Error-resilient packet reordering for compressed video transmission over error-prone networks

Abstract

In wireless video transmission, the bursty packet losses of the channel will severely impair the reconstructed video quality. This paper presents an error-resilient packet reordering scheme to improve the error robustness of the transmitted video. Firstly, we propose a compressed domain transmission distortion model to efficiently estimate the importance of every video packet. Then based on this importance index, we create an intelligent transmission order for all packets in a transmission group to alleviate the impact of bursty losses on the video quality. Simulation results demonstrate that the proposed scheme can improve the reconstructed video quality by up to 1.8 dB compared with conventional methods. Moreover, the computational complexity of the proposed method is very low thus it is highly desirable for practical video streaming applications.