Adaptive Coding For Wireless Video Transmission Based On Cross-Layer Feedback

With the rapid growth of wireless communications and the emerging demands for high quality of wireless video communications, how to provide proper video services no matter where or when becomes a hot research topic nowadays. However, the dynamics and heterogeneity of wireless networks, high error rate, the unpredictability of fluctuation in channel bandwidth, and the sensitivity to error in compressed video streams all make their influence on the quality of wireless video communications; such influences make wireless video transmission face enormous challenges.So far, the traditional hierarchical structure of communication systems (OSI standard) can be well applied in the wired network. This design greatly simplifies the network modeling with strong robustness; however, it lacks the flexibility to adapt to the fading wireless channel. Therefore, in order to better adapt to fluctuating channel conditions, the cross-layer design was proposed in many academic papers. But as many conventional schemes were implemented solely at different layers, the performance was quite limited. By taking parameters from different layers into account, the cross-layer structure design optimizes them according to system constraints and network characteristics to achieve efficient resource allocation and obtain overall high performance of transmission system. In this thesis, we choose H.264/AVC video standard for the study. We apply our proposed cross-layer feedback design on the H.264/AVC standard to improve the quality of reconstructed video by modifying the application-layer video coding algorithm.First, a single cross-layer feedback based video coding scheme is proposed in this paper. This scheme adopts both adaptive modulation and coding (AMC) and automatic repeat request (ARQ) techniques to provide an adaptive wireless video coding solution. Within this scheme, timely channel-awareness can be obtained by the feedback information from the PHY layer. Then the video encoder at application (APP) layer can take accurate rate control to adapt to the varying channel bandwidth. Experimental results show that compared to the existed schemes without feedback and channel-awareness, our proposed method can get a significant improvement in Peak Signal-to-Noise Ratio (PSNR) for wireless video transmission.Second, we further developed a two-level cross-layer feedback based method. On the basis of the single-level feedback described above, we add the feedback information from the Data link (DL) layer to the Application (APP) layer to guide the proper Reference Picture Selection (RPS) so as to reduce the video quality degradation brought by burst fame losses. By combining these two-level feedbacks, the proposed scheme can get a much better performance.Finally, a cross-layer feedback based method considering the features of handover is proposed in this paper to solve the severe packet loss problem during handover. We can get the channel SNR threshold at the PHY layer when the handover happens, then send this threshold to modify the RPS mechanism at the APP layer. Experimental results show that our method can efficiently improve the video transmission quality during handover.