本文選題：Stokes空間 + 認知光網(wǎng)絡(luò)　； 參考：《中山大學(xué)》2017年碩士論文

【摘要】：隨著用戶需求的不斷增長,未來光纖通信網(wǎng)絡(luò)的發(fā)展呈現(xiàn)高突發(fā)性、服務(wù)多樣化以及智能化的特點。認知光網(wǎng)絡(luò),因其能夠根據(jù)具體環(huán)境和用戶需求靈活地分配帶寬以提高網(wǎng)絡(luò)資源效率,被看作是下一代光網(wǎng)絡(luò)的重要選擇之一。為了提高認知光網(wǎng)絡(luò)的快速適配的能力,往往需要光接收機提前獲取信號的某些相關(guān)參數(shù),如調(diào)制格式、光信噪比,以便優(yōu)化后續(xù)的信號處理和解調(diào)過程。但是在光纖通信系統(tǒng)中,這些參數(shù)的獲取或估計往往受到相干光接收機大相位噪聲,光纖偏振相關(guān)損傷和非線性損傷的影響。近年來,基于Stokes空間參量的相干光信號分析,由于不依賴于接收信號的偏振旋轉(zhuǎn)和混合,相干光接收機引入的大頻率偏移和相位噪聲,而受到廣泛關(guān)注。本文主要基于Stokes空間參量的分析方法,對認知相干光接收機中的關(guān)鍵技術(shù)——調(diào)制格式識別以及光信噪比監(jiān)測展開研究,主要研究內(nèi)容及創(chuàng)新點如下:(1)通過用MATLAB進行仿真實驗,探究了光纖信道的一些非理想因素對信號在Stokes空間分布的影響;(2)設(shè)計并提出了一個基于Stokes空間參量和減法聚類的調(diào)制格式識別算法。該算法可以實現(xiàn)對5種常見調(diào)制格式(BPSK,QPSK,8PSK,8QAM,16QAM)的正確識別;同已有的基于Stokes空間參量的調(diào)制格式識別方法比較,該算法具有較低的復(fù)雜度和較好的噪聲容忍度;(3)實現(xiàn)了一種基于Stokes空間參量和人工神經(jīng)網(wǎng)絡(luò)的光信噪比監(jiān)測算法。此方法根據(jù)不同光信噪比下信號在Stokes空間最小二乘平面上星座點彌散度不同,基于人工神經(jīng)網(wǎng)絡(luò)算法學(xué)習(xí)二者之間的映射關(guān)系,進而估算光信噪比。仿真結(jié)果表明在OSNR大于5d B的時候能實現(xiàn)0.5d B以內(nèi)的精確度。相比于一般的函數(shù)擬合方法,這種基于人工神經(jīng)網(wǎng)絡(luò)的算法具有更高的計算精度;(4)搭建長距離偏振復(fù)用相干光通信系統(tǒng),對以上兩種算法進行驗證和評估。在調(diào)制格式識別的實驗中實現(xiàn)了對QPSK和16QAM信號在背靠背傳輸以及長距離傳輸條件下的正確識別,在光信噪比監(jiān)測的實驗中對QPSK信號的數(shù)據(jù)實現(xiàn)了較高精度的監(jiān)測結(jié)果。
[Abstract]:With the increasing demand of users, the development of optical fiber communication network in the future is characterized by high burst, diversified service and intelligence. Cognitive optical networks are regarded as one of the important options for next generation optical networks because they can flexibly allocate bandwidth according to specific environment and user needs to improve network resource efficiency. In order to improve the ability of fast adaptation of cognitive optical networks, it is often necessary for optical receivers to obtain some related parameters of signals in advance, such as modulation format, optical signal-to-noise ratio (SNR), so as to optimize the subsequent signal processing and demodulation process. However, in optical fiber communication systems, the acquisition and estimation of these parameters are often affected by large phase noise, polarization dependent damage and nonlinear damage of coherent optical receivers. In recent years, the analysis of coherent optical signals based on Stokes spatial parameters has attracted wide attention due to the large frequency offset and phase noise introduced by coherent optical receivers, which do not depend on the polarization rotation and mixing of received signals. Based on the analysis method of Stokes spatial parameters, this paper studies the key techniques of cognitive coherent optical receiver, modulation format recognition and optical signal-to-noise ratio monitoring. The main research contents and innovations are as follows: 1) Simulation experiments are carried out with MATLAB. In this paper, the influence of some non-ideal factors on the spatial distribution of signal in Stokes is discussed, and a modulation format recognition algorithm based on Stokes spatial parameter and subtraction clustering is proposed. This algorithm can realize the correct recognition of five common modulation formats: BPSK, QPSK, 8PSK, 8QAM16QAM, and compare with the existing modulation format recognition methods based on Stokes spatial parameters. The algorithm has lower complexity and better noise tolerance. It implements an optical signal-to-noise ratio (SNR) monitoring algorithm based on Stokes spatial parameters and artificial neural networks. This method is based on artificial neural network algorithm to study the mapping relationship between constellation points in Stokes space and estimate the optical signal-to-noise ratio (SNR) according to the different optical signal-to-noise ratio (SNR) signal to noise ratio (OSNR) signal to noise ratio (OSNR). The simulation results show that the accuracy of 0.5 dB can be achieved when the OSNR is larger than 5 dB. Compared with the general function fitting method, the algorithm based on artificial neural network has a higher computational accuracy. (4) A long distance polarization multiplexing coherent optical communication system is built, and the above two algorithms are verified and evaluated. In the experiment of modulation format recognition, the correct recognition of QPSK and 16QAM signals under the condition of back-to-back transmission and long distance transmission is realized. In the experiment of optical signal-to-noise ratio monitoring, the data of QPSK signal are monitored with high accuracy.
【學(xué)位授予單位】：中山大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TN929.1

【參考文獻】

相關(guān)期刊論文 前1條

1 吳承治;;探討用于網(wǎng)格的光網(wǎng)絡(luò)技術(shù)[J];現(xiàn)代傳輸;2009年01期

，

本文編號：1817388