基於位置的服務(location-based service; LBS)已經逐漸進入人們的日常生活中，但全球定位系統(Global Positioning System; GPS)卻受到雜訊、屏蔽效應等干擾，無法在室內提供精準的定位，因此如何在室內精準的定位逐漸成為研究的熱門議題。本研究利用最近火紅的類神經網路(neural network; NN)結合通道狀態資訊(channel state information; CSI)達到高精準的室內定位。通道為訊號在無線通訊中從發射端到接收端的訊號路徑，影響通道狀態的因素與周圍環境息息相關，周圍環境可能會導致訊號散射、衰落等，或者因為距離導致訊號逐漸衰減。多載波通訊系統中，鄰近子載波訊號的通道狀態資訊有高度的相關性，卷積神經網路對於此種輸入資料有很好的學習力。據此，本研究提出改良式的卷積神經網路，閥式(gated)卷積神經網路對於相關的通道狀態資訊又有更好的擬合性，使室內定位有更高的精準度，其平均誤差可以達到7cm 左右。本研究還比較了在不同天線數量下定位的精準度，就算只有兩根天線，定位精準度平均仍可達到28cm 左右。最後，本研究也實驗了不同訓練資料集對定位的影響，當測試資料集中的位置不在訓練資料集中，類神經網路對通道訊號狀態及位置函數的擬合性就未如預期的好。

Location-based service (LBS) has become important part in people’s life in recent years, but the global positioning system(GPS) restricted by the shielding effect and noise isn’t available in indoor environments. Therefore, how to accurtely locate in indoor environment has become a popuplar issue in recent years. This thesis uses the channel state information(CSI) combined with convolutional neural network(CNN) to achieve a highly accurate indoor positioning. The CSI refers to known channel properties of a communication link in wireless communications. This information describes how a signal propagates from the transmitter to the receiver and represents the combined effect of, for example, scattering, fading, and power decay with distance. In multi-carrier comunnication systems, the CSI of adjacent subcarriers has high correlation, and CNN is promising to learn the relationship of these input information. Beyond that, we propose and improve CNN, i.e., the gated CNN, which has more talent to locate in indoor environments than traditional CNNs. Experimental results show that the proposed gated CNN can achieve an accuracy of less than 0.08 m with 16 antennas. We aslo demonstrate the accuracy under different number of antennas. With only 2 antennas, the accuracy can still be within 0.3 m.

[1] Z. Yang, Z. Zhou, and Y. Liu, “From RSSI to CSI: Indoor localization via channel response,” ACM Comput. Surveys, vol. 46, no. 2, pp. 1-32, 2013.
[2] K. Wu, J. Xiao, Y. Yi, D. Chen, X. Luo and L. M. Ni, “CSI-Based Indoor Localization,” IEEE Transactions on Parallel and Distributed Systems, vol. 24, no. 7, pp. 1300-1309, 2013.
[3] Y. Chapre, A. Ignjatovic, A. Seneviratne and S. Jha, “CSI-MIMO: Indoor Wi-Fi fingerprinting system,” 39th Annual IEEE Conference on Local Computer Networks, pp. 202-209, 2014.
[4] Y. Zhang, L. Deng and Z. Yang, “Indoor positioning based on FM radio signals strength,” 2017 First International Conference on Electronics Instrumentation & Information Systems (EIIS), pp. 1-5, 2017.
[5] B. Molina, E. Olivares, C. E. Palau and M. Esteve, “A Multimodal Fingerprint-Based Indoor Positioning System for Airports,” IEEE Access, vol. 6, pp. 10092-10106, 2018.
[6] R. K. Yadav, B. Bhattarai, H. Gang and J. Pyun, “Trusted K Nearest Bayesian Estimation for Indoor Positioning System,” IEEE Access, vol. 7, pp. 51484-51498, 2019.
[7] H. Liu, B. Chen, P. Tseng, K. Feng and T. Wang, “Map-Aware Indoor Area Estimation with Shortest Path Based on RSS Fingerprinting,” 2015 IEEE 81st Vehicular Technology Conference (VTC Spring), pp. 1-5, 2015.
[8] Yifei Jiang, Yun Xiang, Xin Pan, Kun Li, Qin Lv, Robert P. Dick, Li Shang, and Michael Hannigan, “Hallway based automatic indoor floorplan construction using room fingerprints,” Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing(UbiComp '13), pp. 315-324, 2013
[9] G. Shen, Z. Chen, P. Zhang, T. Moscibroda, and Y. Zhang, “Walkie-Markie: Indoor pathway mapping made easy,” Proc. USENIX NSDI, pp. 85–98, 2013.
[10] K. Wu, J. Xiao, Y. Yi, M. Gao, and L. M. Ni, “Fila: Fine-grained indoor localization,” 2012 Proceedings IEEE INFOCOM, pp. 2210-2218, 2012.
[11] G. Wu and P. Tseng, “A Deep Neural Network-Based Indoor Positioning Method using Channel State Information,” 2018 International Conference on Computing, Networking and Communications (ICNC),pp. 290-294, 2018,.
[12] Minsky, Marvin L., Papert, Seymour A. and Bottou, Léon (FRW). Perceptrons: An Introduction to Computational Geometry. America: Mit Pr, 1969.
[13] N. Aloysius and M. Geetha, “A review on deep convolutional neural networks,” 2017 International Conference on Communication and Signal Processing (ICCSP), pp. 0588-0592, 2017.
[14] 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.
[15] A. Krizhevsky, I. Sutskever, G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Proc. Adv. Neural Inf. Process. Syst., vol. 1, pp. 1097-1105, 2012.
[16] K. Simonyan and A. Zisserman, “Very deep convolutional networks for largescale image recognition,” arXiv:1409.1556, 2014.
[17] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, A. Rabinovich, “Going deeper with convolutions,” arXiv:1409.4842, 2014.
[18] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” arXiv:1512.03385, 2015.
[19] Maximilian Arnold, Jakob Hoydis, Stephan ten Brink, “Novel Massive MIMO Channel Sounding Data Applied to Deep Learning-based Indoor Positioning,” arXiv:1810.04126, 2019
[20] W. Kang, S. Nam, Y. Han, and S. Lee, “Improved heading estimation for smartphone-based indoor positioning systems,” Proc. IEEE 23rd Int. Symp. Pers., IndoorMobile Radio Commun.-(PIMRC), pp. 2449-2453, 2012.
[21] L. Lkovi£ová, P. Kajánek, A. Kopácik, “Pedestrian indoor positioning and tracking using smartphone sensors, step detection and map matching algorithm,” Geod. List, vol. 70, no. 1, pp. 1-24, 2016.
[22] Li H, Chen X, Jing G, Wang Y, Cao Y, Li F, Zhang X, Xiao H. “An Indoor Continuous Positioning Algorithm on the Move by Fusing Sensors and Wi-Fi on Smartphones,” Sensors (Basel), vol. 15, no. 12, pp. 31244-31267, 2015.
[23] J. Miranda, R. Abrishambaf, T. Gomes, P. Gonçalves, J. Cabral, A. Tavares and J. Monteiro, “Path loss exponent analysis in Wireless Sensor Networks: Experimental evaluation,” Proc. 11th IEEE Int. Conf. Ind. Inform. (INDIN), pp. 54-58, 2013.
[24] J. Luo and X. Zhan, “Characterization of smart phone received signal strength indication for WLAN indoor positioning accuracy improvement,” JNW., vol. 9, pp. 739-746, 2014.
[25] Q. Wang, R. Sun, X. Zhang, Y. Sun, and X. Lu, “Bluetooth positioning based on weighted K-nearest neighbors and adaptive bandwidth mean shift,” International Journal of Distributed Sensor Networks, vol. 13, no. 5, 2017.
[26] M. S. Lee, H. Ju, and C. G. Park, “Map assisted PDR/Wi-Fi fusion for indoor positioning using smartphone,” Int. J. Control, vol. 15, no. 2, pp. 627-639, 2017.
[27] S. Venkatraman, J. Caffery, and H. R. You, “A novel TOA location algorithm using LOS range estimation for NLOS environments,” IEEE Transactions on Vehicular Technology, vol. 53, no 5, pp. 1515-1524, 2004.
[28] N. Patwari,J. N. Ash,S. Kyperountas, A. O. Hero, R. L. Moses, and N.S. Correal, “Locating the nodes: cooperative localization in wireless sensor networks,” IEEE Signal processing magazine, vol. 22, no.4, pp. 54-69, 2005.
[29] H. Liu, H. Darabi, P. Banerjee, and J. Liu, “Survey of wireless indoor positioning techniques and systems,” IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 37, no. 6, pp. 1067-1080, 2007.
[30] P. Corral, E. Peña, R. Garcia, V. Almenar, and A. D. C. Lima, “Distance estimation system based on ZigBee,” 2008 IEEE 10th International Symposium on Spread Spectrum Techniques and Applications, pp. 817-820, 2008.
[31] S. He, and S.-H. Gary Chan, “Wi-Fi fingerprint-based indoor positioning: Recent advances and comparisons,” IEEE Communications Surveys & Tutorials, vol. 18, no .1, pp. 466-490, 2016.
[32] Yann N., Angela, Michael and, David, “Language Modeling with Gated Convolutional Networks,” arXiv:1612.08083, 2016
[33] A. Graves, A. Mohamed and G. Hinton, “Speech recognition with deep recurrent neural networks,” 2013 IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 6645-6649, 2013.