Wireless Video Multicast Key Techniques Research

Wireless video multicast is a newly arising technique for video transmission. It’s mainly focus on the situation that many people are interesting in the same video in a same area. However, the wireless channel characteristic is time-varying and the terminals are moving frequently, therefore, the different terminals in a same multicast group usually have the different channel conditions, which will make the sender difficult to fulfill all the receivers at the same time. If the traditional video coding applied directly to the wireless video multicast environment, it will have unpredictable consequences, such as cliff effects, i.e., when the SNR of the received signal is less than a certain threshold, the decoded video quality dropped dramatically.In this dissertation, we state the existing problems in wireless video multicast in detail and try to explain these problems form the point view of the deviation of the application layer and the physical layer designing. Besides, we have studied the wireless link adaptive technology that is widely used in current wireless communication networks and the error estimated encoding (EEC). Then, we combine these two technologies together to improve the performance of wireless link adaptive scheme. Further, we amend the EEC through the theoretical and experimental analyses by setting different estimation intervals with different coefficients. After that, we successfully make EEC achieve high estimation accuracy with smaller overhead.In addition, we try to use a new method based on real-value representation to solve the robustness and scalability problems existed in wireless video multicast. Firstly, we describe Soft-Cast, which is one of the real-value representation methods in detail and state its system processes form the aspects of encoder and decoder. Further, we propose an improved decoding scheme for Soft-Cast after analyzing its decoding process. Compared with the original Soft-Cast decoding algorithm, our method can simplify the process of decoding and better handling the decoding failure problem caused by critical information loss.