【摘要】：通信信號(hào)調(diào)制識(shí)別是指對(duì)接收信號(hào)自動(dòng)處理并判定其調(diào)制類(lèi)型的過(guò)程。作為信號(hào)檢測(cè)與解調(diào)的中間環(huán)節(jié)，調(diào)制識(shí)別技術(shù)在認(rèn)知無(wú)線(xiàn)電、智能解調(diào)器、電子偵察等各種民用及軍事應(yīng)用中扮演著重要角色。調(diào)制識(shí)別技術(shù)經(jīng)過(guò)幾十年的發(fā)展，雖然已經(jīng)取得了很多成果，但隨著工程化需求逐漸提高，無(wú)線(xiàn)通信信道環(huán)境日益復(fù)雜，仍有不少問(wèn)題亟待解決。 本文重點(diǎn)致力于似然比調(diào)制識(shí)別算法的工程化應(yīng)用和基于先進(jìn)信號(hào)處理方法的調(diào)制識(shí)別技術(shù)的研究，論文的主要工作和創(chuàng)新性成果主要包括： 1.從通信信號(hào)處理的工程實(shí)用角度出發(fā)，針對(duì)似然比調(diào)制識(shí)別算法計(jì)算復(fù)雜度高的問(wèn)題，提出了一種改進(jìn)的快速算法。該算法引入硬件實(shí)現(xiàn)中預(yù)存查表的思想，通過(guò)一個(gè)查找表的地址讀出待識(shí)別信號(hào)的似然函數(shù)值以便節(jié)省在線(xiàn)處理時(shí)間，為似然比調(diào)制識(shí)別算法的實(shí)時(shí)化應(yīng)用提供了一種解決思路。試驗(yàn)結(jié)果表明算法能夠在保證似然比算法最優(yōu)性的基礎(chǔ)上有效地節(jié)約時(shí)間成本，更能適應(yīng)實(shí)時(shí)性要求高的應(yīng)用場(chǎng)合。 2.在平坦衰落信道環(huán)境下，提出了基于一種自適應(yīng)馬爾可夫鏈蒙特卡羅（MCMC）技術(shù)——自適應(yīng)Metropolis（AM）技術(shù)的調(diào)制識(shí)別算法。該算法能夠在迭代過(guò)程中產(chǎn)生滿(mǎn)足目標(biāo)分布的未知參數(shù)和發(fā)送符號(hào)的各態(tài)歷經(jīng)樣本，從而在實(shí)現(xiàn)似然函數(shù)的近似計(jì)算的同時(shí)完成參數(shù)估計(jì)，可實(shí)現(xiàn)調(diào)制識(shí)別與參數(shù)估計(jì)的一體化處理。相比于傳統(tǒng)Metropolis-Hastings（MH）技術(shù)，AM技術(shù)避免了因建議分布函數(shù)選取不當(dāng)造成的性能損失。仿真結(jié)果表明在平坦衰落信道環(huán)境下，基于AM技術(shù)的調(diào)制識(shí)別算法能夠快速、精確的收斂，具有很好的識(shí)別性能。 3.將平坦衰落信道環(huán)境中基于AM技術(shù)的調(diào)制識(shí)別算法推廣到多徑信道環(huán)境中，為解決AM算法在未知參數(shù)維數(shù)較高時(shí)收斂速度放緩的問(wèn)題，提出了基于單分量自適應(yīng)Metropolis（SCAM）技術(shù)的調(diào)制識(shí)別算法。該算法在迭代過(guò)程中按順序依次對(duì)未知參數(shù)向量中每個(gè)分量單獨(dú)進(jìn)行類(lèi)似AM采樣的操作。仿真結(jié)果表明在未知參數(shù)維數(shù)較高的情況下，，SCAM算法擁有更加優(yōu)越的收斂性能，基于SCAM技術(shù)的調(diào)制識(shí)別算法在多徑信道環(huán)境下識(shí)別性能良好。 4.在低信噪比環(huán)境下，提出了基于混沌理論的調(diào)制識(shí)別算法。文中對(duì)Duffing振子大尺度周期態(tài)特性進(jìn)行了系統(tǒng)研究，考察了激勵(lì)信號(hào)頻率、幅度和相位對(duì)Duffing振子周期解的影響。根據(jù)大尺度周期狀態(tài)下系統(tǒng)解隨激勵(lì)信號(hào)相位變化的規(guī)律，設(shè)計(jì)了基于Duffing振子的MPSK信號(hào)調(diào)制識(shí)別算法。并根據(jù)調(diào)制識(shí)別中的特征提取任務(wù)需求，對(duì)Duffing振子進(jìn)行了模型優(yōu)化，將優(yōu)化后Duffing振子系統(tǒng)解的Poincaré映射作為分類(lèi)特征，實(shí)現(xiàn)了低信噪比下的MPSK信號(hào)的調(diào)制識(shí)別。該算法僅僅利用載波頻率的先驗(yàn)信息，無(wú)需進(jìn)行碼元同步，并且對(duì)信號(hào)幅度、載波初始相位以及載波頻率的變化不敏感。仿真表明該算法具有較強(qiáng)的噪聲免疫力，在信噪比較低的情況下仍能達(dá)到滿(mǎn)意的識(shí)別效果。
[Abstract]:Communication signal modulation identification refers to a process of automatically processing a received signal and determining its modulation type. As the intermediate link of signal detection and demodulation, modulation recognition technology plays an important role in various civil and military applications such as cognitive radio, intelligent demodulator and electronic reconnaissance. After several decades of development of modulation recognition technology, although many achievements have been achieved, with the increasing engineering demand, the wireless communication channel environment is becoming more and more complex, and there are still many problems to be solved. In this paper, we focus on the engineering application of likelihood ratio modulation recognition algorithm and the research of modulation recognition technology based on advanced signal processing method. Including: 1. Engineering from communication signal processing Aiming at the problem of high computational complexity of likelihood ratio modulation recognition algorithm, an improvement is put forward. The algorithm introduces the idea of pre-stored look-up table in hardware implementation, reads the quasi-random function value of the signal to be identified through the address of a look-up table in order to save on-line processing time, and provides a real-time application of the likelihood ratio modulation recognition algorithm. The experimental results show that the algorithm can effectively save time cost on the basis of guaranteeing the optimality of likelihood ratio algorithm, and can meet the requirement of real-time performance. In the context of flat fading channel, a self-adaptive Markov chain Monte Carlo (MCMC) technique is proposed. The algorithm can generate the unknown parameters satisfying the target distribution and the state ergodic samples of the transmitted symbols in the iterative process, Compared with the traditional method of Metropolis-Hastings (MH), AM technology avoids the selection of the proposed distribution function. The simulation results show that the modulation recognition algorithm based on AM technology can rapidly and accurately converge in a flat fading channel environment. In order to solve the problem of slowing convergence speed when the unknown parameter dimension of AM algorithm is higher, a single-component adaptive Polis (SCA) is proposed in this paper. M) a modulation recognition algorithm of the technique, wherein each component in the unknown parameter vector is separated in sequence according to sequence in the iterative process The simulation results show that the SCAM algorithm has better convergence performance when the unknown parameter dimension is high, and the modulation recognition algorithm based on the SCAM technology is in multi-path. in a low signal-to-noise ratio environment, In this paper, a systematic study of Duffing oscillator's large-scale periodic state characteristics is carried out, and the frequency, amplitude and phase pairs of excitation signals are investigated. The effect of periodic solution of uffing oscillator. Based on the law of the phase change of the system solution with the excitation signal under the large-scale periodic state, Duffing oscillation is designed. According to the requirement of feature extraction task in the modulation recognition, the Duffing oscillator is optimized, the Poincare map of the optimized Duffing oscillator system is used as the classification feature, and the low signal noise is realized. the algorithm only uses a priori information of the carrier frequency, does not need to perform symbol synchronization, The simulation results show that the algorithm has strong noise immunity and is low in signal-to-noise ratio.
【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：TN911.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 呂鐵軍,郭雙冰,肖先賜;基于模糊積分的通信信號(hào)調(diào)制識(shí)別方法研究[J];電子學(xué)報(bào);2001年06期

2 王首勇,朱光喜,唐遠(yuǎn)炎;應(yīng)用最優(yōu)小波包變換的特征提取方法[J];電子學(xué)報(bào);2003年07期

3 韓鋼,李建東,李長(zhǎng)樂(lè),蔡雪蓮;一種改進(jìn)的基于累積量的MDPSK信號(hào)分類(lèi)算法[J];電子學(xué)報(bào);2004年10期

4 路鵬,李月;微弱正弦信號(hào)幅值混沌檢測(cè)的一種改進(jìn)方案[J];電子學(xué)報(bào);2005年03期

5 陳衛(wèi)東,楊紹全;基于循環(huán)累量不變量的MPSK信號(hào)調(diào)制識(shí)別算法[J];電子與信息學(xué)報(bào);2003年03期

6 李楊,李國(guó)通,楊根慶;通信信號(hào)數(shù)字調(diào)制方式自動(dòng)識(shí)別算法研究[J];電子與信息學(xué)報(bào);2005年02期

7 柳征;王明陽(yáng);姜文利;周一宇;;一種新的貝葉斯調(diào)制分類(lèi)算法[J];電子與信息學(xué)報(bào);2006年07期

8 王彬;葛臨東;徐立清;劉媛濤;;一種基于信道盲辨識(shí)和盲均衡的多徑信道調(diào)制方式識(shí)別算法[J];電子與信息學(xué)報(bào);2008年08期

9 譚曉衡;鄢海燕;蘇萌;;基于自適應(yīng)小波消噪的數(shù)字調(diào)制識(shí)別優(yōu)化算法[J];電子與信息學(xué)報(bào);2011年02期

10 阮秀凱;張志涌;;基于小波變換的調(diào)制自動(dòng)識(shí)別新方法研究[J];南京郵電大學(xué)學(xué)報(bào)(自然科學(xué)版);2007年01期

相關(guān)博士學(xué)位論文 前1條

1 謝濤;外激勵(lì)型混沌振子微弱信號(hào)檢測(cè)方法研究[D];北京交通大學(xué);2009年

本文編號(hào)：2266446