臺灣博碩士論文加值系統

近年來，隨著硬體圖形處理器的快速發展，使得深度學習這項技術能夠容易且廣泛被使用，更衍伸出機器視覺、語音識別及自然語言處理等多項應用，展現出深度學習具有一定的潛能和可靠度。對此我們針對光通訊領域，利用深度學習技術設計了一種線性區塊碼解碼器，以提升解碼器效能。本文研究了基於深度學習的線性區塊碼解碼器於光分碼多重擷取網路中的應用。光分碼多重擷取網路是一種利用光正交碼來實現多使用者同時非同步傳輸的技術，雖然具有彈性和安全性的優點，但也面臨著多重使用者干擾和位元錯誤率較高的挑戰。為了提升光分碼多重擷取網路的效能，本文採用了深度學習的方法來設計一種線性區塊碼解碼器，該解碼器能夠自行學習從接收訊號中，恢復原始資料的最佳方式，並利用模擬資料來驗證其解碼效能。根據模擬結果顯示，使用深度神經網路搭建的線性區塊碼解碼器能夠在不同的訊號雜訊比和使用者數量下，都能夠優於傳統的解碼器，同時也展現了深度學習在光通訊領域的潛力。

In recent years, the rapid development of hardware graphics processors has led to the easy and widespread use of deep learning technology. This has resulted in the emergence of various applications such as machine vision, speech recognition, and natural language processing, demonstrating the potential and reliability of deep learning. In this regard, a linear block code decoder has been designed using deep learning technology in the field of optical communication to improve decoder performance. This thesis focuses on studying the application of a deep learning-based linear block code decoder in optical code division multiple access (OCDMA) networks. OCDMA technology uses optical orthogonal codes to achieve simultaneous asynchronous transmission of multiple users. Although it has the advantages of flexibility and security, it also faces challenges such as multi-user interference and high bit error rates. To improve the performance of OCDMA networks, this thesis adopts a deep learning approach to design a linear block code decoder that can learn the best way to recover original data from received signals. According to simulation results, a linear block code decoder built using deep neural networks outperforms traditional decoders under different signal-to-noise ratios and user numbers. This demonstrates the potential of deep learning in the field of optical communication and its ability to improve the performance of OCDMA networks.

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 動機與目的......2
1.3 論文架構......2
第二章 光分碼多重擷取網路與線性區塊碼之探討......3
2.1 光分碼多重擷取網路......3
2.1.1 MWC編碼SAC-OCDMA網路......3
2.1.2 網路參數分析......4
2.2 線性區塊碼......6
2.2.1 低密度奇偶校驗碼......6
2.2.2 極化碼......7
2.3 傳輸與解碼......9
2.3.1 調變與雜訊方法......9
2.3.2 傳統解碼演算法......11
第三章 神經網路介紹與結構設計......12
3.1 深度類神經網路......12
3.1.1 多層感知器......12
3.1.2 激勵函數......14
3.1.3 損失函數......15
3.1.4 最佳化器......16
3.2 模擬架構設計......17
3.2.1 神經網路層設計......17
3.2.2 神經網路訓練流程......18
3.2.3 神經網路測試流程......19
3.2.4 效能評估方法......20
第四章 數據模擬分析與討論......22
4.1 光參數資料採樣率模擬......23
4.2 質數碼調整模擬......32
4.3 接收端有效光功率模擬......42
第五章 結論......48
參考文獻......49

[1]J. Hui, "Pattern Code Modulation and Optical Decoding - A Novel Code-Division Multiplexing Technique for Multifiber Networks," IEEE J. Sel. Areas Commun., vol. 3, no. 6, pp. 916-927, 1985, doi: 10.1109/JSAC.1985.1146265.
[2]M. Kavehrad and D. Zaccarin, "Optical code-division-multiplexed systems based on spectral encoding of noncoherent sources," J. Lightwave Technol., vol. 13, no. 3, pp. 534-545, 1995, doi: 10.1109/50.372451.
[3]W. Zou, H. M. H. Shalaby, and H. Ghafouri-Shiraz, "Modified quadratic congruence codes for fiber Bragg-grating-based spectral-amplitude-coding optical CDMA systems," J. Lightwave Technol., vol. 19, no. 9, pp. 1274-1281, 2001, doi: 10.1109/50.948274.
[4]C. C. Yang, "Optical CDMA Passive Optical Network Using Prime Code With Interference Elimination," IEEE Photonics Technol. Lett., vol. 19, no. 7, pp. 516-518, 2007, doi: 10.1109/LPT.2007.893579.
[5]J.-F. Huang, C.-C. Yang, and C.-M. Huang, "On Analyzing Quasi-Cyclic LDPC Codes Over Modified Welch–Costas-Coded Optical CDMA System," J. Lightwave Technol., vol. 27, no. 12, pp. 2150-2158, 2009/06/15 2009. [Online]. Available: https://opg.optica.org/jlt/abstract.cfm?URI=jlt-27-12-2150.
[6]R. Gallager, "Low-density parity-check codes," IRE Trans. Inf. Theory, vol. 8, no. 1, pp. 21-28, 1962, doi: 10.1109/TIT.1962.1057683.
[7]D. J. C. MacKay and R. M. Neal, "Near Shannon limit performance of low density parity check codes," Electron. Lett., vol. 32, no. 18, pp. 1645-1646. [Online]. Available: https://digital-library.theiet.org/content/journals/10.1049/el_19961141
[8]C. Sae-Young, G. D. Forney, T. J. Richardson, and R. Urbanke, "On the design of low-density parity-check codes within 0.0045 dB of the Shannon limit," IEEE Commun. Lett., vol. 5, no. 2, pp. 58-60, 2001, doi: 10.1109/4234.905935.
[9]E. Arikan, "Channel Polarization: A Method for Constructing Capacity-Achieving Codes for Symmetric Binary-Input Memoryless Channels," IEEE Trans. Inf. Theory, vol. 55, no. 7, pp. 3051-3073, 2009, doi: 10.1109/TIT.2009.2021379.
[10]C.-C. Liu, C.-C. Yang, J.-F. Huang, and W.-H. Hung, "Performance analysis of polar codes for one-coincidence-coded OCDM networks," Opt. Fiber Technol., vol. 64, p. 102525, 2021/07/01/ 2021, doi: https://doi.org/10.1016/j.yofte.2021.102525.
[11]G. E. Hinton, S. Osindero, and Y.-W. Teh, "A Fast Learning Algorithm for Deep Belief Nets," Neural Comput., vol. 18, no. 7, pp. 1527-1554, 2006, doi: 10.1162/neco.2006.18.7.1527.
[12]A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet classification with deep convolutional neural networks," presented at the Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1, Lake Tahoe, Nevada, 2012.
[13]G. Hinton et al., "Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups," IEEE Signal Process Mag., vol. 29, no. 6, pp. 82-97, 2012, doi: 10.1109/MSP.2012.2205597.
[14]I. Sutskever, O. Vinyals, and Q. V. Le, "Sequence to Sequence Learning with Neural Networks," p. arXiv:1409.3215doi: 10.48550/arXiv.1409.3215.
[15]H.-T. Cheng et al., "Wide & Deep Learning for Recommender Systems," p. arXiv:1606.07792doi: 10.48550/arXiv.1606.07792.
[16]T. Gruber, S. Cammerer, J. Hoydis, and S. t. Brink, "On deep learning-based channel decoding," in 2017 51st Annual Conference on Information Sciences and Systems (CISS), 22-24 March 2017 2017, pp. 1-6, doi: 10.1109/CISS.2017.7926071.
[17]H. Kim, Y. Jiang, R. Rana, S. Kannan, S. Oh, and P. Viswanath, "Communication Algorithms via Deep Learning," p. arXiv:1805.09317doi: 10.48550/arXiv.1805.09317.
[18]H. Zhu, Z. Cao, Y. Zhao, and D. Li, "Learning to Denoise and Decode: A Novel Residual Neural Network Decoder for Polar Codes," IEEE Trans. Veh. Technol., vol. 69, no. 8, pp. 8725-8738, 2020, doi: 10.1109/TVT.2020.3000345.
[19]C. Wen, J. Xiong, L. Gui, and L. Zhang, "A BP-NN Decoding Algorithm for Polar Codes," in 2019 11th International Conference on Wireless Communications and Signal Processing (WCSP), 23-25 Oct. 2019 2019, pp. 1-5, doi: 10.1109/WCSP.2019.8927910.
[20]G. Larue, L. A. Dufrene, Q. Lampin, H. Ghauch, and G. R. B. Othman, "Neural Belief Propagation Auto-Encoder for Linear Block Code Design," IEEE Trans. Commun., vol. 70, no. 11, pp. 7250-7264, 2022, doi: 10.1109/TCOMM.2022.3208331.
[21]I. Ashour, S. Shaari, P. S. Menon, and H. Bakarman, "Optical Code-Division Multiple-Access and Wavelength Division Multiplexing: Hybrid Scheme Review," J. Comput. Sci., vol. 10, pp. 1718-1729, 01/01 2012, doi: 10.3844/jcssp.2012.1718.1729.
[22]R. Tanner, "A recursive approach to low complexity codes," IEEE Trans. Inf. Theory, vol. 27, no. 5, pp. 533-547, 1981, doi: 10.1109/TIT.1981.1056404.
[23]R. Mori and T. Tanaka, "Performance of Polar Codes with the Construction using Density Evolution," IEEE Commun. Lett., vol. 13, no. 7, pp. 519-521, 2009, doi: 10.1109/LCOMM.2009.090428.
[24]P. Trifonov, "Efficient Design and Decoding of Polar Codes," IEEE Trans. Commun., vol. 60, no. 11, pp. 3221-3227, 2012, doi: 10.1109/TCOMM.2012.081512.110872.
[25]F. R. Kschischang, B. J. Frey, and H. A. Loeliger, "Factor graphs and the sum-product algorithm," IEEE Trans. Inf. Theory, vol. 47, no. 2, pp. 498-519, 2001, doi: 10.1109/18.910572.
[26]W. S. McCulloch and W. Pitts, "A logical calculus of the ideas immanent in nervous activity," Bull. Math. Biol., vol. 5, no. 4, pp. 115-133, 1943/12/01 1943, doi: 10.1007/BF02478259.
[27]F. Rosenblatt, "The perceptron: a probabilistic model for information storage and organization in the brain," (in eng), Psychol Rev, vol. 65, no. 6, pp. 386-408, Nov 1958, doi: 10.1037/h0042519.
[28]D. E. Rumelhart, G. E. Hinton, and R. J. Williams, "Learning representations by back-propagating errors," Nature, vol. 323, no. 6088, pp. 533-536, 1986/10/01 1986, doi: 10.1038/323533a0.
[29]S. R. Dubey, S. K. Singh, and B. B. Chaudhuri, "Activation Functions in Deep Learning: A Comprehensive Survey and Benchmark," p. arXiv:2109.14545doi: 10.48550/arXiv.2109.14545.
[30]K. Janocha and W. M. Czarnecki, "On Loss Functions for Deep Neural Networks in Classification," p. arXiv:1702.05659doi: 10.48550/arXiv.1702.05659.
[31]S. Ruder, "An overview of gradient descent optimization algorithms," p. arXiv:1609.04747doi: 10.48550/arXiv.1609.04747.