臺灣博碩士論文加值系統

摘要
傳統的適應性類神經網路等化器(adaptive neural networks equalizer)，須配合使用硬決定器(hard decision)並使用倒傳遞演算法(back propagation algorithm)[4,5]來設計等化器。一般傳統的等化器，在估測訊號時，必須使用一個參考訊號(reference signal)來進行估測，當參考訊號未知時，便無法正確估測出訊號[4,5,17]，所以此種方法的運用限制頗大。在無線通訊系統上已有多種自主式等化器被提出，如以LMS演算法 (least mean square algorithm) [2,4,5]、RLS演算法 (recursive least square algorithm) [2,4,5]為基礎之自主式等化器，而自主式等化器(blind equalizer)最大的優點是不須使用參考訊號[2,4,6,7,8]，只需利用接收端的訊號便可將傳輸訊號順利估測出來。
本篇論文中，我們提出一種有別於傳統方法之自主式類神經網路等化器(blind neural networks equalizer)，應用高階統計訊號理論之Maximum Kurtosis Deconvolution (MKD)方法[6,7,8]來設計自主式類神經網路等化器，改善傳統類神經網路等化器之缺點，並以類神經網路為基礎結合沃特拉等化器利用Maximum Kurtosis Deconvolution (MKD)方法設計一個非線性結構之等化器，直接以接收端之訊號，來分析與模擬自主式等化器之特性，並將研究設計方法應用於非線性無線通道系統上。

ABSTRACT
The conventional method of a adaptive neural network equalizer used back propagation algorithm with hard decision device. And a reference signal is usually needed. In recent years, in digital communications has many types of blind equalizers been present, which they didn't need to use the reference signals.
In this thesis, we present a new method of the adaptive neural networks with Maximum Kurtosis Deconvolution (MKD) to improve the disadvantages of a adaptive neural networks equalizer. Finally, we develop the second order Volterra filter based on neural networks with Maximum Kurtosis Deconvolution (MKD) method, to improve the disadvantages of the LMS second order Volterra equalizer. we use only the received signals to develop the nonlinear equalizer, and apply them for nonlinear communication system.

目錄
中文摘要…………………………………………………………………I
英文摘要………………………………………………………………III
目錄……………………………………………………………………IV
圖目錄………………………………………………………………VI
第一章 導論……………………………………………………………1
第二章 高階統計訊號理論……………………………………………3
2.1 高階統計訊號理論之Cumulant定義…………………………3
2.2 高階統計訊號理論Kurtosis介紹……………………………5
第三章 自主式等化器應用於線性通道系統…………………………7
3.1 自主式等化器通道特性………………………………………7
3.2 LMS演算法為基礎之自主式等化器………………………9
3.3 RLS演算法為基礎之自主式等化器………………12
3.4 Neural networks演算法為基礎之自主式等化器…………13
3.5 電腦模擬結果………………………………………………16
第四章 自主式等化器應用於非線性通道系統……………………23
4.1 等化器非線性通道特性………………23
4.2 沃特拉級數………25
4.3 自主式二階沃特拉等化器.............................................26
4.4 以類神經網路為基礎結合二階沃特拉之自主式等化器........29
4.5 電腦模擬結果........................................................35
第五章 結論…………………………………………………………42
參考文獻………………………………………………………………44

參考文獻
[1]吳岳龍“行動式無線通道高階統計類神經網路等化器設計”, 高雄
工學院電機工程研究所碩士論文,1997
[2]葉茂森 “非線性快速演算法自主式等化器之設計與應用”, 高雄工
學院電機工程研究所碩士論文,1997
[3]許國棟 “具格狀前置濾波器之適應性最小均方值沃特拉通道等化
器研究”, 中山大學電機工程研究所碩士論文,1999
[4]S. Haykin, “Adaptive Filter Theory, 3rd ed.,”, Prentice Hall, 1996.
[5]Widrow & Walach, “Adaptive Inverse Control”, Prentice Hall, 1996
[6]J.N. Li, “Entwurf und Analyse neure Algorithmen fiir die blinde Kana
-lentzerrung unter Verwendung hoherer Momente,” Dissertation,
THDarmstadt, Germany, 1995.
[7]J.N. Li, “Carrier Phase Recovery and Maximum Kurtosis Deconvolut-
ion Based Blind Equalization ”, "FREQUENZ,PP.151-155 July 1995
[8]J.N. Li, “A Variable Step-Size Blind Adaptive Equalizer Based on
Maximum Kurtosis Deconvolution”, 8.Aachener Kolloquium, 1994
[9]J.M. Mendel , “Tutorial on Higher-Order Statistics(Spectra) in Signal
Processing and System Theory：Theoretical Results and Some Applic-
ations”, Proceedings of the IEEE,Vol.79,PP.278-305,1991
[10]A. Papoulis, “Probability , Random Variables , and Stochastic Proce-
sses”, Mc Graw-Hill,Inc. Third Edition, 1991
[11]L.L. Scharf, “Statistical Signal Processing-Detection, Estimation, and
Time Series Analysis”, Addison Wesley, 1991
[12]E.W. Kamen and J.K. Su, “Introduction to Optimal Estimation”,
Springer, 1999
[13]M. Schetzen, “ The Volterra and Wiener Theories of the Nonlinear
System”, New Yoke, Wiley, 1980
[14]T. Koh and E.J. Powers, “Second-order Volterra filtering and its appl-
ication to nonlinear system identification”, IEEE, Trans. On
Acoustics, Speech and Signal Processing, Vol. ASSP-33, No.6, Dec
1985,pp1445-1455
[15]Miroslav V. Dokic and Petter M. Clarkson, “On the Performance of a
Second-Order Adaptive Volterra Filter”, IEEE Trans. On Signal Proce
-ssing, Vol.41,No.5, May 1993
[16]H. V. Poor and G.W. Wornell, “ Wireless Communications (signal pro
-cessing perspective),” Prentice Hall Edition, 1998.
[17]莊國賢 “自主式等化器應用於多輸入多輸出非線性通道系統”,義
守大學電機工程研究所碩士論文,2001