基于HTTP流化的自适应码率混合控制算法

HTTP流化自适应流媒体技术已逐渐应用于媒体视频传输. 提出一种新的自适应码率控制算法, 将视频资源缓存、网络吞吐量共同作为混合控制的主要参数. 在视频资源缓存控制机制中, 通过采用缓存空间均衡的方式, 调整单个视频分片进入缓存区所消耗的时间, 使视频数据总量保持在均衡范围内. 在网络吞吐量控制机制中, 根据当前网络状况, 利用保守策略提升视频码率, 并利用逻辑斯谛方程确定码率下调幅度. 通过使用双参数进行码率决策, 能有效避免单一使用某一个参数而产生的决策精度上的不足, 从而实现控制性能最优. 在实际网络带宽情况下, 该算法已在使用MPEG-DASH标准的参考平台上得到验证.

Adaptive Hybrid Rate Control Algorithm Based on HTTP Streaming

Dynamic adaptive streaming over HTTP streaming has gradually been employed for video content delivery over internet. A novel adaptive rate control algorithm is proposed, in which video resource cache and network throughput are the main parameters for hybrid control. In the video resource cache control mechanisms, the method of balancing cache space is used to adjust consumed time of a single video segment pouring into the cache. This way, the total amount of video data can be maintained in a balanced range. In the network throughput control mechanism, under the current network conditions the media bitrates is switched up and detected using a conservative strategy, and the logistic equation determines bitrates reduction. Compared with single parameter, two parameters improve the accuracy, and optimal control performance can be achieved. The proposed algorithm has been verified on the MPEG-DASH standard reference platform.