【摘要】 为解决现有多源流化方案存在的无法适应Internet网络异构性和动态性的问题,提出了一种新的多源流化方案。其中,传统的细粒度可扩展编码被改造成一种适合于多源流化的多子流传输模型。在此基础上,提出了基于优化峰值信噪比(PSNR)的多源流化码流分配算法。实验表明,较之平均分配码流,该算法能在相同条件下将视频的图像质量提高约1dB。整个多源流化方案对于异构而动态变化的In-ternet网络环境具有良好的适应能力,可用于在内容分发网络(CDN)中提供流媒体服务。更多还原

【Abstract】 In the best-effort Internet, single source video streaming often suffers from unreliability and bandwidth bottleneck along a single route. Multiple source video streaming has been suggested to improve delivery but most proposed schemes can not adapt to the heterogeneity and bandwidth variations of the Internet. This paper presents a method for multiple source video streaming in which the traditional fine granular scalable coding is changed into a multiple sub-streams-based transmission model. The performance of a peek signal-to-noise ratio (PSNR)-based stream rate allocation algorithm for multiple sources video streaming was 1 dB better than the average rate allocation method. This scheme can efficiently combine the available bandwidth of multiple routes to bypass the bandwidth bottleneck along a single route and smooth the jitter to achieve high quality streaming video.更多还原