請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/87191完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 謝宏昀 | zh_TW |
| dc.contributor.advisor | Hung-Yun Hsieh | en |
| dc.contributor.author | 楊大寬 | zh_TW |
| dc.contributor.author | Ta-Kuan Yang | en |
| dc.date.accessioned | 2023-05-18T16:16:49Z | - |
| dc.date.available | 2023-11-09 | - |
| dc.date.copyright | 2023-05-10 | - |
| dc.date.issued | 2023 | - |
| dc.date.submitted | 2023-02-14 | - |
| dc.identifier.citation | [1] J.-X. Liao, “An indoor localization system for cellular networks based on transfer learning with reduced cost on site survey,” no. 2021, pp. 1–143, 2021.
[2] W. Njima, M. Chafii, A. Chorti, R. M. Shubair, and H. V. Poor, “Indoor localization using data augmentation via selective generative adversarial net works,” IEEE Access, vol. 9, pp. 98 337–98 347, 2021. [3] G. Draft, “Feasibility study on new services and markets technology enablers stage 1 (release 14),” International Telecommunication Union, 2016. [4] F. Zafari, A. Gkelias, and K. K. Leung, “A survey of indoor localization sys tems and technologies,” IEEE Communications Surveys & Tutorials, vol. 21, no. 3, pp. 2568–2599, 2019. [5] H. Zhang, Z. Zhang, S. Zhang, S. Xu, and S. Cao, “Fingerprint-based local ization using commercial lte signals: A field-trial study,” in 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall). IEEE, 2019, pp. 1–5. [6] L. Xiao, A. Behboodi, and R. Mathar, “A deep learning approach to finger printing indoor localization solutions,” in 2017 27th International Telecom munication Networks and Applications Conference (ITNAC). IEEE, 2017, pp. 1–7. [7] R. S. Sinha, S.-M. Lee, M. Rim, and S.-H. Hwang, “Data augmentation schemes for deep learning in an indoor positioning application,” Electronics, vol. 8, no. 5, p. 554, 2019. [8] R. S. Sinha and S.-H. Hwang, “Improved rssi-based data augmentation tech nique for fingerprint indoor localisation,” Electronics, vol. 9, no. 5, p. 851, 2020. [9] L. Chen, S. Zhang, H. Tan, and B. Lv, “Progressive rss data augmenter with conditional adversarial networks,” IEEE Access, vol. 8, pp. 26 975–26 983, 2020. [10] H. Rizk, M. Abbas, and M. Youssef, “Device-independent cellular-based in door location tracking using deep learning,” Pervasive and Mobile Computing, vol. 75, p. 101420, 2021. [11] F. Alhomayani and M. H. Mahoor, “Deep learning methods for fingerprint based indoor positioning: a review,” Journal of Location Based Services, vol. 14, no. 3, pp. 129–200, 2020. [12] H. Mukhtar, “Machine learning enabled-localization in 5g and lte us ing image classification and deep learning,” Ph.D. dissertation, Université d’Ottawa/University of Ottawa, 2021. [13] A. Blanco, N. Ludant, P. J. Mateo, Z. Shi, Y. Wang, and J. Widmer, “Perfor mance evaluation of single base station toa-aoa localization in an lte testbed,” in 2019 IEEE 30th annual international symposium on personal, Indoor and Mobile Radio Communications (PIMRC). IEEE, 2019, pp. 1–6. [14] Z. Yang, Z. Zhou, and Y. Liu, “From rssi to csi: Indoor localization via channel response,” ACM Computing Surveys (CSUR), vol. 46, no. 2, pp. 1– 32, 2013. [15] E. Goldoni, A. Savioli, M. Risi, and P. Gamba, “Experimental analysis of rssi-based indoor localization with ieee 802.15. 4,” in 2010 European Wireless Conference (EW). IEEE, 2010, pp. 71–77. [16] P. Kumar, L. Reddy, and S. Varma, “Distance measurement and error es timation scheme for rssi based localization in wireless sensor networks,” in 2009 Fifth international conference on wireless communication and sensor networks (WCSN). IEEE, 2009, pp. 1–4. [17] W. Stallings, Data and computer communications. Pearson Education India, 2007. [18] S.-K. T. Raphael, “Selecting robust features for cellular indoor localization in environments with human activities,” Master’s thesis, Jan 2022. [19] M. Aljumaily, “A survey on wifi channel state information (csi) utilization in human activity recognition,” 2016. [20] I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial networks,” 2014. [21] Z. Wang, Q. She, and T. E. Ward, “Generative adversarial networks in com puter vision: A survey and taxonomy,” ACM Computing Surveys (CSUR), vol. 54, no. 2, pp. 1–38, 2021. [22] M. Mirza and S. Osindero, “Conditional generative adversarial nets,” arXiv preprint arXiv:1411.1784, 2014. [23] P. Isola, J.-Y. Zhu, T. Zhou, and A. A. Efros, “Image-to-image translation with conditional adversarial networks,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 1125–1134. [24] S. Saxena and M. N. Teli, “Comparison and analysis of image-to-image gen erative adversarial networks: A survey,” arXiv preprint arXiv:2112.12625, 2021. [25] G. Oguntala, R. Abd-Alhameed, S. Jones, J. Noras, M. Patwary, and J. Ro driguez, “Indoor location identification technologies for real-time iot-based applications: An inclusive survey,” Computer Science Review, vol. 30, pp. 55–79, 2018. [26] N. Singh, S. Choe, and R. Punmiya, “Machine learning based indoor local ization using wi-fi rssi fingerprints: an overview,” IEEE Access, 2021. [27] T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785–794. [28] J. Friedman, T. Hastie, and R. Tibshirani, “Additive logistic regression: a statistical view of boosting (with discussion and a rejoinder by the authors),” The annals of statistics, vol. 28, no. 2, pp. 337–407, 2000. [29] J.-Y. Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image-to-image translation using cycle-consistent adversarial networks,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2223–2232. [30] eriklindernoren, “Keras-GAN,” 2021. Online Available at: https://github.com/eriklindernoren/Keras-GAN [31] L. Weng, “From gan to wgan,” arXiv preprint arXiv:1904.08994, 2019. [32] S. Vallender, “Calculation of the wasserstein distance between probability distributions on the line,” Theory of Probability & Its Applications, vol. 18, no. 4, pp. 784–786, 1974. [33] A. Ramdas, N. García Trillos, and M. Cuturi, “On wasserstein two-sample testing and related families of nonparametric tests,” Entropy, vol. 19, no. 2, p. 47, 2017. [34] Y. Wang, L. Zhang, and J. van de Weijer, “Ensembles of generative adver sarial networks,” 2016. [35] H. Zhang, M. Cisse, Y. N. Dauphin, and D. Lopez-Paz, “mixup: Beyond empirical risk minimization,” arXiv preprint arXiv:1710.09412, 2017. [36] S. Yun, D. Han, S. J. Oh, S. Chun, J. Choe, and Y. Yoo, “Cutmix: Reg ularization strategy to train strong classifiers with localizable features,” in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 6023–6032. [37] L. Van der Maaten and G. Hinton, “Visualizing data using t-sne.” Journal of machine learning research, vol. 9, no. 11, 2008. [38] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE transactions on image processing, vol. 13, no. 4, pp. 600–612, 2004. [39] M. F. Naeem, S. J. Oh, Y. Uh, Y. Choi, and J. Yoo, “Reliable fidelity and diversity metrics for generative models,” in International Conference on Ma chine Learning. PMLR, 2020, pp. 7176–7185. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/87191 | - |
| dc.description.abstract | 在基於機器學習的室內定位系統中，當環境改變時，原本為特定環境建立的模型可能不再能有效的運作。因此，我們必須重新收集目標場域中大量的數據，這需要消耗許多時間。因此，這種方法是不實際的。在我們的研究中，我們的目標是設計一個數據增強系統，以減少蒐集資料的時間成本；我們使用軟件定義無線電（SDR）硬體和開源的平台（OAI 5G）建立定位系統。我們可以通過這個系統收集基於LTE的特徵，並將它們放入機器學習模型中以預測用戶的坐標。在此之後，我們使用Cycle GAN進行數據增強，這通常用於改變圖像的風格。我們提出了一種基於Cycle GAN的方法，稱為Semi Cycle GAN，用於使用在新場域中的少量資料和使用在舊場域中的大量資料以產生類似於新場域的充足資料量。這種方法可以省下大量時間，並提高模型的精準度。它比 Cycle GAN 更適用於低維度資料。與Vanilla GAN相比，我們可以提高定位精準度，並從目標場域和源頭場域中使用數據。我們的研究可以通過上述方法提高資料使用率，並減少時間。此外，我們使用帶有不同評分機制的選擇方法以及數據混合技術，以提高模擬數據的質量和多樣性。總而言之，我們可以通過平均距離誤差（MDE）提高36％的定位性能。 | zh_TW |
| dc.description.abstract | In machine learning-based indoor localization systems, the initial model built for a particular environment may no longer be useful when the environment is altered. As a result, we must recollect a big amount of data, which takes time. This approach, however, is ineffective. In our work, we aim to design a data augmenta tion system to reduce the time cost of the site survey; we used the Software-Defned Radio (SDR) hardware platform and open-source software platform (OAI 5G) to build the localization system. We can collect LTE-based features by this system and ft them into a machine-learning model to predict the user’s coordinates. Af ter that, we used the cycle Generative Adversarial Network (cycleGAN) for data augmentation, which is usually used to change the style of images. We propose a method based on Cycle GAN called Semi Cycle GAN to utilize a small amount of data in a new domain and a large amount of data in an old domain to produce a suffcient amount of data similar to the new domain. This method can save lots of time and incline the model’s accuracy. It is more suitable for low-dimension data than Cycle GAN. Compared with Vanilla GAN, we can enhance localization accuracy and utilize data from both the target and source domains. Our research can increase the data usage rate and decrease time by the above methods. In addition, we use the selection method with different scoring mechanisms and data mixing techniques to increase the quality and diversity of simulation data. In conclusion, we can improve the localization performance by 36% in Mean Distance Error (MDE). | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2023-05-18T16:16:49Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2023-05-18T16:16:49Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | TABLE OF CONTENTS
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii CHAPTER 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . 1 CHAPTER 2 BACKGROUND AND RELATED WORK . . . . 5 2.1 Localization Methods . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1.1 Received Signal Strength Indicator (RSSI) . . . . . . . . 5 2.1.2 Channel State Information (CSI) . . . . . . . . . . . . . 6 2.1.3 Fingerprinting Method . . . . . . . . . . . . . . . . . . . 7 2.1.4 Comparison and Conclusion . . . . . . . . . . . . . . . . 7 2.2 Localization Evaluation Methods . . . . . . . . . . . . . . . . . . 8 2.3 Generated Adversarial Network(GAN) . . . . . . . . . . . . . . . 9 2.3.1 Vanilla GAN (GAN) . . . . . . . . . . . . . . . . . . . . 9 2.3.2 Conditional GAN (cGAN) . . . . . . . . . . . . . . . . . 12 2.3.3 Image-to-Image Translation with Conditional Adversarial Nets (Pix2Pix) . . . . . . . . . . . . . . . . . . . . . . . 13 2.3.4 Comparison of a Different kind of GANs . . . . . . . . . 14 2.4 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 2.4.1 Data Augmentation Without Using Machine Learning . . 16 2.4.2 Data Augmentation Using Machine Learning . . . . . . . 17 CHAPTER 3 SYSTEM MODEL . . . . . . . . . . . . . . . . . . . 21 3.1 System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.2 LTE Signal Features . . . . . . . . . . . . . . . . . . . . . . . . . 22 3.3 Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.4 Machine Learning Models . . . . . . . . . . . . . . . . . . . . . . 25 3.4.1 Target Problem . . . . . . . . . . . . . . . . . . . . . . . 25 3.5 Upper Bound and Lower Bound . . . . . . . . . . . . . . . . . . 26 3.5.1 Upper bound . . . . . . . . . . . . . . . . . . . . . . . . 26 3.5.2 Lower bound . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.6 Fine-tuning Method . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.6.1 Localization Model Description . . . . . . . . . . . . . . 26 3.6.2 Conservative Training . . . . . . . . . . . . . . . . . . . . 29 3.6.3 Evaluation of Fine-tuning Model . . . . . . . . . . . . . . 31 3.7 GAN-based Method . . . . . . . . . . . . . . . . . . . . . . . . . 31 CHAPTER 4 DATA AUGMENTATION . . . . . . . . . . . . . . 33 4.1 Cycle GAN Method . . . . . . . . . . . . . . . . . . . . . . . . . 33 4.1.1 Objective Loss Function . . . . . . . . . . . . . . . . . . 35 4.1.2 Generator Network . . . . . . . . . . . . . . . . . . . . . 36 4.1.3 Problem of Cycle GAN in Our Work . . . . . . . . . . . 37 4.2 Semi Cycle GAN Method (SCG) . . . . . . . . . . . . . . . . . . 38 4.2.1 Objective Loss Function . . . . . . . . . . . . . . . . . . 39 4.2.2 Complexity Analysis . . . . . . . . . . . . . . . . . . . . 41 4.3 Selection Method . . . . . . . . . . . . . . . . . . . . . . . . . . 42 4.3.1 Scoring by Discriminator . . . . . . . . . . . . . . . . . . 43 4.3.2 Scoring by the Wasserstein Distance . . . . . . . . . . . . 46 4.4 Mixing Data from Different Methods . . . . . . . . . . . . . . . 49 4.4.1 Standard ensemble of GAN . . . . . . . . . . . . . . . . . 49 4.4.2 Mix-up . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4.4.3 Cut-mix . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.6 Evaluation Metrics . . . . . . . . . . . . . . . . . . . . . . . . . 53 4.6.1 t-Distributed Stochastic Neighbor Embedding (t-SNE) . 54 4.6.2 Density and Coverage . . . . . . . . . . . . . . . . . . . . 56 CHAPTER 5 PERFORMANCE EVALUATION . . . . . . . . . 58 5.1 Evaluation of GAN-based Method . . . . . . . . . . . . . . . . . 58 5.1.1 Evaluation of Vanilla GAN Method . . . . . . . . . . . . 58 5.1.2 Evaluation of Semi Cycle GAN Method . . . . . . . . . . 61 5.2 Evaluation of Selected Semi Cycle GAN . . . . . . . . . . . . . . 65 5.2.1 t-SNE . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 5.2.2 Model Performance Evaluation . . . . . . . . . . . . . . . 65 5.2.3 Scoring by Wasserstein distance . . . . . . . . . . . . . . 70 5.2.4 Discussion of the number of simulation data . . . . . . . 70 5.2.5 Evaluation using Density and Coverage . . . . . . . . . . 75 5.3 Evaluation of Data Mixing Method . . . . . . . . . . . . . . . . 75 5.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 CHAPTER 6 CONCLUSION AND FUTURE WORK . . . . . 84 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 | - |
| dc.language.iso | zh_TW | - |
| dc.title | 室內定位系統跨領域訓練之資料擴增方法研究 | zh_TW |
| dc.title | Reducing Site-Survey Cost for Cellular Indoor Localization via Data Augmentation | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 111-1 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 吳沛遠;馮輝文 | zh_TW |
| dc.contributor.oralexamcommittee | Pei-Yuan Wu;Huei-Wen Ferng | en |
| dc.subject.keyword | 室內定位,跨領域訓練,資料擴增, | zh_TW |
| dc.subject.keyword | indoor localization,cross domain,data augmentation,Generative Adversarial Network, | en |
| dc.relation.page | 88 | - |
| dc.identifier.doi | 10.6342/NTU202300465 | - |
| dc.rights.note | 同意授權(全球公開) | - |
| dc.date.accepted | 2023-02-15 | - |
| dc.contributor.author-college | 電機資訊學院 | - |
| dc.contributor.author-dept | 電信工程學研究所 | - |
| 顯示於系所單位： | 電信工程學研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-111-1.pdf | 2.89 MB | Adobe PDF | 檢視/開啟 |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。