Bit Serial Architecture for the Two-Dimensional DCT

We present an architecture for the calculation of the Two Dimensional Discrete Cosine Transform and its Inverse that admits a high data rate. It is based on the row-column decomposition, the use of a fast algorithm, serial digit arithmetic and redundant coding. The critical path is set by the delay of a multiplexer plus a binary adder with as many digits as the width of the serial digits to be processed. We discuss its implementation for processing 8 bit 8x8 pixel blocks or 12 bit coefficients. Its implementation using standard cell 1 microm CMOS technology presents a 100MHz data rate and a core area of 42mm .

[1]  Ting Chen,et al.  VLSI implementation of a 16*16 discrete cosine transform , 1989 .

[2]  Francis Jutand,et al.  A one chip VLSI for real time two-dimensional discrete cosine transform , 1988, 1988., IEEE International Symposium on Circuits and Systems.

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

[4]  F. Jutand,et al.  A single chip video rate 16×16 discrete cosine transform , 1986, ICASSP '86. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  U. Totzek,et al.  CMOS VLSI implementation of the 2D-DCT with linear processor arrays , 1990, International Conference on Acoustics, Speech, and Signal Processing.