通信学报

• 学术论文 • 上一篇    下一篇

基于半马尔科夫决策过程的视频传输拥塞控制算法

田 波,杨宜民,蔡述庭   

  1. 广东工业大学 自动化学院,广东 广州 510090
  • 出版日期:2014-08-25 发布日期:2014-08-15
  • 基金资助:
    国家自然科学基金资助项目(61201392)

Semi-Markov decision process based congestion control algorithm for video transmission

  • Online:2014-08-25 Published:2014-08-15

摘要: 针对传输控制协议(TCP, transmission control protocol)的拥塞控制算法未能满足视频传输质量要求的问题,提出了一种基于半马尔科夫决策过程的视频传输拥塞控制算法。首先, 为克服目前基于峰值信噪比的视频质量评估方法实时性低的缺点,设计了一种可在线运行的无参考视频质量评估方法。其次,根据接收端视频质量的反馈,采用半马尔科夫决策过程对拥塞控制进行建模,并通过求解此模型得到拥塞控制参数的调整策略。仿真实验结果表明,与目前典型的拥塞控制算法相比,该算法不但具备更好的TCP友好性,而且有效地提高了解码后视频序列的主观和客观质量。

Abstract: Due to the problem that the congestion control algorithm of transmission control protocol (TCP) cannot meet the requirements of quality of experience for Internet video transmission, a semi-Markov decision process based congestion control algorithm for video transmission was proposed. First, in order to improve the accuracy of video quality assessment, an online operation no-reference video quality assessment was presented. Then, the congestion control problem was formulated as a semi-Markov decision process according to the feedback of video quality assessment, and the optimal parameters of congestion control were obtained by solving the semi-Markov decision process. Simulation results reveal that proposed algorithm achieves better subjective and objective video quality, and obtains well TCP-friendliness ratio, compared with the conventional congestion control algorithms.

No Suggested Reading articles found!