Svc Unequal Error Protection In Wireless Channel

Recently, the two major international video coding standards organizations developed the latest video coding standards, H.264. Because of the difference of the transmission network and the different processing capabilities and the terminal display ability, the scalable extension standard (SVC) of the H.264 was developed. It can make the same source to transport in different network, display and reconstruct in different processing capabilities terminal.From the perspective of information theory, the joint source and channel coding method has the better transmission performance than other means in the communication system. It can comprehensively consider the two conditions of the source and channel, and achieve the overall optimization of transmission performance, get the best transmission quality. As the SVC source is temporal scalable, resolution scalable and quality scalable, so it is composed by different levels. Based on the different importance of the scalable video coding information, this paper proposes one joint source and channel coding method, which can distinguish the different importance of the data, using the unequal error protection method to transport data, and gives the theoretical basis and experimental results. The main content of the text is summarized as follows:Firstly, the research status quo of the wireless video transmission technology is introduced, and the key technologies of the video coding standard, the temporal scalability, resolution scalability and quality scalability of the scalable video coding are given.Then, from the information theory view, the paper analyzes joint source and channel coding theory and gives theoretical basis of its superior performance in the communication system. So the joint source and channel coding for SVC is taken into account.In this paper, a joint source/channel coding method using low density parity check codes (LDPC) is proposed for transmission of the scalable video coding extension of H.264/AVC over noisy channels. It takes a group of picture (GOP) as a unit, using the two level method of inter-frame bit budget and intra-frame code-rate allocation, achieving the optimal code-rate allocation. Based on the temporal hierarchy prediction structure, the method of inter-frame bit budget can optimize the end to end distortion performance. Meanwhile, the method of intra-frame code-rate allocation makes the resolution level and quality level as a unit, using the simple search arithmetic choose the LDPC channel coding rate, achieving the optimal unequal error protection for different level in one frame.The paper gives the system structure diagram of the transport process, and introduces the basic theory of each part briefly. At last the simulation experiment is given, the configuration of the experiment parameters are listed in detail, and obtain the reconstructed performance curve. The result shows that the method can reasonably change the bit budget and coding code-rate allocation, when bit rate and Signal-to-Noise condition is changed. Compared with other method, we can see that the method can effectively protect the basic and the lower layer data in the low Signal-to-Noise and low channel bandwidth, and reduce the end to end image distortion; the reconstruct image quality is superior to EEP method.