【作者】 李鹏；

【导师】 常义林； 卢朝阳；

【作者基本信息】 西安电子科技大学 ， 通信与信息系统， 2011， 博士

【摘要】 伴随着移动用户对无线视频业务需求的不断增长，视频业务本身具有的特性和无线信道固有的特点等因素决定了在无线环境中传输高质量的视频业务势必面临着巨大的技术挑战，如何提高无线视频传输的通信质量已成为通信领域需要迫切解决的问题之一。本论文针对如何提高基于OFDM/OFDMA技术的下一代无线通信网络中的视频通信质量进行了深入的探索与研究，主要的研究工作着眼于提高视频通信质量的OFDM无线传输技术。分别从结合视频信源特性的分级调制、跨层的分组调度和无线资源分配等方面进行了深入的研究。主要的研究成果包括：1.研究了用于无线视频传输的自适应分级正交幅度调制不等保护方法。针对基于OFDM调制的无线传输系统中不同子载波的信道质量存在由于无线信道的频率选择性衰落和时变特性导致的差异，论文提出了一种用于无线视频传输的自适应分级正交幅度调制不等保护方法，根据子载波的信道质量和视频数据对解码端恢复视频质量的重要性自适应地确定各个子载波上所使用的分级正交幅度调制参数，为码流中具有不同重要性的数据提供不等保护。仿真结果表明，相比于已有基于分级正交幅度调制的不等保护传输方法，所提方法可使恢复视频质量平均提高1dB。2.研究了用于无线视频传输的子载波分类与分级正交幅度调制相结合的不等保护方法。对于OFDM无线系统中不同子载波之间存在的信道质量差异，提出了一种子载波分类与正交幅度调制相结合的无线视频传输不等保护方法，通过将所有可用的子载波按照其质量分类，并采用不同的调制参数实现对视频码流中具有不同重要性数据的不等保护，避免了将码流中的重要数据映射至处于深衰落的子载波所导致的视频质量下降，同时克服了已有分级调制技术对视频码流中具有不同重要性数据须有固定比例的限制，提高了视频通信的质量。3.研究了用于无线视频传输质量优化的跨层分组调度和无线资源分配算法。为了提高现有分层无线网络中可用资源的利用率，改善视频通信的质量，提出了一种优化视频质量的跨层分组调度和无线资源分配算法。该算法以最小化网络中各用户的视频传输失真为目标，按照视频分组对解码端恢复视频质量的重要性对网络中不同用户的视频分组进行合理的调度，并根据调度的结果为不同用户分配可用的子载波以及发送功率，降低无线环境对不同用户视频分组丢失概率的影响，提高了视频通信的质量。4.对用于无线视频传输的跨层分组调度和资源分配算法中的公平性问题进行了研究。对于无线通信系统而言，不同用户之间的公平性是系统设计过程中必须考虑的重要因素之一。对于OFDMA无线系统中传输的已压缩编码的流媒体视频，结合所提出质量优化的跨层分组调度和无线资源分配算法，在考虑不同用户之间接受环境差异的基础上，提出了一种具有用户公平性的跨层分组调度和无线资源分配算法，在提高视频通信质量的同时保证处于不同接收环境的用户之间的公平性。更多还原

【Abstract】 With the increasing of the requirement for the video services in the wirelessnetworks, the inherent nature of the video services and the characteristics of the wirelesschannels determined that transmission high quality of video services over the wirelesschannel are bounded to face huge technical challenges. How to improve the quality ofthe video services in wireless environment has became an urgent problem in theresearch of wireless video communications. This dissertation is mainly concerned withthe design and implementation of various transmission schemes to improve the qualityof video services in the OFDM-based wireless transmission systems. The authors’ majorcontributions are outlined as follows:1. An unequal error protection (UEP) method for video transmission using adaptivehierarchical quadrature amplitude modulation (HQAM) is proposed to reduce theimpact of frequency-selective fading and the time-varying nature of wireless channels.Firstly, the video streams are divided into substreams with unequal priorities. Then, theHQAM modulation parameters of each subcarrier are adaptively determined based onthe channel quality of each subcarrier and the upper limit of the bit error rate of thesubstreams with unequal priorities. Finally, UEP is implemented via modulating videostreams onto subcarriers with different channel quality. Simulation results show that theproposed adaptive HQAM method reduces the impact of frequency-selective fading andthe time-varying nature of the wireless channels and improves the quality of thereconstructed video by an average of1dB compared with HQAM-based methods.2. For band-limited wireless video transmission systems with multi-subcarriers, anovel UEP strategy for video transmission over the wireless channel by using HQAM isproposed. Firstly, all the subcarriers available are divided into different classes based ontheir channel quality, and then the HQAM modulation parameters are determinedaccording to the channel quality of those subcarriers and the important of different partsof the video bitstream to the reconstructed video. The UEP of the video bitstream isimplemented by modulating different parts in the video stream to subcarriers withunequal HQAM parameters. Experimental results demonstrate that compared with thereported methods, the proposed method effectively improves the quality of thereconstructed video.3. The problem of packet scheduling and resource allocation for OFDMA wirelessvideo transmission is studied. The problem is first formulated as an optimizationproblem with the objective of minimizing the reconstructed video distortion of each user under the radio resource constraints. In order solve the optimization problem, across-layer optimization algorithm is proposed by taking the radio resource constraintsinto account. Simulation results shows that the proposed cross-layer algorithmoutperforms the state-of art packet scheduling and resource allocation algorithms by upto3dB in terms of the quality of the reconstructed video.4. Fairness across multi-user with different channel quality in the OFDMA wirelesstransmission system is an important issue during the packet scheduling and wirelessresource allocation. In order to maintain the fairness among users in the same cell, across-layer optimization algorithm is proposed by taking the radio resource constraintsand the fairness across multi-user into account. The situation that those subcarriers wthhigher channel quality are all taken up by users with better channel quality is avoided tomaintain the fairness. Simulation results show that the proposed cross-layer algorithmoutperforms the state-of-the-art packet scheduling and resource allocation whilemaintain the fairness across multi-users.更多还原