I. Introduction
The state-of-the-art video coding standard H.264/AVC [1], which provides high compression efficiency and friendly network access capability, has been considered as one of the most suitable video coding standards for wireless video transmission [2], [3]. However, due to the predictive coding and the variable length coding (VLC) used in H.264/AVC, the compressed video streams are very sensitive to the transmission errors. In the spatio-temporal prediction framework, a single error may result in obvious reconstructed video quality degradation of the current frame, and the effects may propagate to the successive frames. Likewise, because of the utilization of VLC, a single bit error can cause the decoder to lose synchronization. As a result, the subsequent correctly received bits can become useless [4]. Thus, to minimize the impact of the transmission errors on the reconstructed video quality, error resilient tools, such as data partitioning and flexible macroblock ordering (FMO), are utilized to improve the robustness of the video streams during transmission. Meanwhile, unequal error protection (UEP) can be implemented by exploiting the characteristics of the unequally distributed significance of bits in the compressed video streams, i.e. bits with higher significance are assigned more protection than those with less significance during transmission. Since the reconstructed video quality would be severely degraded when errors occur in those bits with higher significance, the UEP scheme is expected to lead to an improved reconstructed video quality.