臺灣博碩士論文加值系統

本論文提出利用深度學習(Deep Learning, DL)中的Pix2Pix生成對抗網路(Generative Adversarial Network, GAN)技術來改善麥克生(Michaelson Interference,MI)干涉感測器訊號解調結果，在傳統相位生成載波(Phase Generated Carrier, PGC)中的微分交叉相乘(Differentiation and Cross Multiplication, DCM)解調方案上，有相位延遲與調變深度問題，導致結果不符合預期，會出現波型失真的現象，所以我們利用深度學習的方法解決傳統干涉光纖感測器解調的相位延遲與調變深度問題。比起傳統硬體解調方法，在相位解調方面，我們提出利用深度學習的技術能夠解調出兩個振動訊號的麥克生感測器，讓訊號解調的過程更加準確和高效，這相對於傳統硬體解調更能有效的解調麥克生感測器的訊號。我們的研究，經過實驗結果顯示，使用基於生成對抗網路的深度學習解調，能夠有效地解調麥克生感測器訊號，這對光纖感測領域有很大的意義，也為解調問題提供了一條新的思路。

This paper proposes the use of Pix2Pix Generative Adversarial Network (GAN) technology, a deep learning (DL) technique, to improve the signal demodulation results of Michaelson Interference (MI) interferometric sensors. In traditional Phase Generated Carrier (PGC) schemes employing Differentiation and Cross Multiplication (DCM) for demodulation, issues such as phase delays and modulation depths often lead to unexpected results and waveform distortions. Therefore, we employ deep learning methods to address these problems in the demodulation of traditional interferometric fiber sensors. Compared to conventional hardware demodulation methods, our proposed deep learning-based technique for phase demodulation enables accurate and efficient decoding of two oscillating signals from the Michaelson interferometric sensor. This approach proves to be more effective in demodulating the sensor's signals compared to traditional hardware-based demodulation methods. Our research, supported by experimental results, demonstrates that utilizing deep learning-based demodulation using Generative Adversarial Networks effectively decodes the signals from Michaelson interferometric sensors. This advancement holds significant implications for the field of fiber-optic sensing, providing a new perspective in addressing demodulation challenges.

摘要...i
Abstract...ii
誌謝...iii
目錄...iv
表目錄...vi
圖目錄...vii
第一章 緒論...1
1.1 前言...1
1.2 研究動機與目的...2
1.3 文獻探討...3
1.4 論文架構...5
第二章 光纖感測器研究背景與理論介紹...6
2.1 光纖感測器基本介紹...6
2.1.1 光纖感測器的分類...6
2.1.2 光纖感測器的相位變化...7
2.2 光纖感測理論分析...10
2.3 光纖感測解調技術...12
2.3.1 反正切演算法解調技術...12
2.3.2 微分交叉相乘法解調技術...13
2.3.3 相關解調技術...14
第三章 Pix2Pix GAN背景與原理介紹和文獻探討...16
3.1 Pix2Pix GAN 介紹...16
3.2 生成器架構...18
3.3 鑑別器架構...20
3.4 卷積層結構...21
第四章 實驗內容與結果...22
4.1 實驗架構...22
4.1.1 實驗前置流程...22
4.1.2 使用生成對抗網路技術於麥克生感測器訊號解調系統...24
4.2 實驗結果...25
4.2.1 PGC-DCM與PGC-Atan解調技術實驗...25
4.2.2 一個震動訊號的實驗...28
4.2.3 當相位振幅∆l1=1μm的兩個待側震動訊號實驗...30
4.2.4 當相位振幅∆l1=2μm的兩個待側震動訊號實驗...35
4.2.5 當相位振幅∆l1=3μm的兩個待測震動訊號實驗...40
第五章 結論...45
參考文獻...46
Extended Abstract...49

[1]詹汝平, "麥克生干涉儀感測器之 數位式相位生成載波解調系統之研究," 碩士, 光電工程系光電與材料科技碩士班, 國立虎尾科技大學, 雲林縣, 2019.
[2]S. Yuan, Y. Hu, Q. Hao, and S. Zhang, "High-accuracy phase demodulation method compatible to closed fringes in a single-frame interferogram based on deep learning," Optics Express, vol. 29, no. 2, pp. 2538-2554, 2021/01/18 2021.
[3]L. Kong, K. Cui, J. Shi, M. Zhu, and S. Li, "1D Phase Unwrapping Based on the Quasi-Gramian Matrix and Deep Learning for Interferometric Optical Fiber Sensing Applications," Journal of Lightwave Technology, vol. 40, no. 1, pp. 252-261, 2022.
[4]J. Hu and D. Li, "Simulation and testing of a noise-limited demodulation system for a fiber-optic hydrophone system based on a Michelson interferometer," in 2016 IEEE/OES China Ocean Acoustics (COA), 2016, pp. 1-5.
[5]L. Zhang, H. Zhang, Y. Jiang, and Z. Wu, "Intelligent and Reliable Deep Learning LSTM Neural Networks-Based OFDM-DCSK Demodulation Design," IEEE Transactions on Vehicular Technology, vol. 69, no. 12, pp. 16163-16167, 2020.
[6]A. D. Kersey, A. Dandridge, and A. B. Tveten, "Dependence of visibility on input polarization in interferometric fiber-optic sensors," (in eng), Opt Lett, vol. 13, no. 4, pp. 288-90, Apr 1 1988.
[7]A. D. Kersey, M. J. Marrone, and A. Dandridge, "Observation of input-polarization-induced phase noise in interferometric fiber-optic sensors," Optics Letters, vol. 13, no. 10, pp. 847-849, 1988/10/01 1988.
[8]A. D. Kersey, M. J. Marrone and M. A. Davis, "Polarisation-insensitive fibre optic Michelson interferometer," in Electronics Letters, vol. 27, no. 6, pp. 518-520, 1991.
[9]A. D. Kersey and M. J. Marrone, "Input-polarisation scanning technique for overcoming polarisation-induced signal fading in interferometric fibre sensors," Electronics Letters, vol. 24, no. 15, pp. 931-933Available: https://digital-library.theiet.org/content/journals/10.1049/el_19880634
[10]A. D. Kersey, M. J. Marrone, and A. Dandridge, "Analysis of input-polarization-induced phase noise in interferometric fiber-optic sensors and its reduction using polarization scrambling," Journal of Lightwave Technology, vol. 8, no. 6, pp. 838-845, 1990.
[11]D. A. Jackson, R. Priest, A. Dandridge, and A. B. Tveten, "Elimination of drift in a single-mode optical fiber interferometer using a piezoelectrically stretched coiled fiber," Applied Optics, vol. 19, no. 17, pp. 2926-2929, 1980/09/01 1980.
[12]S. K. Sheem, T. G. Giallorenzi, and K. Koo, "Optical techniques to solve the signal fading problem in fiber interferometers," Applied Optics, vol. 21, no. 4, pp. 689-693, 1982/02/15 1982.
[13]K. P. Koo, A. B. Tveten, and A. Dandridge, "Passive stabilization scheme for fiber interferometers using (3×3) fiber directional couplers," Applied Physics Letters, vol. 41, no. 7, pp. 616-618, 1982.
[14]A. Dandridge, A. B. Tveten, and T. G. Giallorenzi, "Homodyne Demodulation Scheme for Fiber Optic Sensors Using Phase Generated Carrier," IEEE Transactions on Microwave Theory and Techniques, vol. 30, no. 10, pp. 1635-1641, 1982.
[15]J. H. Cole, B. A. Danver, and J. A. Bucaro, "Synthetic-Heterodyne Interferometric Demodulation," IEEE Transactions on Microwave Theory and Techniques, vol. 30, no. 4, pp. 540-543, 1982.
[16]吳振彰, "干涉式光纖感應器PGC 解調系統的靈敏歸一化及FPGA 數位訊號處理技術," 碩士, 光電與通訊工程學系, 國立高雄師範大學, 高雄市, 2011.
[17]P. Isola, J. Y. Zhu, T. Zhou, and A. A. Efros, "Image-to-Image Translation with Conditional Adversarial Networks," in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 5967-5976.
[18]A. Mino and G. Spanakis, "LoGAN: Generating Logos with a Generative Adversarial Neural Network Conditioned on Color," in 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA), 2018, pp. 965-970.
[19]O. Ronneberger, P. Fischer, and T. Brox, "U-Net: Convolutional Networks for Biomedical Image Segmentation," in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Cham, 2015, pp. 234-241: Springer International Publishing.
[20]Q. Hao et al., "High signal-to-noise ratio reconstruction of low bit-depth optical coherence tomography using deep learning," (in eng), J Biomed Opt, vol. 25, no. 12, Nov 2020.
[21]U. Demir and G. Unal, "Patch-Based Image Inpainting with Generative Adversarial Networks," 03/20 2018.
[22]廖柏閎, "使用生成對抗網路技術於頻域光學同調斷層掃描圖像加強之研究," 碩士, 光電工程系光電與材料科技碩士班, 國立虎尾科技大學, 雲林縣, 2022.
[23]N. Wu, X. Wang, B. Lin, and K. Zhang, "A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems," IEEE Access, vol. 7, pp. 110197-110204, 2019.
[24]H. Ye, L. Liang, G. Y. Li, and B. H. Juang, "Deep Learning-Based End-to-End Wireless Communication Systems With Conditional GANs as Unknown Channels," IEEE Transactions on Wireless Communications, vol. 19, no. 5, pp. 3133-3143, 2020.
[25]A. Ahmad, S. Agarwal, S. Darshi, and S. Chakravarty, "DeepDeMod: BPSK Demodulation Using Deep Learning Over Software-Defined Radio," IEEE Access, vol. 10, pp. 115833-115848, 2022.
[26]M. Chojnacki and N. Palka, "Demodulation of output signals from unbalanced fibre optic Michelson interferometer," in Modern Problems of Radio Engineering, Telecommunications and Computer Science (IEEE Cat. No.02EX542), 2002, pp. 249-250.
[27]Y. Liu, B. Qi, and D. E. Winder, "Faraday Michelson Interferometers for Signal Demodulation of Fiber-Optic Sensors," Journal of Lightwave Technology, vol. 39, no. 8, pp. 2552-2558, 2021.
[28]康晉瑜, "麥克生干涉儀感測器之水下振動感測器系統之研究," 碩士, 光電工程系光電與材料科技碩士班, 國立虎尾科技大學, 雲林縣, 2020.