頻移相位同調光纖通信系統的效能分析; Performance Analysis of Coherent Optical Fiber Systems Using

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Electrical Engineering > Electronic Thesis & Dissertation > Item 987654321/8862

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/8862

Title:	頻移相位同調光纖通信系統的效能分析;Performance Analysis of Coherent Optical Fiber Systems Using
Authors:	王家男;Ja-Nan Wang
Contributors:	電機工程研究所
Keywords:	同調光通訊;頻移調變;Coherent optical communication;FSK
Date:	2001-07-06
Issue Date:	2009-09-22 11:36:39 (UTC+8)
Publisher:	國立中央大學圖書館
Abstract:	同調光纖系統 (coherent optical fiber systems), 因在接收機端, 加入雷射模組作為本地震盪器 (local oscillator), 引用了光的波動性質, 使得雷射的相位噪音 (phase noise), 會顯著影響接收機的特性。因相位噪音的統計特性不易計算, 使得同調光纖系統的效能分析變得具有挑戰性。其中頻移 (FSK) 調變在同調光纖系統中, 有兩個不同的範疇: 一種是窄的頻移調變, 其頻移的差額小於資料傳輸量, 其接收機的形式, 以相位分散 (phase-diversity) 方式為主。另一種是寬的頻移調變, 其頻移的差額大於兩倍的資料傳輸量, 則可使用兩個中頻濾波器解調。本文即針對這二種調變方式, 提出了效能的分析計算和模擬方法。在分析計算方面, 利用隨機過程上的技巧, 求得了高斯近似法的收斂解。因在數值計算上無累進誤差, 使之應用於評估各類的操作參數, 和更複雜的系統模型。且更進一步, 求得矩量生成函數 (moment generating function), 並率先採用特殊變數變換的技巧, 使得數值積分能夠有效的收斂。在模擬方法方面, 所發展出的模擬方式, 可由一般的個人電腦執行。我們又引入了 importance sampling 的方法, 使得電腦模擬可以延伸到計算極低的機率。 The BERs of CPFSK and FSK receivers are numerical computed efficiently. The effects of shot and laser phase noise are taken into account. For a particular realizations of laser phase noises, the BER is expressed as a Laplace inversion integral. It is then averaged with respect to the distribution of the phase noise. The method of changes of variable is used twice in the computation, and the results converge rapidly. Gaussian approximation of outputs of CPFSK receivers is derived. However, for ideal and Gaussian frequency shapes, Gaussian approximation methods are shown to be not so accurate for the case of low BER. We have established the models of noises and FSK signals in coherent communication systems. These simulations can be implemented systematically to compute the BER of the receivers and histograms of probability distributions of outputs. They can also be applied to checking the validity of performance analyzation. By using importance sampling method we computed the low BER without consuming much computer CPU time. So we can use these simulation methods to assistant the practical system design, which is considered for very low BER. In fact, it is so efficient that we only utilized common personal computers as simulation platforms. Furthermore, a random number generator with a very long period is no longer necessary for the shorter convergent time. From the results of computer simulation, we found that Gaussian approximation overestimate the system performance. One reason is that shot noises in coherent frequency modulation systems may not obey any type of large number theory well enough to utilize Gaussian approximation.
Appears in Collections:	[Graduate Institute of Electrical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Size	Format

社群 sharing

Loading...