Variable Length Reconfigurable Algorithms and Architectures for DCT/IDCT Based on Modified Unfolded Cordic

A coordinate rotation digital computer (CORDIC) based variable length reconfigurable DCT/IDCT algorithm and corresponding architecture are proposed. The proposed algorithm is easily to extend to the 2 n -point DCT/IDCT. Fur- thermore, we can easily construct the N-point DCT/IDCT with two N/2-pt DCTs/IDCTs based the proposed algorithm. The architecture based on the proposed algorithm can support several power-of-two transform sizes. To speed up the computation of DCT/IDCT without losing accuracy, we develop the modified unfolded CORDIC with the efficient carry save adder (CSA). The rotation angles of CORDIC used in proposed algorithm are arithmetic sequence. For convenience, we develop the architecture of N-point IDCT with the orthogonal property of DCT and IDCT transforms. The proposed architecture are modeled with MATLAB language and performed in DCT-based JPEG process, the experimental results show that the peak signal to noise ratio (PSNR) values of proposed architectures are higher than the existing CORDIC based architectures at both different quantization factors and different test images. Furthermore, the proposed architectures have higher regularity, modularity, computation accuracy and suitable for VLSI implementation.

[1]  G.S. Moschytz,et al.  Practical fast 1-D DCT algorithms with 11 multiplications , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[2]  Earl E. Swartzlander,et al.  A scaled DCT architecture with the CORDIC algorithm , 2002, IEEE Trans. Signal Process..

[3]  Jianguo Liu,et al.  Fixed-point IDCT without multiplications based on B.G. Lee's algorithm , 2009, Digit. Signal Process..

[4]  M. Omair Ahmad,et al.  An Efficient Unified Framework for Implementation of a Prime-Length DCT/IDCT With High Throughput , 2007, IEEE Transactions on Signal Processing.

[5]  Constantinos E. Goutis,et al.  A fast DCT processor, based on special purpose CORDIC rotators , 1994, Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94.

[6]  Jürgen Götze,et al.  Low-complexity multi-purpose IP Core for quantized Discrete Cosine and integer transform , 2009, 2009 IEEE International Symposium on Circuits and Systems.

[7]  Kaushik Roy,et al.  A low power reconfigurable DCT architecture to trade off image quality for computational complexity , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  Hai Huang,et al.  A novel VLSI linear array for 2-D DCT/IDCT , 2010, 2010 3rd International Congress on Image and Signal Processing.

[9]  N. Ahmed,et al.  Discrete Cosine Transform , 1996 .

[10]  B. Lee A new algorithm to compute the discrete cosine Transform , 1984 .

[11]  Liang-Gee Chen,et al.  A cost-effective architecture for 8×8 two-dimensional DCT/IDCT using direct method , 1997, IEEE Trans. Circuits Syst. Video Technol..

[12]  Shen-Fu Hsiao,et al.  Design and implementation of a novel linear-array DCT/IDCT processor with complexity of order log2 N , 2000 .

[13]  Chein-Wei Jen,et al.  A simple processor core design for DCT/IDCT , 2000, IEEE Trans. Circuits Syst. Video Technol..

[14]  S.-F. Hsiao,et al.  New matrix formulation for two-dimensional DCT/IDCT computation and its distributed-memory VLSI implementation , 2002 .

[15]  Zhongfeng Wang,et al.  An improved scaled DCT architecture , 2009, IEEE Transactions on Consumer Electronics.

[16]  A. J. Al-Khalili,et al.  Low-power data-dependent 8/spl times/8 DCT/IDCT for video compression , 2003 .

[17]  Shanq-Jang Ruan,et al.  Low-power and high-quality Cordic-based Loeffler DCT for signal processing , 2007, IET Circuits Devices Syst..

[18]  Jinsang Kim,et al.  Low-power multiplierless DCT architecture using image correlation , 2004, IEEE Trans. Consumer Electron..