An area-efficient IDCT architecture for multiple video standards

In this paper, we introduce a unified 1-D IDCT architecture for a video decoder supporting multiple standards including AVS, VC-1, MPEG-4 and H.264. The low-cost architecture is implemented by the method of matrix decomposition and sharing resources such as adders and shifters as far as possible. We make use of the coefficients in AVS and VC-1 to reduce the complexity of IDCT based on MPEG-4 greatly. And multipliers are replaced by shifters and adders, which are much more area-efficient. Researching on the relationship of coefficients among AVS, VC-1, and MPEG-4, we propose a low-cost hardware sharing architectures requiring lower hardware cost than that of the individual and separate implementations.

[1]  Bin-Da Liu,et al.  High throughput 2-D transform architectures for H.264 advanced video coders , 2004, The 2004 IEEE Asia-Pacific Conference on Circuits and Systems, 2004. Proceedings..

[2]  Henrique S. Malvar,et al.  Low-complexity transform and quantization in H.264/AVC , 2003, IEEE Trans. Circuits Syst. Video Technol..

[3]  Kyeongsoon Cho,et al.  Architecture of transform circuit for video decoder supporting multiple standards , 2008 .

[4]  Ieee Standards Board,et al.  IEEE standard specifications for the implementations of 8x8 inverse discrete cosine transform , 1991 .

[5]  Seonyoung Lee,et al.  Design of Transform and Quantization Circuit for Multi-Standard Integrated Video Decoder , 2007, 2007 IEEE Workshop on Signal Processing Systems.

[6]  J. Niittylahti,et al.  Row-column decomposition based 2D transform optimization on subword parallel processors , 2005, International Symposium on Signals, Circuits and Systems, 2005. ISSCS 2005..