跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 張吉良
Ji-Lian Chang
論文名稱： 利用進化演算法在多層感知機結構上之判別回授等化器
指導教授： 賀嘉律
Chia-Lu Ho
口試委員:
學位類別： 碩士
Master
系所名稱： 資訊電機學院 - 電機工程學系
Department of Electrical Engineering
畢業學年度： 89
語文別： 中文
論文頁數： 104
中文關鍵詞： 類神經網路 、符元干擾 、等化器 、多層感知器 、進化演算法 、判別回授等化器 、交配 、突變
外文關鍵詞： Neural Networks, ISI, Equalizer, MLP, EA, DFE, crossover, mutation
相關次數： 點閱：10  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 在近幾年來，類神經網路(Neural Networks)十分被重視，它是一個解決非線性問題的有力工具，它被應用在許多方面，而在調適性等化器上面，也得到非常好的效果，在數位通訊系統中，為了消除符元干擾(Inter Symbol Interference, ISI)和Noise，等化器是十分必要的，對於通訊系統而言，訊號間干擾的ISI效應和Noise不僅是造成本身傳送訊號的失真，而且可能還會造成接收端的判別錯誤，使得接收到的訊號發生錯誤，資料不正確，接收端的等化器(Equalizer)可消除ISI效應和Noise，資料的正確率更是靠它才能大大提升，而調適性等化器通常使用參數的學習演算法，傳統的做法是使用最小均方差演算法(Least Mean Square, LMS)。
    這篇論文提出一個以新的進化演算法(Evolution Algorithm, EA)應用在多層感知器(Multi-Layer Perceptron, MLP)的後遞式判別式回授化器(Decision Feedback Equalizer, DFE)。是一種利用類神經網路(Neural Networks)，模仿生物神經元、生物基因進化遺傳，經由交配(crossover)、突變(mutation)、選擇(selection)、求得好的等化器係數，並且希望由進化演算法中與電腦模擬的結果中，比較出和其他做法的差異和性能。

    目錄 頁碼 摘要 誌謝 目錄I 圖目、表目III 第一章 續論1 1.1 Introduction1 1.2 等化器(Equalizer)3 1.2.1 等化器之需求3 1.2.2 等化器之分類5 1.2.3 非線性等化器之必要6 1.3 研究動機結構與流程7 第二章 多層感知器(Multi-Layer Perceptron, MLP)8 2.1 多層感知器簡介8 2.1.1 生物神經元結構9 2.1.2 類神經元模型10 2.1.3 感知器11 2.2 多層感知機結構13 2.3 將多層感知機架構在等化器上15 2.4 Norm back propagation algorithm19 第三章 進化演算法(Evolution Algorithm, EA)24 3.1 進化演算法簡介24 3.2 進化演算法的演算流程及各步驟30 3.3 架構EA在多層感知器所構成的等化器上37 第四章 模擬與結果(Simulation and Results)45 4.1 適存度(Fitness)46 4.2 均方差(Mean Square Error)51 4.3 位元錯誤率(Bit Error Rate)54 4.4 決策區間(Decision Region)74 4.5 三度空間決策區間(3D Space Decision Region)89 第五章 結論(Conclusion)97 參考文獻(Reference)99 圖目、表目 頁碼 圖1.1.1 Schematic of data transmission system2 圖1.2.1 Baseband Communication System3 圖1.2.2 等化器之型態、結構與演算法5 圖1.3.1 進化演算法主要流程7 圖2.1.1 生物神經細胞模型9 圖2.1.2 類神經元模型10 圖2.1.3 感知器11 圖2.1.4 單一Neuron11 圖2.1.5 Neuron結構11 圖2.1.6 Sigmoid Function12 圖2.2.1 多層感知器的結構13 圖2.2.2 Multilayer perceptron architecture14 圖2.3.1 Multilayer perceptron decision feedback equalizer15 圖2.3.2 Channel17 圖2.3.3 jth neuron with feedback signals in first layer17 圖2.4.1 for different values of p20 圖3.1.1 進化演算法流程26 圖3.1.2 主要EA作法29 圖3.3.1 DFE using MLP structure37 圖4.1 Channel45 圖4.1.1 (4,1)DFE using (9,3,1)MLP structure46 圖4.1.2~圖4.1.8 (4,1)DFE using (9,3,1)MLP structure Fitness47-50 圖4.2.1~圖4.2.3 (4,1)DFE using (9,3,1)MLP structure MSE51-52 圖4.2.4 Simulation results showing relative convergence rate performance53 圖4.3.1~圖4.3.18 BER Performance of (4,1)DFE using (9,3,1) Evolution-based MLP structure55-72 圖4.4.1 (2,0)DFE using (9,3,1)MLP structure74 圖4.4.2 2D平面上的基本點產生75 圖4.4.3 最佳分界線76 圖4.4.4~圖4.4.5 Gaussian Noise distribution77 圖4.4.6~圖4.4.11 最佳分界線與實際模擬結果比較78-80 圖4.4.12~圖4.4.25 3D of Boundary Error81-87 圖4.5.1 (3,0)DFE using (9,3,1)MLP structure89 圖4.5.2 (3,0)DFE using (9,3,1)MLP structure 在3D立體空間中基本點的產生90 圖4.5.3 (3,0)DFE using (9,3,1)MLP structure 在3D空間中基本點91 圖4.5.4 (3,0)DFE using (9,3,1)MLP structure 在3D空間中用EA模擬的分界結果91 圖4.5.5~圖4.5.12 3D Space Boundary92-95

    [1] S. Siu, G.J. Gibson, and C.F.N. Cowan, “Decision feedback equalizer using neural network structures and performance comparison with stand architecture”, IEE Proc. Part I, Communication, Speech, and Vision, Vol. 137, NO.4, pp 221-225,August, 1990.
    [2] S. Siu, and C.F.N. Cowan, “Performance analysis of the norm back propagation algorithm for adaptive equalizer” IEE Proc.Part F, Vol. 140, No. 1, Febr. 1993.
    [3] G.J. Gibson, S. Siu,and C.F.N. Cowan, “The application of nonlinear structures to the reconstruction of binary signal”, IEEE Trans. On Signal Processing, Vol. 39, No. 8,
    pp.1877-1884, Aug, 1991.
    [4] S. Qureshi, “Adaptive Equalisation”, Proceedings of the IEEE, Vol.73, No.9 pp.1349-1387,Sept.1985.
    [5] T. Back, Evolutionary algorithm in theory and practice: evolution strategies, evolutionary programming, genetic algorithms. Oxford, 1996.
    [6] L. Zadeh, “Outline of a new approach to the analysis of complex systems and decision process ”, IEEE Trans. Syst., Man, Cybern., vol. SMC-3, pp.28-44, 1973.
    [7] L. Schlaffi (1814-1895), “Gesammelte Mathtematische Abhandlungen”, vol. 1, Birkhauser, Basel, 1950, pp209-212.
    [8] J. Makhoul, R. Schwartz, and A. E. Jaroudi, “Classifications Capability of two layer neural nets”, Proc. IEEE Int. Conf. ASSP, pp.635-638, May, 1989.
    [9] J. Mirchandani, and W. Cao, “On hidden nodes for neural nets”, IEEE Trans. Circuit and System, vol. 36, No. 5, pp.661-664, May 1989.
    [10] G. J. Gibson, and C.F.N. Cowan, “On the decision regions of multiplayer perceptrons”, Proc. of the IEEE, vol. 78, pp. 258-262.
    [11] F.Sweeney, P. Power, and C.F.N. Cowan, “An analysis of evolutionary techniques for channel equalisation”, Irish Signals and Systems Conference 1998.
    [12] Thomas Back, Frank Hoffmeister, Hans-Paul Schwefel, “A Survey of Evolution Strategies”, University of Dortmund Department of Computer Science XI P.O Box 50 05 00 . D-4600 Dortmund 50 . Germany.
    [13] Ching-Haur Chang and Sammy Siu, Che-Ho Wei, “Complex Backpropagation Decision Feedback Equalizer With Decision Using Neural Nets”, journal of the Chinese institute of electrical engineering, vol.7, no.1 pp63-69(2000).
    [14] Thmos Back, Hans-Paul Schwefel, “Evolutionary Computation: An Overview”, Evolutionary Computation, 1996., Proceedings of IEEE International Conference on , 1996 , Page(s): 20 -29
    [15] Sammy Siu, Ching-Haur Chang, and Che-HO Wei, Transactions Briefs,“ Norm Back Propagation Algorithm For Adaptive Equalizer”, IEEE Transactions on circuits and systems-II: Analog and digital sigital processing, vol.42, no.9, September 1995.
    [16] P.Power, F. Sweeney, C.F.N. Cowan, “EA Crossover Schemes For a MLP Channel Equalizer”, Electronics, Circuits and Systems, 1999. Proceedings of ICECS ''99. The 6th IEEE International Conference on Volume: 1 , 1999 , Page(s): 407 -410 vol.1
    [17] G. Gibson, S.Siu, C. Cowan, “Multi-Layer Perceptron Structure Applied to Adaptive Equalizers for Data Communications”, IEEE Proceedings ICASSP Glasgow, Scotland, May 1989, pp. 1183-1186.
    [18] F.Sweeney, P Power, C.Cowan, “The Use of Evolutionary
    Optimisation in Channel Equalisation”, Eusipco `98 proceeding, Rhodes, Greece, Sept. 1998, Vol.4, pp.2221-2224.
    [19] D. Goldberg, “Genetic Algorithms in Search Optimisation”, Addison-Wesley,1989.
    [20] C. Lau, “Neural Networks, Theoretical foundations and analysis”, IEEE Press,Piscataway, NJ, 1991.
    [21] C.H. Chang, S. Siu, C.H. Wei, “A Polynomial —Perceptron based Decision Feedback Equalizer with a Robust Learning Algorithm”. Signal Processing, Vol.47, No.2 pp.145-158, 1996.
    [22] S. Siu, C.H. Chang, C.H. Wei, “On the Effect of p of the Norm Back Propagation Algorithm for Adaptive Equalization”. IEEE Transactions on Ciucuits and Systems II: Analog and Digital Signal Processing, Sept, 1995,, Vol.42, No.9, pp.604-607.
    [23] C.H. Chang, S. Siu, and C.H. Wei, “A Decision Feedback Equalizer Using Higher-Order Correction”, IEEE, Int. Symposium on Circuits and Systems, Proc. Vol.1, pp.707-710, Sheraton, Chicago Hotel & Towers Chicago, IIIinois, May 3-6, 1993.
    [24] S. Siu, C.H. Chang, and C.H. Wei, “Square- Root Recursive Prediction Error Algorithm for Perceptron-Based Adaptive Equalization Over Frequency Selcetive Fading Channel”, IEEE Int. Symposium on Circuits and System, Seattle, Washington, USA, April 30 May 3, 1995.
    [25] S. Siu, G.J. Gibson, and C.F.N. Cowan, “Decisino Feedback Equalization Using Neural Network Structures”, Proc. IEE Int. Conf. On Neural Networks, pp.124-128, Oct.16-18, 1989, London, U.K.
    [26] G.J. Gibson, S. Siu, and C.F.N. Cowan, “Application of Multi-layer Perceptron as Adaptive Channel Equalizers”, Adaptive Systems in Control and Signal Processing 1989. Selected papers from the IFAC Symposium, Glasgow, U.K.,19-21 April 1989 (Oxford, U.K. : Pergamon), pp.573-578.
    [27] S. Siu, and C.F.N. Cowan, “Adaptive Equalization Using Back Propagation learning Algorithm”, Proc. IEE Int. Conf. On Neural Networks ,Nov. 18-20, 1991, U.K.
    [28] C.H. Chang, S. Siu,and C.H. Wei, “Decision Feedback Equalization Using Complex Back Propagation Algorithm”, IEEE Int. Symposium on Circuits ans Systems, H.K. 1997.
    [29] S. Siu, J.L.Shi, C.S. Liao, and M.J. Wu, “Digital Synchronization Network”, Teleom. Technical Quarterly, Vol.10, No.3, Febr., 1991.
    [30] S. Siu, J.L.Shi, C.S. Liao, and M.J. Wu, “The Performance of the Synchronization Nwtwork Clocks in Digital Network”, Telcom, Technical Quarterly, Vol.12, No.2, 1992.
    [31] S. Siu, C.H. Chang, C.H. Wei, “Reduction of Nonlinear Distortion Using Adaptive Polynomial-Perceptron Equalizer”, Proc. of the First(1995) Radio Science Symposium, Kaohsiung, Taiwan, ROC, pp.245-249, Aug.1995.
    [32] 蘇木春, 張孝德, “機器學習─類神經網路、模糊系統以及基因演算法則”, 全華科技圖書股份有限公司, 二版, 88.
    [33] 王炤棋, “進化計算與並行式類神經網路應用於無線網路頻道配置之研究”, 國立中央大學資訊工程研究所碩士論文, 民86.
    [34] 施柏屹, “倒傳遞類神經網路學習收斂之初步探討”, 國立中央大學機械工程研究所碩士論文, 民89.
    [35] 陳建錦, “應用進化演算法解二次分配問題”, 國立中央大學資訊工程研究所碩士論文, 民88.
    [36] 陳君瑋, “全數位化π/4-shifted DQPSK之分析與實現”, 國立中央大學電機工程研究所碩士論文, 民89.
    [37] 孫春在博士, “人工智慧課程”, 交大資訊科學研究所.
    [38] 張清濠, “使用健全學習法則的多項式類神經網路等化器”, 國立交通大學電子研究所博士論文, 民84.
    [39] 林繼洲, “函數連結與模糊適應等化器效能評估”, 元智大學電機工程研究所碩士論文, 民88.
    [40] Simon Haykin, “Neural Networks”, McMaster University, Hamilton, Ontario. Canada
    [41] Yoh-Han Pao, “Adaptive Pattern Recognition and Neural Networks”, Case Western Reserve University.

    QR CODE
    :::