隨著人工智慧和物聯網技術的發展，將有越來越多的傳統應用與服務趨向智慧化與自動化。舉例而言，利用自動化機器人在辦公室的座位間遞送紙本文件、在醫院的病房與檢驗實驗室之間傳送病人的血液樣本，或是在博物館琳瑯滿目的展品之間提供準確且智慧化的導覽解說。諸如此類的室內適地性服務(Location Based Service, LBS)，皆需仰賴精確的室內定位系統才能達成與實現。
本論文提出一個雙重多數決(Double Majority Decision)室內定位機制，同時採取RSSI與室內磁場變異性(Indoor Magnetic Field Anomalies)作為定位的特徵。此外，結合支持向量機(Support Vector Machine, SVM)以提升定位的準確度。SVM會針對接收訊號強度(Received Signal Strength Indicator, RSSI)、磁場強度及整合這兩者特徵訓練出對應的模型，分別為RSSI模型、磁場模型與混合模型。在測試定位階段，每一筆特徵樣本都會利用這三個模型預測的結果進行第一次多數決，以多數決的結果作為此筆特徵樣本的預測位置。而每一個測試點都收集五筆特徵樣本，根據這五筆特徵樣本的預測結果進行第二次多數決，此結果即為測試點最終的定位位置。
此定位系統實現並測試於兩個不同的室內場域。第一個位於台大醫院二樓的走廊，長寬尺寸為6.3m×2.4m。透過本論文提出的定位機制所得出的定位平均誤差為0.003m，定位準確率為99.99%。第二個場域位於台大醫學院五樓的空地，長寬尺寸為6.3m×6.3m。其定位平均誤差為0.007m，定位準確率為99.99%。實驗結果顯示，本論文所提出的機制是一個可靠且穩定的室內定位方法。

As artificial intelligence and the Internet of Things continue to develop, more legacy applications and services will become smart and automatic in the future. For example, document delivery in offices and blood sample transfer between laboratories in a hospital can be done by automatic robots. Smart docents can provide guided tours of exhibitions for visitors in museums. These indoor location-based services all rely on indoor positioning systems with high positioning accuracy.
In this study, we propose a double majority decision mechanism that leverages RSSI and magnetic anomalies as features of indoor environments. A Support Vector Machine (SVM) is also employed to improve accuracy. The SVM trained three models, i.e. RSSI Model, Magnetic Model and Hybrid Model for RSSI, magnetic fields and hybrid of them, respectively. For each tested sample, the first majority decision was made based on the results of the three models to determine the predicted location of the sample in the online phase. S number of samples were collected at each testing point. The predicted locations of the S samples are the basis for the second majority decision-making to get the final estimated locations at the testing points.
The experiments with the proposed system are conducted in two indoor scenarios: a 6.3m×2.4m corridor on the second floor of National Taiwan University Hospital, for which the average error distance and the accuracy were 0.003m and 99.99%, and a 6.3m×6.3m opening area on the fifth floor of the College of Medicine, National Taiwan University, for which the average error distance and the accuracy were 0.007m and 99.99%. The experimental results confirm that the Double Majority Decision mechanism is a stable and reliable indoor positioning method.

[1] Amazon Prime Air. [Online]. Available: https://www.amazon.com/Amazon-Prime-Air/b?ie=UTF8&node=8037720011
[2] Galileo (satellite navigation). [Online]. Available: https://en.wikipedia.org/wiki/Galileo_(satellite_navigation)
[3] BeiDou Navigation Satellite System. [Online]. Available: https://en.wikipedia.org/wiki/BeiDou_Navigation_Satellite_System
[4] R. Carvalho, Shan-Ho Yang, Yao-Hua Ho and Ling-Jyh Chen, “Indoor Localization Using FM and DVB-T Signals,” IEEE Annual Consumer Communications & Networking Conference, pp. 862-867, 2016.
[5] A. Syberfeldt, M. Ayani, M. Holm, L. Wang and R. Lindgren-Brewster, “Localizing Operators in the Smart Factory: A Review of Existing Techniques and Systems,” International Symposium on Flexible Automation, pp. 179-185, 2016.
[6] H. Sharifi, A. Kumar, F. Alam and K. M. Arif, “Indoor Localization of Mobile Robot with Visible Light Communication,” IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, pp. 1-6, 2016.
[7] A. Yassin, Y. Nasser, M. Awad, A. Al-Dubai, R. Liu,C. Yuen, R. Raulefs and E. Aboutanios,” Recent Advances in Indoor Localization: A Survey on Theoretical Approaches and Applications,” IEEE Communications Surveys & Tutorials, 2017.
[8] A. A. Wahab, A. Khattab, Y. A. Fahmy, “Two-way TOA with Limited Dead Reckoning for GPS-Free Vehicle Localization Using Single RSU,” International Conference on ITS Telecommunications, pp.244-249, 2013.
[9] Rida, M. E., Fuqiang, Liu, Jadi, Y., Algawhari, A. A. A., Askourih, A., “Indoor Location Position Based on Bluetooth Signal Strength,” Information Science and Control Engineering, pp. 769-773, 2015.
[10] Palumbo, F., Barsocchi, P., Chessa, S., Augusto, J. C., “A Stigmergic Approach to Indoor Localization Using Bluetooth Low Energy Beacons,” Advanced Video and Signal Based Surveillance, pp.1-6, 2015.
[11] N. Pirzada, M. Y. Nayan, M. F. Hassan, F. Subhan, H. Sakidin, “WLAN Location Fingerprinting Technique for Device-Free Indoor Localization System,” International Conference on Computer and Information Sciences, pp. 650-655, 2016.
[12] A. Thaljaoui, T. Val, N. Nasri and D. Brulin,” BLE Localization Using RSSI Measurements and iRingLA,” IEEE International Conference on Industrial Technology (ICIT), pp. 2178-2183,2015.
[13] N. Kuxdorf-Alkirata, D. Brückmann, “Reliable and Low-cost Indoor Localization Based on Bluetooth Low Energy,” International Symposium on Wireless Systems within the Conferences on Intelligent Data Acquisition and Advanced Computing Systems, pp.92-96, 2016.
[14] S. He, S. H. Chan, “INTRI: Contour-based Trilateration for Indoor Fingerprint-based Localization,” IEEE Transactions on Mobile Computing, 2016.
[15] Chowdhury, T. I., Rahman, M. M., Parvez, S. A., Alam, A. K. M. M., Basher, A., Alam, A., Rizwan, S., “A Multi-step Approach for RSSI-based Distance Estimation Using Smartphones,” International Conference on Networking Systems and Security, pp.1-5, 2015.
[16] Y. Shu, Y. Huang, J. Zhang, P. Coue, P. Cheng, J. Chen and K. G. Shin,” Gradient-Based Fingerprinting for Indoor Localization and Tracking,” IEEE Transactions on Industrial Electronics, Vol. 63, No. 4, pp. 2424-2433, 2016.
[17] Faragher, R., Harle, R., “Location Fingerprinting with Bluetooth Low Energy Beacons,” IEEE Journal on Selected Areas in Communications, Vol. 33, No. 11, pp. 2418-2428, 2015.
[18] Lohan, E. S., Talvitie, J., Figueiredo e Silva, P., Nurminen, H., Ali-Loytty, S., Piche, R., “Received Signal Strength Models for WLAN and BLE-based Indoor Positioning in Multi-floor Buildings,” International Conference on Localization and GNSS, pp. 1-6, 2015.
[19] C. Dong, J. Ding, J. Lin, “Segmented Polynomial RSSI-LQI Ranging Modelling for ZigBee-based Positioning Systems,” Chinese Control Conference, pp. 8387-8390, 2016.
[20] Yoon, P. K., Zihajehzadeh, S., Bong-Soo, Kang, Park, E. J., “Adaptive Kalman Filter for Indoor Localization Using Bluetooth Low Energy and Inertial Measurement Unit,” International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 825-828, 2015.
[21] T. Li, J. Wang, Z. Chen, “Research of Indoor Localization based on Inertial Navigation Technology,” Chinese Control and Decision Conference, pp. 2860-2864, 2016.
[22] K. Yamazaki, K. Kato, K. Ono, H. Saegusa, K. Tokunaga, Y. Iida, S. Yamamoto, K. Ashiho, K. Fujiwara, N. Takahashi, “Analysis of Magnetic Disturbance due to Buildings,” IEEE Transactions on Magnetics, Vol. 39, No. 5, pp. 3226-3228, 2003.
[23] M. Angermann, M. Frassl, M. Doniec, B. J. Julian, P. Robertson, “Characterization of The Indoor Magnetic Field for Applications in Localization and Mapping,” International Conference on Indoor Positioning and Indoor Navigation, pp. 1-9, 2012.
[24] Seong-Eun, Kim, Yong, Kim, Jihyun, Yoon, Eung Sun, Kim, “Indoor Positioning System Using Geomagnetic Anomalies for Smartphones,” International Conference on Indoor Positioning and Indoor Navigation, pp.1-5, 2012.
[25] H. Xie, T. Gu, X. Tao, H. Ye, J. Lu, “A Reliability-Augmented Particle Filter for Magnetic Fingerprinting Based Indoor Localization on Smartphone,” IEEE Transactions on Mobile Computing, Vol.15, No.8, pp. 1877-1892, 2016.
[26] B. Kim, S. H. Kong, “A Novel Indoor Positioning Technique Using Magnetic Fingerprint Difference,” IEEE Transactions on Instrumentation and Measurement, Vol.65, No.9, pp.2035-2045, 2016.
[27] Y. Shu, C. Bo, G. Shen, C. Zhao, L. Li, F. Zhao, “Magicol: Indoor Localization Using Pervasive Magnetic Field and Opportunistic WiFi Sensing,” IEEE Journal on Selected Areas in Communications, Vol.33, No.7, pp.1443-1457, 2015.
[28] M. Zhang, W. Shen, J. Zhu, “WIFI and Magnetic Fingerprint Positioning Algorithm Based on KDA-KNN,” Chinese Control and Decision Conference, pp. 5409-5415, 2016.
[29] Cortes, C., Vapnik, V., “Support-vector Networks,” Machine Learning, Vol.20, Issue.3, pp.273-297, doi:10.1007/BF00994018, 1995.
[30] K. Tbarki, S. Ben Said, R. Ksantini, Z. Lachiri, “RBF Kernel Based SVM Classification for Landmine Detection and Discrimination,” International Image Processing, Applications and Systems, pp.1-6, 2016.
[31] Android API Guides-Bluetooth Low Energy. [Online]. Available: https://developer.android.com/guide/topics/connectivity/bluetooth-le.html
[32] Android API Guides – SensorEvent. [Online]. Available: https://developer.android.com/reference/android/hardware/SensorEvent.html
[33] Chih-Chung Chang and Chih-Jen Lin, “LIBSVM : A Library for Support Vector Machines.” ACM Transactions on Intelligent Systems and Technology, 2:27:1--27:27, 2011. Software available at http://www.csie.ntu.edu.tw/~cjlin/libsvm
[34] Xiaoqing, Lu, Haitao, Liu, Feng, Liu,” A Novel Algorithm for Enhancing Accuracy of Indoor Position Estimation,” World Congress on Intelligent Control and Automation, pp. 5528-5533, 2014.