸入測資進行定位目標位置預測。在模型優化及提升定位精準度方面，本論文使用卡爾曼濾波器 (Kalman Filter) 及高斯模糊 (Gaussian Blur) 兩種資料前處理方式將擷取到的 RSSI 訊號先行濾波、送入神經網路的灰階圖像也先進行平滑化的處理，使得最終模型定位精準度能達到定位誤差落在 0.79m-1.89m 間並且標準差在 0.27m~0.49m 間。

Indoor locality services are not only widely used and have high commercial value, such as long-term medical care (elderly tracking), factory storage (asset monitoring), airport stations (route guidance), department stores (indoor shopping guide).

In this thesis, we propose an indoor positioning system based on Bluetooth mesh network. First, we replaced the traditional lights in the experimental field with Delta's Bluetooth mesh lights, which can measure the signal strength of the beacon and send the results to the gateway through the Bluetooth network. We can judge the positioning target based on the RSSI value retrieved from the gateway.

In this thesis, we propose a Convolutional Neural Network indoor positioning model, which can achieve an average positioning error of 1.25m in our experimental field. We store the trained network model and sequentially send the testing data into the model to predict the location of the target Beacon.

[1] Daniel Minoli, Kazem Sohraby, and Benedict Occhiogrosso (2017), "IoT considerations, requirements, and architectures for smart buildings—Energy optimization and next-generation building management systems", IEEE Internet of Things Journal 4.1: 269-283.
[2] Petros Spachos, Ioannis Papapanagiotou, and Konstantinos N. Plataniotis (2018), "Microlocation for smart buildings in the era of the internet of things: A survey of technologies, techniques, and approaches", IEEE Signal Processing Magazine 35.5: 140-152.
[3] Junjie Liu (2014) , "Survey of Wireless Based Indoor Localization Technologies", https://www.cse.wustl.edu/~jain/cse574-14/ftp/indoor/index.html
[4] Faheem Zafari, Athanasios Gkelias, and Kin K. Leung (2019) , "A survey of indoor localization systems and technologies", IEEE Communications Surveys & Tutorials 21.3: 2568-2599.
[5] Mesh Working Group, "Mesh Profile Bluetooth® Specification", https://www.bluetooth.com/specifications/mesh-specifications
[6] Jan Blumenthal, Ralf Grossmann, Frank Golatowski, Dirk Timmermann (2007) , "Weighted centroid localization in zigbee-based sensor networks", 2007 IEEE international symposium on intelligent signal processing.
[7] Kiran Thapa and Steven Case (2003) , "An Indoor Positioning Service for Bluetooth Ad Hoc Networks", Midwest Instruction and Computing Symposium (MICS). 45
[8] V.S. Abhayawardhana, I.J. Wassell, D. Crosby, M.P. Sellars, M.G. Brown (2005), "Comparison of empirical propagation path loss models for fixed wireless access systems", 2005 IEEE 61st Vehicular Technology Conference.
[9] http://wits.ice.nsysu.edu.tw/course/pdfdownload/953G/第 07 章-ChannelModel.pdf
[10] Greg Welch and Gary Bishop (2006) , "An Introduction to the Kalman Filter ".
[11] I. Guvenc and O. Dedeoglu (2003), "Enhancements to RSS based indoor tracking systems using Kalman filters", Diss. University of New Mexico.
[12] Jun Wang, Paulo Urriza, Yuxing Han, Danijela Cabric (2011) , "Weighted centroid localization algorithm: theoretical analysis and distributed implementation", IEEE Transactions on wireless communications 10.10: 3403-3413.
[13] Santosh Subedi, Goo-Rak Kwon, Seokjoo Shin, Suk-seung Hwang, Jae-Young Pyun (2016) , "Beacon based indoor positioning system using weighted centroid localization approach", 2016 Eighth International Conference on Ubiquitous and Future Networks (ICUFN).
[14] Fazli Subhan, Halabi Hasbullah, Azat Rozyyev, Sheikh Tahir Bakhsh (2011) , "Indoor positioning in bluetooth networks using fingerprinting and lateration approach", 2011 International Conference on Information Science and Applications.
[15] Tamer Dag and Taner Arsan (2018) , "Received signal strength based least squares lateration algorithm for indoor localization", Computers & Electrical Engineering 66 (2018): 114-126.
[16] Long Cheng, Chengdong Wu and Yunzhou Zhang (2012) , "A survey of localization in wireless sensor network", International Journal of Distributed Sensor Networks 8.12: 962523. 46
[17] R. Bajaj, S.L. Ranaweera, D.P. Agrawal (2002) , "GPS: location-tracking technology", Computer 35.4: 92-94.
[18] Jae Hyung An and Lynn Choi (2016) , "Inverse fingerprinting: server side indoor localization with Bluetooth low energy", 2016 IEEE 27th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).
[19] Ju Wang, Jie Xiong, Hongbo Jiang, Kyle Jamieson, Xiaojiang Chen, Dingyi Fang, Chen Wang (2016) , "LiFS: low human-effort, device-free localization with fine-grained subcarrier information", Proceedings of the 22nd Annual International Conference on Mobile Computing and Networking.
[20] J-Y Chang (2017) , "BLE-based Indoor Positioning and Tracking System using Server-Side n-dependent Fingerprinting associated with Trilateration ", in Master thesis, Institute of Computer and Communication Engineering, National Cheng-Kung University, Taiwan.
[21] Mesh Working Group , "Mesh Model Bluetooth® Specification", https://www.bluetooth.com/specifications/mesh-specifications
[22] Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov (2014) , "Dropout: a simple way to prevent neural networks from overfitting", Journal of machine learning research. 15(1):1929–1958.
[23] Marco Altini, Davide Brunelli, Elisabetta Farella, Luca Benini (2010) , "Bluetooth indoor localization with multiple neural networks", IEEE 5th International Symposium on Wireless Pervasive Computing 2010:295-300
[24] S. Feldmann, K. Kyamakya and Zighuo Lue (2003) , "An Indoor Bluetooth-Based Positioning System: Concept, Implementation and Experimental Evaluation", International Conference on Wireless Networks. vol. 272. 47
[25] Christoph Fuchs, Nils Aschenbruck, Peter Martini, MonikaWieneke (2011) , "Indoor tracking for mission critical scenarios: A survey", Pervasive and Mobile Computing. 7(1):1–15.
[26] Zohaib Iqbal, Da Luo, Peter Henry, Samaneh Kazemifar, Timothy Rozario, Yulong Yan, Kenneth Westover, Weiguo Lu, Dan Nguyen, Troy Long, Jing Wang, Hak Choy, Steve Jiang (2018) , "Accurate real time localization tracking in a clinical environment using Bluetooth Low Energy and deep learning".
[27] A. Bekkelien, Michel Deriaz, S. Marchand-Maillet (2012) , "Bluetooth indoor positioning", Master’s thesis, University of Geneva.
[28] Anil Madhavapeddy and Alastair Tse (2005) , "A study of bluetooth propagation using accurate indoor location mapping", UbiComp 2005: Ubiquitous Computing: 903-903.