蜂窝网络环境下多速率视频多播中视频质量补偿机制研究

Aim to the quality imbalance problem which widely exists in multi-rate video multicast systems, this project base on in-depth study of the characteristic of multi-rate video streams, cellular networks, mobile channel and user’s behavior, firstly explore the user perceived video quality compensation mechanism, and establish a multi-user cooperative cognitive relay based user perceived video quality compensation model for multi-rate video multicast; Secondly, to against the channel state changing of the cognitive link, study efficient relay channel selection schemes and low-latency channel switching algorithm, to reduce the channel access delay in relay process; Thirdly, consider the characteristic of scalable video coding and energy-constrained of mobile devices, study efficient relay rate schedule model and cooperative power allocation model, to improve the system performance; Finally, consider the delay-sensitive of video stream, combine with the characteristic of proposed models, study the efficient optimal algorithms. The proposed project not only to provide a theoretical basis and practical guidance for the next generation of mobile video multicasting standards development, and can make the next generation of video multicast service to get breakthrough.

针对多速率视频多播中广泛存在的用户收视质量不均衡问题，本项目拟在分析研究视频流特征、蜂窝网络特征、通信信道特征以及用户行为特征的基础上，首先探究移动视频多播中的用户收视质量补偿机制，构建基于认知协同中继技术的用户收视质量补偿框架模型；其次针对认知链路多变的特点，重点研究高效的中继信道选择算法及低时延的信道切换方案，以减少中继过程中的信道接入时延；同时，针对可分级编码流的特征及移动设备能量受限的特点，研究高效的中继速率调度与协同功率分配算法，以提高中继效率和减少能耗；最后，针对视频流时延敏感的特点，结合所提质量补偿机制中相关理论与技术，寻求高效的优化算法。本项目的研究成果，不仅为下一代移动视频多播标准的制定提供理论依据与实践指导，而且可以使得下一代视频多播业务获得突破性的进展。

本项目主要研究了蜂窝网络环境下小区边缘用户的视频质量补偿机制。三年来，项目组根据研究内容开展了该项目的各项研究，圆满的完成了研究计划。在研究目标上，建立了蜂窝网络环境下移动视频多速率视频多播中多用户认知协同中继传输模型；建立了时延约束下认知中继信道选择与资源最优调度模型；研究了多用户认知协同中继补偿机制模型的优化求解算法，并验证了所建立的协同中继补偿机制模型和相关求解算法的有效性。在项目执行期间，项目组共计发表相关论文13篇，其中SCI论文3篇，EI论文2篇，中文核心5篇，会议论文3篇。另外申请发明专利8项，授权软件著作权2项，获得湖北省科技进步2等奖1项，培养了8名硕士，其中3名硕士已经毕业，另有2名硕士即将毕业。