FGS Video Coding And Transmitting In Wireless IP Networks

With the rapid development of information technology and the great progress of society, the demand for obtaining information has become greater and greater. Therefore, it is not surprising that people hope to obtain information in mobile. User will be provided with services of different contents and qualities by way of data and multimedia application on wireless terminal. The above will be the shows of the future of wireless communication. Video is a very important part of multimedia data. Wireless video coding and transmitting technology has become the biggest buzz words of present multimedia communication research area.Modern network is entering the era of developing wireless IP network, due to the different types of wireless networks are converging into all IP networks today. The proposal of the concept of IP networks is both the trademark of IP network development and the milestone of wireless network development.However, heterogeneity of wireless IP network challenges video communication greatly. To meet the challenge, the following should be taken into serious consideration: Firstly, the wired link of wireless IP network has bandwidth capacity and bandwidth change which are unsymmetrical to wireless link. Secondly, which the error of transmitting video data is variant and time varying. Thirdly, the delay and jitter of transmitting video data are constrained. Finally, guarantee of QoS is complicated.The laboratory, at which the author works, is undertaking the national high technology research and development program of China (No.2002AA119010) and the National Natural Science Foundation of China (No.60496315). Based on the research work at the laboratory, the author makes a detailed and profound analysis of video coding and transmission system of wireless IP network, thesis tries to find out solutions to the problem described above.From the angel of system, joint source-channel coding is adopted to optimize rate allocation so as to increase self-adaptation of video transmitting system. In addition, FGS video coding is taken to adapt to the different bandwidth and variation of bandwidth. Multiple products coding scheme is suggested to correct the hybrid errors of wireless IP network. Furthermore, FEC is adopted to decrease the delay for real time video application.Angled at the fatal weakness of low coding efficiency of FGS and PFGS video coding, based on its analysis and research, a new coding scheme of the sign of residue coefficient in bit-plane is advanced, which raises coding efficiency to a certain degree. In the meantime, bitstream structure has been greatly improved by classifying it into two grades: two-grade and three-grade bitstream structure. It not only raises error resilience of bitstream, but also is required by the following various packaging schemes in multiple product FEC. Combing source rate-distortion theory, rate-distortion model of FGS enhancement layer is introduced. Rate allocation arithmetic among frames of FGS enhancement layer is discussed so as to pave the way for future joint optimization.Owing to the above-mentioned improved bitstream structure, two transmitting package scheme are suggested correspondingly: lengthened package and equally-lengthened package; to correct the hybrid errors of wireless IP network and realize the real time video application, multiple product FEC (MPFEC) scheme could be one suitable solution. It requires the realization of intra-package FEC by way of containing RCPC in the data link layer, of inter-package FEC by way of using systematic RS coding in transport layer. The model of estimating transmission distortion at sending end has been set up.Based on the general rate distortion theory, combined with FGS technology at source and MPFEC transmitting strategy, adopting joint source-channel technology, FGS enhancement layer rate and channel rate optimization allocation arithmetic is put forward. Experimental imitating results prove it to be a very effective way to the practical transmission.Considering the problem of base layer (or single layer) video transmission, error propagation is avoidless duo to inter-roll dependency in which motion-compensation predicating coding technology results. Constrained by target rate, channel coding performance, error propagation attenuation and distortion function are all modeled to decrease the complicated computation. Experiments prove that under given condition computing results of these models very approximate to those of the practical simulation results. Particularly, the introduction of attenuation model of error propagation and construction of new rate distortion model has simple greatly the generating of rate-distortion surface.