隨著智慧型手機的普及，全球定位系統 (Global Positioning System, GPS) 廣泛地被應用於室外定位 (Outdoor Positioning)，許多與生活息息相關的服務也都依靠GPS，例如近期為了防止新冠肺炎 (Coronavirus, COVID-19) 疫情擴張，政府必須掌握自我隔離者、已知感染者與其密集接觸者的行蹤，依照定位資料及時間，使得政府能夠發送資訊提醒與感染者在同一時間與地點活動的人民注意自己的身體狀況，以提升防疫效果。
不過受到建築遮蔽所造成訊號衰減等因素，GPS在室內定位的準確度與室外相差甚遠。即使GPS在室內的訊號不佳，但對於室內定位的需求並無減少，故近年來在室內環境中實現高精確度之定位系統成為了熱門的研究議題。
本研究使用手持裝置 (例如：平板電腦) 來模擬未來發展至成熟且普及的穿戴式裝置 (例如：智慧眼鏡) ，在不額外架設Wi-Fi分享器或iBeacon等硬體設備，純粹依靠手持裝置上的鏡頭擷取影像，利用卷積神經網路 (Convolutional Neural Network, CNN)，來實現室內定位系統。本篇論文所提出之系統於實際場域進行試驗，並以不同路徑、角度、天氣、時間實測，以驗證本文提出的方法。同時，透過我們提出來的合理定位過濾 (Reasonable Positioning Filter, RPF) 程序改良後，平均定位精準度可達到94.181%。

As smart phone has become more popular, Global Positioning System (GPS) has been widely used in outdoor positioning. Many services in our life are related to the position. For example, to prevent the expansion of coronavirus (COVID-19) outbreak, recently, the government tries to track self-isolators or infected people and then notice those people who are active at the same times and places for raising the efficiency of epidemic prevention.
However, the performance of indoor positioning based on GPS is not good enough because the signal is declined by the buildings and the obstacles. The need of indoor positioning is still desired even if GPS performance is worse in an indoor environment. Hence, developing a precise Indoor Positioning System (IPS) becomes a hot topic in recent years.
In this study, we use handheld devices (e.g. tablet) to mimic future wearable devices (e.g. smart glasses). Without relying on any additional hardware (e.g. Wi-Fi access point or iBeacon), we employ handheld devices with a camera to fetch real-time images and recognize them with the assistance of Convolutional Neural Network (CNN) to carry out an indoor positioning system (IPS). To validate the concept, we conducted several experiments in real indoor environments and the experimental scenarios include walking on different paths with a variety of speeds in different weathers at different times. Meanwhile, the proposed Reasonable Positioning Filter (RPF) can make the average accuracy of indoor positioning achieve 94.181%.

[1] L. Yuan-Cheng, H. Frannie, Y. Yi-Hsuan, L. Ching-Neng, and S. Yu-Chin, "A GPS navigation system with QR code decoding and friend positioning in smart phones," in 2010 2nd International Conference on Education Technology and Computer, 2010, vol. 5, pp. V5-66-V5-70.
[2] X. Bao, H. Bie, and M. Wang, "Integration of multimedia and location-based services on mobile phone tour guide system," in 2009 IEEE International Conference on Network Infrastructure and Digital Content, 2009, pp. 642-646.
[3] C. Yang, P. Tsai, M. Liao, C. Huang, and C. E. Yeh, "Location-Based Mobile Multimedia Push System," in 2010 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, 2010, pp. 181-184.
[4] C. C. Ho and R. Lee, "Real-Time Indoor Positioning System Based on RFID Heron-Bilateration Location Estimation and IMU Inertial-Navigation Location Estimation," in 2015 IEEE 39th Annual Computer Software and Applications Conference, 2015, vol. 3, pp. 481-486.
[5] P. Bahl and V. N. Padmanabhan, "RADAR: an in-building RF-based user location and tracking system," in Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064), 2000, vol. 2, pp. 775-784 vol.2.
[6] J. Leu and H. Tzeng, "Received Signal Strength Fingerprint and Footprint Assisted Indoor Positioning Based on Ambient Wi-Fi Signals," in 2012 IEEE 75th Vehicular Technology Conference (VTC Spring), 2012, pp. 1-5.
[7] C. Yang and H. Shao, "WiFi-based indoor positioning," IEEE Communications Magazine, vol. 53, no. 3, pp. 150-157, 2015.
[8] T. Shuo, M. Lin, and X. Yubin, "A compression and reconstruction algorithm of indoor WLAN positioning Radiomap data based on neural network," in 11th International Conference on Wireless Communications, Networking and Mobile Computing (WiCOM 2015), 2015, pp. 1-6.
[9] G. Anastasi, R. Bandelloni, M. Conti, F. Delmastro, E. Gregori, and G. Mainetto, "Experimenting an indoor bluetooth-based positioning service," in 23rd International Conference on Distributed Computing Systems Workshops, 2003. Proceedings., 2003, pp. 480-483.
[10] Z. Jianyong, L. Haiyong, C. Zili, and L. Zhaohui, "RSSI based Bluetooth low energy indoor positioning," in 2014 International Conference on Indoor Positioning and Indoor Navigation (IPIN), 2014, pp. 526-533.
[11] W. Kang and Y. Han, "SmartPDR: Smartphone-Based Pedestrian Dead Reckoning for Indoor Localization," IEEE Sensors Journal, vol. 15, no. 5, pp. 2906-2916, 2015.
[12] J. Kim and H. Jun, "Vision-based location positioning using augmented reality for indoor navigation," IEEE Transactions on Consumer Electronics, vol. 54, no. 3, pp. 954-962, 2008.
[13] Z. Liu, Y. Xu, Z. Liu, and J. Wu, "A large scale 3D positioning method based on a network of rotating laser automatic theodolites," in The 2010 IEEE International Conference on Information and Automation, 2010, pp. 513-518.
[14] C. Lim, S. Kang, H. Cho, S. Park, and J. Park, "An Enhanced Indoor Localization Algorithm Based on IEEE 802.11 WLAN Using RSSI and Multiple Parameters," in 2010 Fifth International Conference on Systems and Networks Communications, 2010, pp. 238-242.
[15] E. C. L. Chan, G. Baciu, and S. C. Mak, "Using the Newton Trust-Region Method to Localize in WLAN Environment," in 2009 IEEE International Conference on Wireless and Mobile Computing, Networking and Communications, 2009, pp. 363-369.
[16] R. Azuma, Y. Baillot, R. Behringer, S. Feiner, S. Julier, and B. MacIntyre, "Recent advances in augmented reality," IEEE Computer Graphics and Applications, vol. 21, no. 6, pp. 34-47, 2001.
[17] Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998.
[18] K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," arXiv e-prints, p. arXiv:1512.03385
[19] K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," arXiv e-prints, p. arXiv:1409.1556
[20] C. Szegedy et al., "Going Deeper with Convolutions," arXiv e-prints, p. arXiv:1409.4842
[21] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, "Rethinking the Inception Architecture for Computer Vision," arXiv e-prints, p. arXiv:1512.00567
[22] S. Ioffe and C. Szegedy, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift," arXiv e-prints, p. arXiv:1502.03167
[23] S. Ruder, "An overview of gradient descent optimization algorithms," arXiv e-prints, p. arXiv:1609.04747
[24] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. Alemi, "Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning," arXiv e-prints, p. arXiv:1602.07261