| With the development of streaming service,high frame-rate ultra-high definition(UHD)real-time communication(RTC)service like Cloud Gaming and Virtual Reality is being widely used.Under this high frame-rate and high bitrate scenario,users could sometimes suffer from severe service degradation.Due to the fluctuations of frames incoming and decoding at the client-side,a decoder queue could be formulated before the streaming decoder at the clientside.Those fluctuations could easily overload the decoder queue and introduce a noticeable queuing delay for those queued frames,resulting in a higher total delay of user input to the corresponding image display to degenerate the user experiments.We demonstrate the severity of decoder queue overloading based on the massive real word high frame-rate UHD RTC streaming traces.Moreover,we figured out that severe decoder queue overloading is caused by video frame burst incoming and decoding speed degeneration,building the backbone of subsequent algorithm design.In this paper,we propose the Frame-Pausing and Frame-Skipping mechanism to reduce the queuing delay effectively by pausing the video frame production and actively managing the frames inside the decoder queue.We jointly optimize the enqueue rate of the decoder queue under Frame-Pausing and manage frames with skipping under Frame-Skipping to maintain the end-to-end delay while ensuring the decoding quality of video codec.We also mathematically quantify the potential performance with a queuing theory and Markovian chain.We evaluate the Frame-Skipping mechanism with our trace-driven simulation with real word UHD RTC traces.Our experiments demonstrate that Frame-Pausing can reduce the ratio of severe decoder queue delay by up to 4.3 × and the ratio of severe total delay by up to 1.9× compared to the WebRTC streaming algorithm.As for Frame-Skipping,the ratio of severe decoder queue delay can be reduced by up to 23×,and the ratio of severe total delay can be reduced by up to 2.6 ×. |
PDF Full Text Request