Dissertation > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Image communication, multimedia communication

Research on Adaptive Video Communication over Wireless Channels

Author QiuJinBo
Tutor ZhuGuangXi
School Huazhong University of Science and Technology
Course Information and Communication Engineering
Keywords Wireless Coummnication Video Communication Cross-layer Adaptive Adaptive Modulation and Coding Adaptive OFDM Unequal Protection Error Control
CLC TN919.8
Type PhD thesis
Year 2006
Downloads 674
Quotes 1
Download Dissertation

Despite the rapid growth of wireless communications, wireless video transmission still faces enormous challenges. The dynamic and heterogeneity of wireless networks, high error rate, the unpredictability of transfer links, and the sensitivity to error in highly compressed video streams all make their influence on the quality of wireless video communications. The classic layered protocol architecture uses static communications between layers, thus not able to make full use of limited resources of spectrum or power. In order to deal with these problems and to satisfy the QoS demanded by various applications, innovations have to be made to traditional layered architecture. Cross-layer interaction for adaptive resource allocation becomes a hot topic of next generation wireless network protocols. The idea of cross-layer interaction does not deny the classic network model. Instead, by taking parameters from different layers into account, it designs and optimizes them according to system constraint condition and network characteristics, to achieve efficient resource allocation and overall performance of networks. Cross-layer design makes it possible to satisfy video applications under very rigor demands. Imbursed by the item“Research of Future Communication System Fundamental Theory and Technology”, the dissertation studies several key techniques of adaptive video communication over wireless channels.Highly compressed video bitstream is very sensitive to transmission errors. We have to add redundancies either in video coding or in transmission in order to enhance the error resilient ability of the compressed bitstream and avoid the quality degradation. These redundancies enhance the robustness of video bitsteam at the cost of decreasing coding efficiency. In this thesis, we introduce the end-to-end video communication model firstly, and then analyze the video coding distortion, channel distortion and the forecast of the following frames. Model the relationship of video coding and channel distortion with the redundancies. We study the joint source and channel coding and present the target function of cross layer joint optimize.We proposed two different cross-layer optimize method to cope with video communication over flat fading channels and frequency selective fading channels. We studied various adaptive transmission schemes and corresponding spectrum efficiency, analyzed and simulated the performance of adaptive MQAM over Raleigh fading channel. We present an adaptive bottom up cross layer method to optimize the parameters at the physical layer, the link layer and the application layer jointly. At physical layer, AMC is adopted to take advantage of time-varying of the channel for a coarse data rate selection and provide a high average spectral efficiency. The adaptive cross layer controller chooses an optimized parameter combination of the other layers to obtain a fine tradeoff between source distortion and channel distortion to satisfy the QoS constraints.Adaptive OFDM can transform the frequency selective fading channel to a group of flat fading sub channels. These flat fading channels have different channel conditions. We can get an unequal protection to the video bitstream with different importance by mapping them to sub channels with different channel conditions. We studied the adaptive OFDM under frequency selection fading channel, analyzed and simulated three kinds of adaptive algorithms for AOFDM. Meanwhile, we studied the layered scalable video coding, modeled the relationship of distortion for coding and transmission of FGS video coding. We proposed a top down cross layer unequal protection method on base of them. The FGS video encoder at the application layer generates base layer and enhancement layer bitstreams with different importance. And the importance of each packet is transferred to the lower layers through top down cross layer information. The link layer and physical layer adjust the retransmission strategy and sub channel allocation according to current channel condition and these importance information. This cross layer adaptation method provides levels of protection to different parts of FGS bitstream which have unequal degrees of importance and thus improve the video quality.Most researches on error control for video communication are focused on how to improve the average quality. However, the quality jitter is also one of the key issues for visual impression. We investigated the quality degradation due to channel distortion and their propagation. Then we proposed an importance metric parameter for macro block, and based on this parameter, we proposed a video packetization method. Our method can minish the quality jitter for each packet error. We also proposed a bidirection PGOP method. Our improved method can adjust the direction of intra macro block refresh according to the variation of the scene. The bidirection PGOP enhances the coding efficiency while maintains the error resilient of the traditional PGOP.Lastly, we discuss the application of video codec method for interactive distance education system. We proposed a novel three level architecture to achieve multimedia communication across heterogeneous networks.

Related Dissertations
More Dissertations