摘要
针对现有PANDA算法将平滑因子设为固定值的不足,提出一种基于PANDA算法的视频码率自适应切换算法。以缓存状态和网络吞吐量变化率动态地构建平滑因子,进一步通过指数加权移动平均(EWMA)对带宽做平滑处理;通过设立一个延缓片段请求发送策略,确保缓存区始终维持在一个均衡的状态,保证视频播放的流畅性。使用libdash对所提算法性能进行验证,实验结果表明,在时刻变化的网络环境中,所提算法能很好地应对网络中出现的这些变化。
Aiming at the deficiency of existing probe and adapt (PANDA) algorithm that sets the smoothing factor as a fixed va-lue, a video rate adaptive switching algorithm based on PANDA algorithm was proposed. The smoothing factor was constructed dynamically with cache state and network throughput rate of change. The bandwidth was further smoothed by an exponential weighted moving average smoother (EWMA ). A delay sending strategy of next fragment request was set up, to keep buffer level in a balanced range as far as possible and to keep video play smoothly. The algorithm was verified on the platform of libdash. Experimental results show that under constantly changing network environment, the proposed algorithm can perform well under varying network conditions.
作者
封红旗
王和
杨长春
顾寰
FENG Hong-qi;WANG He;YANG Chang-chun;GU Huan(School of Information Science and Engineering,Changzhou University,Changzhou 213164,China)
出处
《计算机工程与设计》
北大核心
2019年第6期1512-1517,1553,共7页
Computer Engineering and Design
基金
赛尔网络下一代互联网技术创新基金项目(NGII20160703)