题名数字电视地面广播中均衡器的研究
作者武晓华
学位类别博士
答辩日期2007-05-25
授予单位中国科学院声学研究所
授予地点声学研究所
关键词数字电视地面广播 判决反馈均衡器 误差传播 分数间隔均衡器
其他题名Research on Equalizers in Digital Television Terrestrial Broadcasting
学位专业信号与信息处理
中文摘要数字电视广播是电视广播发展史上的一次历史性飞跃。数字电视地面广播以其能够满足现代信息社会“信息到人”需求的特点,在数字电视体系中占据了非常重要的地位。经过多年研究,数字电视地面广播基本形成两大发展道路:基于OFDM技术和基于VSB技术。在OFDM系统中,以多载波传输结合FFT技术来对抗信道衰落;在VSB系统中,以复杂的自适应均衡器来对抗信道衰落。 本文主要研究了VSB系统中使用的判决反馈均衡器(DFE),并重点研究了如何提高DFE的性能。DFE是VSB接收机中最重要的组成部分,其决定了接收机对抗信道衰落的能力,不同接收机的设计主要区别在DFE的设计上。 本文的主要贡献和创新工作如下: 1. 误差传播是DFE性能提高面临的最大障碍之一,本文针对误差传播的产生、传播及抽头权值更新三个环节,提出了三种减小误差传播的新算法和一种快速仿真算法。 1)提出一种使用Lagrange乘子法限制抽头权值的新算法。该算法在均衡过程出现误差传播后,通过给DFE的反馈部分抽头权值加恰当的限制,达到限制误差传播的目的。仿真表明,与不使用限制算法相比,使用该算法后突发错误总数减少为原来的25.16%。 2)提出一种将优化软判决器用于DFE的新算法。该算法使用软判决,为下一步的处理提供了额外信息,从而可以达到减小误差的目的。仿真表明,该算法使突发错误总数减少为原算法的15.28%。 3)提出一种针对原有盲均衡算法的改进算法。该算法通过判断均衡算法输出是否可信,避免将错误权值更新信号引入系统,从而减小了误差传播对抽头权值更新过程的影响。仿真表明,该算法使突发错误总数减少为原算法的11.37%。 4)提出一种用于误差传播仿真的快速仿真算法。使用该算法使仿真过程中误差传播出现概率大大增加,从而大大节约了仿真时间,提高了工作效率。 2. 分数间隔均衡器对位定时要求低,可以补偿严重的时延和幅度失真,为系统提供更低的均方误差。本文提出一个适用于VSB系统均衡器使用的分数间隔均衡器方案,论证了分数间隔均衡器不仅适用于DFE的前馈部分,也适用于反馈部分的观点。与通常的设计相比,使用该方案后系统性能有了显著的提高,在低信噪比的情况下,系统的均方误差甚至减小了0.3。
英文摘要Digital television broadcasting is a revolution in the history of television broadcasting. Digital television terrestrial broadcasting can meet the demand of “information to people” in the modern information society and plays a very important role in the system of digital television. After a long time development, digital television terrestrial broadcasting evolves along two directions: OFDM technology based and VSB technology based. In an OFDM system, multicarrier and FFT technology are used to resist channel fading. In a VSB system, complex adaptive equalizer is used to resist channel fading. In this thesis, decision feedback equalizers (DFE) in VSB systems are researched, and how to improve the performance of DFE is further researched. DFE is the most important part of a VSB receiver, it decide the ability of receivers in resisting channel fading. The mainly difference between different receivers is the design of DFE. The main contributions and innovations of this thesis are shown below: 1. Error propagation is one of the main obstacles that DFE suffered. In this thesis, to improving performance over reducing error propagation in DFE, three new algorithms are proposed corresponding to the following considerations: producing and propagation of error propagation, and updating process of tap weight. A new fast simulation algorithm is also proposed. 1) A constraint algorithm optimized by the Lagrange multiplier method and imposed on the feedback equalizer is proposed. In this algorithm, an appropriate constraint is imposed on the tap weight of the feedback equalizer when error propagation occurs in equalization process, then constraints the propagation of error. Simulation results show that after using the proposed constraint algorithm, the total number of burst error is reduced to 25.16% when compared with the algorithm without constraint. 2) An optimized soft decision algorithm is proposed. In this algorithm, soft decision is used, so extra information is provided for the processing of next step, then error can be reduced. Simulation results show that after using the proposed soft decision algorithm, the total number of burst error is reduced to 15.28% when compared with the old algorithm. 3) An improved algorithm based on the old adaptive algorithm is proposed. It judges if the output of equalization algorithm is credible, avoids importing the error tap updating signal into DFE, then reduces the effect of error propagation to tap weight updating process. Simulation results show that after using the proposed new algorithm, the total number of burst error is reduced to 11.37% when compared with the old algorithm. 4) A new fast simulation algorithm used in the simulation of error propagation is proposed. Through making error propagation occurring more frequently, the new algorithm saves the simulation time greatly and improves the efficiency of research. 2. A new fractionally spaced equalizer scheme suit for equalizer in VSB system is proposed, and the opinion that fractionally spaced equalizer not only fit for feedback equalizer but also fit for feedforward equalizer is introduced and demonstrated. Fractionally spaced equalizers demand less on symbol synchronization, and can compensate severe distortion of delay and amplitude. So they can provide lower mean square error. Compared with traditional designs, the new scheme has distinct improvement in performance. Mean square error of system can be even reduced 0.3 when SNR is low.
语种中文
公开日期2011-05-07
页码137
内容类型学位论文
源URL[http://159.226.59.140/handle/311008/178]  
专题声学研究所_声学所博硕士学位论文_1981-2009博硕士学位论文
推荐引用方式
GB/T 7714
武晓华. 数字电视地面广播中均衡器的研究[D]. 声学研究所. 中国科学院声学研究所. 2007.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace