Optimization And Implementation Of MPEG-4 Decoder In Embeded System

With the rapid development of computer, communication and electronic technologies, our society has entered a digital era. And the digital multimedia technologies make our life more colorful. In digital multimedia technologies, video compression technology is the core. By far, there have been several international standards in video compression technology, they are MPEG-1, MPEG-2, MPEG-4, H.261, H.263 and H.264, etc. Among these standards, MPEG-4 Simple Profile has the advantage ofbetter video quality under the situations of low bit rate and high compression ratio and is very suitable for embedded system applications. There is great potential in using embedded systems to implement MPEG-4 SP video codec. And the goal of the researches in this paper is to optimize an MPEG-4 SP decoder and make it have better decoding frame rate on an embedded platform.Firstly, this paper describes in detail the MPEG-4 video coding standard and its main technologies regarding video decoding such as inverse quantization (IQ), IDCT, motion compensation and decoding, etc. They are not only the key technologies of video decoding, but also the theoretic basis of the further works on decoder optimization.Next, this paper gives an all-around introduction to embedded systems. Based on this introduction, the embedded hardware environment and embedded operation system which can be used to show the value of the research in this paper are chosen.Thirdly, the computational complexity of the MPEG-4 SP decoder is analyzed under ADS1.2, The optimization and testing of MPEG-4 decoder is also carried out in the same environment. The main optimization strategies include: adoption of an improved IDCT algorithm—the row-column based FIDCT, optimization of MPEG-4 SP decoder structure, elimination of decoding redundancy, and the optimization motion compensation. By comparing the results before and after optimization, it is found that the performance of decoder is greatly improved.Finally, the optimized decoder is partly modified to compile successfully under the arm-linux-gcc environment, and the executable decoder after compilation is transplanted into the destination platform.