請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/82614完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 曹恆偉(Hen-Wai Tsao) | |
| dc.contributor.author | You-Jie Peng | en |
| dc.contributor.author | 彭祐頡 | zh_TW |
| dc.date.accessioned | 2022-11-25T07:47:51Z | - |
| dc.date.available | 2023-10-27 | |
| dc.date.copyright | 2021-11-02 | |
| dc.date.issued | 2021 | |
| dc.date.submitted | 2021-10-29 | |
| dc.identifier.citation | R. Pan, “深度學習工作流程,” 2019, [Online] available: https://www.slideshare.net/rouyunpan/ss-205278853. C. T. PEI, “Dropout —隨機關閉神經元 | 模擬人腦神經元原理 | 避免模型過擬合,” 2020, [Online] available: https://medium.com/@a5560648/dropout- 5fb2105dbf7c. A. Radford, L. Metz, and S. Chintala, “Unsupervised representation learn- ing with deep convolutional generative adversarial networks,” arXiv preprint arXiv:1511.06434, 2015. T. Bai, H. Zhang, J. Wang, C. Xu, M. Elkashlan, A. Nallanathan, and L. Hanzo,“Fifty years of noise modeling and mitigation in power-line communications,” IEEE Communications Surveys Tutorials, 2020. T. Instruments, “TIDM-TMDSPLCKIT-v3 reference design,” 2014, [Online] available: https://www.electronicsdatasheets.com/manufacturers/texas- instruments/reference-designs/TIDM-TMDSPLCKIT-V3. G. Prasad and L. Lampe, “Full-duplex power line communications: Design and applications from multimedia to smart grid,” IEEE Communications Magazine, vol. 58, no. 2, pp. 106–112, 2019. G. Artale, A. Cataliotti, V. Cosentino, D. Di Cara, R. Fiorelli, S. Guaiana, and G. Tinè, “A new low cost coupling system for power line communication on medium voltage smart grids,” IEEE Transactions on Smart Grid, vol. 9, no. 4, pp. 3321–3329, 2016. N. A. Letizia, A. M. Tonello, and D. Righini, “Learning to synthesize noise: The multiple conductor power line case,” in 2020 IEEE International Symposium on Power Line Communications and its Applications (ISPLC). IEEE, 2020, pp. 1–6. “機器學習 sklearn——-線性迴歸（房價與房屋尺寸關係的線性擬合）,” 2019, [Online] available: https://www.itread01.com/content/1550258285.html. I. Goodfellow, Y. Bengio, and A. Courville, Deep learning. MIT press, 2016. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” in Advances in neural information processing systems, 2017, pp. 5998–6008. 廖星宇, 一直學不會 Tensorflow？PyTorch 更好用更強大更易懂！. 114 台北市內湖區瑞光路 578 號 2 樓: 深石數位, 2018, ch. 3, pp. 45–50. S. Albawi, T. A. Mohammed, and S. Al-Zawi, “Understanding of a convolu- tional neural network,” in 2017 International Conference on Engineering and Technology (ICET). Ieee, 2017, pp. 1–6. W. Zaremba, I. Sutskever, and O. Vinyals, “Recurrent neural network regular- ization,” arXiv preprint arXiv:1409.2329, 2014. I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets,” Advances in neural information processing systems, vol. 27, 2014. N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: a simple way to prevent neural networks from overfitting,” The journal of machine learning research, vol. 15, no. 1, pp. 1929–1958, 2014. T. Kanungo, D. M. Mount, N. S. Netanyahu, C. D. Piatko, R. Silverman, and A. Y. Wu, “An efficient k-means clustering algorithm: Analysis and imple- mentation,” IEEE transactions on pattern analysis and machine intelligence, vol. 24, no. 7, pp. 881–892, 2002. P. Cunningham, “Dimension reduction,” in Machine learning techniques for multimedia. Springer, 2008, pp. 91–112. M. Nixon and A. Aguado, Feature extraction and image processing for computer vision. Academic press, 2019. A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Advances in neural information processing systems, vol. 25, pp. 1097–1105, 2012. P. Druzhkov and V. Kustikova, “A survey of deep learning methods and soft- ware tools for image classification and object detection,” Pattern Recognition and Image Analysis, vol. 26, no. 1, pp. 9–15, 2016. G. Hu, Y. Yang, D. Yi, J. Kittler, W. Christmas, S. Z. Li, and T. Hospedales, “When face recognition meets with deep learning: an evaluation of convolu- tional neural networks for face recognition,” in Proceedings of the IEEE inter- national conference on computer vision workshops, 2015, pp. 142–150. X. Luo, R. Shen, J. Hu, J. Deng, L. Hu, and Q. Guan, “A deep convolution neural network model for vehicle recognition and face recognition,” Procedia Computer Science, vol. 107, pp. 715–720, 2017. M. Arjovsky, S. Chintala, and L. Bottou, “Wasserstein generative adversarial networks,” in International conference on machine learning. PMLR, 2017, pp. 214–223. Y. Chen, “剖析深度學習 (2)： 你知道 cross entropy 和 kl divergence 代表什麼意義嗎？ 談機器學習裡的資訊理論,” 2020, [Online] available: https://www.ycc.idv.tw/deep-dl_2.html. 廖星宇,一直學不會 Tensorflow？PyTorch 更好用更強大更易懂！. 114 台北市內湖區瑞光路 578 號 2 樓: 深石數位, 2018, ch. 6, pp. 22–26. I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin, and A. Courville, “Im- proved training of wasserstein gans,” arXiv preprint arXiv:1704.00028, 2017. A. A. Smith, “Power line noise survey,” IEEE Transactions on Electromagnetic Compatibility, no. 1, pp. 31–32, 1972. M. Antoniali, M. De Piante, and A. M. Tonello, “Plc noise and channel char- acterization in a compact electrical car,” in 2013 IEEE 17th international sym- posium on power line communications and its applications. IEEE, 2013, pp. 29–34. A. N. Milioudis, K. N. Syranidis, G. T. Andreou, and D. P. Labridis, “Model- ing of medium-voltage power-line communication systems noise levels,” IEEE Transactions on power delivery, vol. 28, no. 4, pp. 2004–2013, 2013. L. Guerrieri, G. Masera, I. S. Stievano, P. Bisaglia, W. R. G. Valverde, and M. Concolato, “Automotive power-line communication channels: Mathemati- cal characterization and hardware emulator,” IEEE Transactions on Industrial Electronics, vol. 63, no. 5, pp. 3081–3090, 2016. B. Han, V. Stoica, C. Kaiser, N. Otterbach, and K. Dostert, “Noise character- ization and emulation for low-voltage power line channels across narrowband and broadband,” Digital Signal Processing, vol. 69, pp. 259–274, 2017. L. Bai, M. Tucci, S. Barmada, M. Raugi, and T. Zheng, “Impulsive noise char- acterization in narrowband power line communication,” Energies, vol. 11, no. 4, p. 863, 2018. S. Liu, F. Yang, and J. Song, “An optimal interleaving scheme with maxi- mum time-frequency diversity for PLC systems,” IEEE Transactions on Power Delivery, vol. 31, no. 3, pp. 1007–1014, 2014. J. A. Cortés, A. Sanz, P. Estopinán, and J. I. García, “On the suitability of the middleton class a noise model for narrowband PLC,” in 2016 International Symposium on Power Line Communications and its Applications (ISPLC). IEEE, 2016, pp. 58–63. M. Zimmermann and K. Dostert, “Analysis and modeling of impulsive noise in broad-band powerline communications,” IEEE transactions on Electromagnetic compatibility, vol. 44, no. 1, pp. 249–258, 2002. H. Meng, Y. L. Guan, and S. Chen, “Modeling and analysis of noise effects on broadband power-line communications,” IEEE Transactions on Power delivery, vol. 20, no. 2, pp. 630–637, 2005. H. Philipps, “Development of a statistical model for powerline communication channels,” in Proc. 2000 April Intl. Symp. on Powerline Com., Limerick, Ire- land, 2000. P. Van Der Gracht and R. Donaldson, “Communication using pseudonoise mod- ulation on electric power distribution circuits,” IEEE transactions on commu- nications, vol. 33, no. 9, pp. 964–974, 1985. M. Katayama, T. Yamazato, and H. Okada, “A mathematical model of noise in narrowband power line communication systems,” IEEE Journal on Selected areas in Communications, vol. 24, no. 7, pp. 1267–1276, 2006. M. Nassar, A. Dabak, I. H. Kim, T. Pande, and B. L. Evans, “Cyclostationary noise modeling in narrowband powerline communication for smart grid applica- tions,” in 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2012, pp. 3089–3092. M. H. Hayes, Statistical digital signal processing and modeling. John Wiley Sons, 2009. M. Elgenedy, M. Sayed, A. El Shafie, I. H. Kim, and N. Al-Dhahir, “Cyclosta- tionary noise modeling based on frequency-shift filtering in NB-PLC,” in 2016 IEEE Global Communications Conference (GLOBECOM). IEEE, 2016, pp. 1–6. W. A. Gardner, “Cyclic wiener filtering: theory and method,” IEEE Transac- tions on communications, vol. 41, no. 1, pp. 151–163, 1993. J. Tian, H. Guo, H. Hu, and H. H. Chen, “Frequency-shift filtering for OFDM systems and its performance analysis,” IEEE Systems Journal, vol. 5, no. 3, pp. 314–320, 2011. N. Shlezinger and R. Dabora, “Frequency-shift filtering for OFDM signal recov- ery in narrowband power line communications,” IEEE Transactions on Com- munications, vol. 62, no. 4, pp. 1283–1295, 2014. S. Moaveninejad, A. Kumar, M. Elgenedy, N. Al-Dhahir, A. M. Tonello, and M. Magarini, “Simpler than fresh filter: A parametric approach for cyclosta- tionary noise generation in NB-PLC,” IEEE Communications Letters, vol. 24, no. 7, pp. 1373–1377, 2020. S. Moaveninejad, A. Kumar, M. Elgenedy, M. Magarini, N. Al-Dhahir, and A. M. Tonello, “Gaussian-middleton classification of cyclostationary correlated noise in hybrid MIMO-OFDM WiNPLC,” in ICC 2019-2019 IEEE Interna- tional Conference on Communications (ICC). IEEE, 2019, pp. 1–7. G. Laguna-Sanchez and M. Lopez-Guerrero, “An experimental study of the effect of human activity on the alpha-stable characteristics of the power-line noise,” in 18th IEEE international symposium on power line communications and its applications. IEEE, 2014, pp. 6–11. S. P. Herath, N. H. Tran, and T. Le-Ngoc, “Optimal signaling scheme and capacity limit of PLC under bernoulli-gaussian impulsive noise,” IEEE Trans- actions on Power Delivery, vol. 30, no. 1, pp. 97–105, 2014. M. Mirahmadi, A. Al-Dweik, and A. Shami, “Ber reduction of OFDM based broadband communication systems over multipath channels with impulsive noise,” IEEE transactions on communications, vol. 61, no. 11, pp. 4602–4615, 2013. E. O. Elliott, “Estimates of error rates for codes on burst-noise channels,” The Bell System Technical Journal, vol. 42, no. 5, pp. 1977–1997, 1963. D. Griffin and J. Lim, “Signal estimation from modified short-time fourier trans- form,” IEEE Transactions on acoustics, speech, and signal processing, vol. 32, no. 2, pp. 236–243, 1984. I. S. Association et al., “Ieee standard for low-frequency (less than 500 khz) nar- rowband power line communications for smart grid applications,” Oct, vol. 31, p. 269, 2013. M. Padala, D. Das, and S. Gujar, “Effect of input noise dimension in GANs,” arXiv preprint arXiv:2004.06882, 2020. C. Donahue, J. McAuley, and M. Puckette, “Adversarial audio synthesis,” arXiv preprint arXiv:1802.04208, 2018. D. Righini and A. M. Tonello, “Automatic clustering of noise in multi-conductor narrow band PLC channels,” in 2019 IEEE International Symposium on Power Line Communications and its Applications (ISPLC). IEEE, 2019, pp. 1–6. C. Donahue, J. McAuley, and M. Puckette, “Synthesizing audio with generative adversarial networks,” arXiv preprint arXiv:1802.04208, vol. 1, 2018. K. G. Hartmann, R. T. Schirrmeister, and T. Ball, “EEG-GAN: Generative adversarial networks for electroencephalograhic (EEG) brain signals,” arXiv preprint arXiv:1806.01875, 2018. K. F. Nieman, J. Lin, M. Nassar, K. Waheed, and B. L. Evans, “Cyclic spectral analysis of power line noise in the 3–200 khz band,” in 2013 IEEE 17th In- ternational Symposium on Power Line Communications and Its Applications. IEEE, 2013, pp. 315–320. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/82614 | - |
| dc.description.abstract | 通訊系統的接收機設計需要考慮到外加雜訊的干擾，並適當地設計系統抑制這些干擾影響資訊傳輸效能。在接收機開發的階段，我們會利用雜訊建模的方式來評估雜訊干擾消除的效能。然而，傳統的「由下而上式」數學建模方法，不容易模擬複雜統計特性的雜訊信號。本研究針對窄頻電力線通訊系統中的主要干擾成分—循環穩態脈衝雜訊，進行建模的研究，目的在設計一個時間序列產生器，其產生的訊號不但在統計特性上能與真實窄頻電力線通訊中的加成性干擾很接近，而且也必須具備足夠的樣本多樣性。本論文基於生成對抗網路進行架構的改良修正，包括了: (1)擴大深度學習模型接受訊號的長度，使模型可以觀察到雜訊循環穩態的特性;(2)將損失函數替代為Wasserstein距離便於能更好的評估訓練集與生成集資料的分佈;(3)設計以雜訊的波形特徵對模型進行訓練的架構。訓練資料集除了採用兩種常見的模型來生成訓練資料外，我們也進行實際環境的量測作為訓練集資料，最後透過定量及定性分析選定最佳的生成架構。研究結果顯示以波形特徵作為訓練的生成對抗網路時間序列產生器在生成雜訊的品質上更接近於量測資料集。雖然此模型在頻譜響應上、時間波形特性上均有不錯的表現，但其循環頻率特性的表現上還有可改善的空間，建議未來可增加遞歸神經網路以及Transformer的架構，提高時間序列之間的相關特性之學習效果。 | zh_TW |
| dc.description.provenance | Made available in DSpace on 2022-11-25T07:47:51Z (GMT). No. of bitstreams: 1 U0001-2710202100331500.pdf: 38803487 bytes, checksum: 4a6e2a9e4f3d294f8345d75a76f2d511 (MD5) Previous issue date: 2021 | en |
| dc.description.tableofcontents | "目錄 致謝 i 摘要 ii Abstract iii 目錄 v 圖目錄 viii 表目錄 xiii 第一章 緒論 1 1.1 前言 1 1.2 研究主題與主要貢獻 2 1.3 論文架構 3 第二章 機器學習與深度學習簡介 5 2.1 深度學習基礎與資料集 7 2.1.1 資料集前處理 9 2.1.2 多層全連結網路(fully connected neural network)介紹 10 2.1.3 加速深度學習運算及訓練的方法 12 2.2 監督式與非監督式學習 14 2.3 卷積神經網路 15 2.3.1 卷積層 15 2.3.2 反卷積層 17 2.3.3 池化層 17 2.3.4 全連結層 18 2.4 生成對抗網路 19 2.4.1 生成對抗網路架構改良 20 2.4.2 損失函數改良 22 第三章 電力線雜訊模型 28 3.1 傳統電力線雜訊種類 28 3.2 窄頻電力線通訊之雜訊模型特性及影響 30 3.3 窄頻電力線循環穩態脈衝雜訊模型 31 3.3.1 基於時變變異數高斯穩態雜訊模型 32 3.3.2 基於頻譜濾波器之高斯穩態脈衝雜訊模型 33 3.3.3 基於頻率平移濾波器(frequency-shift filter, FRESH)之循環穩態脈衝雜訊模型 34 3.3.4 簡化版頻率平移濾波器之循環穩態脈衝雜訊 36 3.4 簡化式窄頻電力線通訊雜訊數學模型 39 3.5 應用生成對抗網路於窄頻電力線雜訊模型 41 第四章 基於生成對抗網路之時間序列產生器設計 46 4.1 資料集介紹 46 4.1.1 資料集(一) 46 4.1.2 資料集(二) 47 4.1.3 資料集(三) 48 4.2 基於電力線通訊雜訊之生成對抗網路架構規劃設計 51 4.2.1 架構1:基於時頻圖特徵學習之生成對抗網路 52 4.2.2 架構2:基於時域波形特徵學習之生成對抗網路 61 4.2.3 架構3:結合時頻域特徵學習生成雜訊架構流程 65 第五章 系統模擬結果分析 67 5.1 基於電力線雜訊生成對抗網路模型評估 67 5.1.1 雜訊定性分析 67 5.1.2 雜訊定量分析 68 5.2 模擬結果與比較 76 5.2.1 資料集(一)應用於生成對抗網路結果分析 76 5.2.2 資料集(二)應用於生成對抗網路結果分析 91 5.2.3 資料集(三)應用於生成對抗網路結果分析 106 5.2.4 定量指標總分析 120 第六章 結論與未來展望 121 6.1 結論 121 6.2 未來展望 123 參考文獻 124" | |
| dc.language.iso | zh-TW | |
| dc.subject | 循環穩態脈衝雜訊 | zh_TW |
| dc.subject | 深度學習 | zh_TW |
| dc.subject | 生成對抗網路 | zh_TW |
| dc.subject | Wasserstein距離 | zh_TW |
| dc.subject | 窄頻電力線通訊 | zh_TW |
| dc.subject | generative adversarial network | en |
| dc.subject | narrow-band power line communication | en |
| dc.subject | cyclostationary impulsive noise | en |
| dc.subject | deep learning | en |
| dc.subject | Wasserstein distance | en |
| dc.title | 基於生成對抗網路時間序列產生器研究: 以窄頻電力線通訊循環穩態脈衝為例 | zh_TW |
| dc.title | Time Series Generator based on Generative Adversarial Network: Cyclic Stationary Impulse Noise for NB-PLC Systems | en |
| dc.date.schoolyear | 109-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.coadvisor | 錢膺仁(Ying-Ren Chien) | |
| dc.contributor.oralexamcommittee | 李宏毅(Hsin-Tsai Liu),馬文忠(Chih-Yang Tseng) | |
| dc.subject.keyword | 窄頻電力線通訊,循環穩態脈衝雜訊,深度學習,生成對抗網路,Wasserstein距離, | zh_TW |
| dc.subject.keyword | narrow-band power line communication,cyclostationary impulsive noise,deep learning,generative adversarial network,Wasserstein distance, | en |
| dc.relation.page | 132 | |
| dc.identifier.doi | 10.6342/NTU202104285 | |
| dc.rights.note | 同意授權(限校園內公開) | |
| dc.date.accepted | 2021-10-29 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 電信工程學研究所 | zh_TW |
| dc.date.embargo-lift | 2023-10-27 | - |
| 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 | |
|---|---|---|---|
| U0001-2710202100331500.pdf 授權僅限NTU校內IP使用（校園外請利用VPN校外連線服務） | 37.89 MB | Adobe PDF |
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。