【正文】
I 本 科 生 畢 業(yè) 論 文 題 目: 基于 M C MA 與 DD L MS 的聯(lián)合 盲 均衡 院 系: 信息科學(xué)與技術(shù)學(xué)院電子與通信工程系 專 業(yè): 通信工程 學(xué)生姓名: 學(xué) 號(hào): 指導(dǎo)教師: (職 稱) 副教授 二〇 一 三 年 六 月 II 學(xué)術(shù)誠信聲明 本人所呈交的畢業(yè)論文,是在導(dǎo)師的指導(dǎo)下,獨(dú)立進(jìn)行研究工作所取得的成果,所有數(shù)據(jù)、圖片資料均真實(shí)可靠。除文中已經(jīng)注明引用的內(nèi)容外,本論文不包含任何其他人或集體已經(jīng)發(fā)表或撰寫過的作品或成果。對本論文的研究作出重要貢獻(xiàn)的個(gè)人和集體,均已在文中以明確的方式標(biāo)明。本畢業(yè)論文的知識(shí)產(chǎn)權(quán)歸屬于培養(yǎng)單位。本人完全意識(shí)到本聲明的法律結(jié)果由本人承擔(dān)。 本人簽名: 日期: III 摘要 在現(xiàn)在不斷發(fā)展的通信領(lǐng)域,隨著信息量的變大,需要不斷提升信息的傳輸速率。而 提升信息的傳輸速率,可以改善通信發(fā)送端和接收端的信號(hào)處理硬件和軟件,或者通過采用更高的碼制來提升信息傳輸速率。以 十六進(jìn)制的正交幅度調(diào)制( 16Quadrature Amplitude Modulation,16QAM) 星座圖為基礎(chǔ)的調(diào)制方式,可以提供現(xiàn)代通信系統(tǒng)一個(gè)較高的碼率,符合現(xiàn)代通信發(fā)展的需求。在接收端,為了消除碼間干擾,傳統(tǒng)上一般采用常模算法( Constant Modulus Algorithm, CMA)進(jìn)行均衡,本文采用修正 恒模算法( Modify Constant Modulus Algorithm , MCMA )和 判 決 引 導(dǎo) 的 最 小 均 方 算 法 ( DecisionDirected LeastMean Square , DDLMS ) 聯(lián) 合 均 衡 算 法 , 對 接 收 端 的 碼間干擾( InterSymbolInterference, ISI) 進(jìn)行均衡,以獲得更優(yōu)的均衡效果,在低信噪比的情況下準(zhǔn)確快速地傳輸信息。 本文對盲均衡通信系統(tǒng)進(jìn)行了簡要分析,詳細(xì)說明了每個(gè)階段信號(hào)序列的變換過程,然后推導(dǎo)了 CMA算法, MCMA 算法, DDLMS 算法及其聯(lián)合均衡算法的實(shí)現(xiàn)過程和迭代公式誤差分析等,分析了聯(lián)合均衡的技術(shù)的實(shí)現(xiàn)方式和切換門限,還對四種不同的均衡算法進(jìn)行了 的仿真實(shí)驗(yàn),得出了各種算法的均衡后星座圖及其剩余誤差分析,還有仿真了 MCMA 與 DDLMS 聯(lián)合均衡 算法在不同的信噪比之下的誤碼率分析,而且對不同的迭代步長因子進(jìn)行了仿真,得出了不同的剩余誤差 變化 曲線。 仿真結(jié)果得到了如下結(jié)論: 與 DDLMS 聯(lián)合均衡 算法較之其它三種均衡算法,能夠?qū)⑹S嗾`差迭代得更小,而且可以有效地修復(fù)傳輸過程中產(chǎn)生的相位偏移; , MCMA 與 DDLMS 聯(lián)合均衡 能夠在相同信噪比的情況下,比 CMA 算法達(dá)到更低的誤碼率, 聯(lián)合均衡算法在信噪比大于 10dB 的時(shí)候,可以將誤碼率減少至 310? 以下, 而且,在信噪比較高的情況下,能夠非常逼近 16QAM 的理論誤碼率; MCMA 與 DDLMS 聯(lián)合均衡 中,較大的 MCMA 算法的步長因子 (本文仿真中使用了最大的步長因子為 ) 能夠快速地收斂到DDLMS 的切換門限值,然后再以 DDLMS 算法將剩余誤差收斂得更小,實(shí)現(xiàn)更快 IV 的收斂,克服了僅用 MCMA 算法的缺陷,但是同時(shí)步長因子也不能過大而造成剩余誤差無法收斂到 DDLMS的切換門限。 盲均衡算法在多個(gè)領(lǐng)域都有了廣泛的應(yīng)用,譬如通信、聲吶、雷達(dá)、地質(zhì)勘探、生物醫(yī)學(xué)工程、控制工程等都具有廣大的應(yīng)用前景和巨大的應(yīng)用市場,而盲均衡算法本身也成為了通信、檢測理論、數(shù)字信號(hào)處理等學(xué)科中的熱門研究對象,在日后必將發(fā)展成為出更為優(yōu)越的盲均衡算法,以適應(yīng)行業(yè)和應(yīng)用的需求。 通過對均衡算法的研究,為通信 接收端的均衡器應(yīng)用提供了更多了理論和依據(jù)。 關(guān)鍵詞: 修正恒模算法( MCMA) ; 判決引導(dǎo)的最小均方算法( DDLMS) ;盲均衡算法;聯(lián)合均衡 V Abstract Nowadays as the development of munication, the increasing information of munication needs to improve the transfer rate of information. Improving the transfer rate of information, can either improve the munication of the sender and the receiver signal processing hardware and software, or by a higher code system to improve the information transmission rate. In 16Quadrature Amplitude Modulation (16QAM) constellation graph based modulation way, modern munication system can provide a higher code rate and meet the demand of the development of modern munication. At the receiving end, in order to eliminate the intersymbol interference, we used to use the Constant Modulus Algorithm (CMA) in equilibrium. This article will use the Modify Constant Modulus Algorithm (MCMA) and DecisionDirected Least Mean Square (DDLMS) joint equalization Algorithm to balance InterSymbolInterference (ISI) at the receiving end, in order to obtain better balance effects in the case of low signaltonoise ratio of transmission information accurately and rapidly. In this paper, the blind equalization munication system is analyzed briefly. The CMA algorithm, the MCMA algorithm, DDLMS algorithm were deduced and its realization process of joint equalization algorithm and iterative formula of error analysis, etc. We analyzed the joint equalization technology implementation and switching threshold. Also we had carried on simulation experiments in the of four different equalization algorithms, and got the constellation diagram of the various algorithms and its residual error analysis. BER simulation of MCMA and DDLMS joint equalization algorithm in different SNR were under analyzed for different iteration step length factor are simulated. Finally we got the residual error of different transform curve. Simulation results obtained the following conclusions: 1. MCMA and DDLMS joint equalization algorithm can get smaller residual error iteration, also can effectively repair the transmission process of phase shift。 2. In White Gaussian Noise channel, under the condition of the same SNR, MCMA and VI DDLMS joint equalization algorithm can achieve lower bit error rate than the CMA algorithm. In the case of high signal noise ratio (SNR), MCMA and DDLMS joint equalization algorithm can be very close to 16QAM theoretical bit error rate。 3. In MCMA and DDLMS joint equalization algorithm, the larger step factors of the MCMA algorithm can quickly converge to the DDLMS threshold switching value. Then we use DDLMS to get smaller residual error convergence, so as to achieve faster convergence. It overes the defect of only using MCMA algorithm. Also step factor cannot be too much which will cause residual error cannot converge to the DDLMS switching threshold. Through the research of the equalization algorithm, we provide more theory and basis of the receiver equalizer in munication applications. Blind equalization algorithm is widely used in many fields, such as munication, sonar, radar, geological exploration, biomedical engineering, control engineering, etc. It has broad application prospect and huge application market. Blind equalization algorithm has bee a munication detection theory, digital signal processing in popular subjects such as the research object. In the future it will develop into a more superior blind equalization algorithm in order to meet the needs of industries and applications. Based on the research of the equalization algorithm, we provide the receiver equalizer in munication applications for more theory and basis. Keywords: Modify Constant Modulus Algorithm( MCMA) ; DecisionDirected LeastMean Square ( DDLMS ) ; Blind Equalization Algorithm ; Joint Equalization VII 目 錄 摘要 ................................................................................................................