基于丢包率预测的视频传输纠错算法

发布时间：2018-12-12 16:34

【摘要】：为解决视频图像在互联网中进行传输时,其质量易受网络丢包率、时延等因素的影响而显著降低的问题,提出了一种基于丢包率预测的视频传输纠错算法。该算法采用隐马尔可夫模型预测网络丢包率,根据丢包率的大小自适应地选择FEC或ARQ对视频图像进行纠错操作。当预测出的丢包率较高时,为避免FEC算法在丢包率较高时降低带宽利用率,采用选择性ARQ算法恢复丢失的视频数据包,并通过限制其重传次数使视频传输的实时性得到了保证;当预测出的丢包率较低时,则采用优化了RS冗余值的FEC算法进行纠错操作。在OPNET modeler中进行的仿真实验表明,与HARQ算法相比,使用该纠错算法,视频图像的PSNR的平均值提高了1.6 dB,平均时延减少了0.24 s左右。该算法不但降低了视频传输的平均时延和丢包率,而且提高了接收端视频图像的重建质量,具有复杂度低、实现简单的特点。
[Abstract]:In order to solve the problem that the quality of video image is easily reduced by the network packet loss rate and delay, a video transmission error correction algorithm based on packet loss rate prediction is proposed. The algorithm uses the hidden Markov model to predict the packet loss rate of the network and adaptively selects FEC or ARQ to correct the error of the video image according to the size of the packet loss rate. When the predicted packet loss rate is high, in order to avoid the FEC algorithm reducing bandwidth utilization when the packet loss rate is high, the selective ARQ algorithm is used to restore the lost video packets, and the real-time performance of video transmission is guaranteed by limiting the number of retransmissions. When the predicted packet loss rate is low, the FEC algorithm, which optimizes the redundant value of RS, is used to correct errors. The simulation results in OPNET modeler show that compared with the HARQ algorithm, the average PSNR of the video image is increased by 1.6 dB, and the average delay is reduced by 0.24 s compared with the HARQ algorithm. This algorithm not only reduces the average delay and packet loss rate of video transmission, but also improves the quality of video image reconstruction at the receiving end. It has the advantages of low complexity and simple implementation.
【作者单位】：广东工业大学自动化学院;铜仁学院数学与信息工程系;
【基金】：国家自然科学基金青年科学基金资助项目(61201392) 铜仁学院科研启动基金资助项目(TS10014)
【分类号】：TN919.8

【参考文献】