Digital signal coding method and circuit for calculating scalar products of vectors and corresponding discrete cosinus transform

摘要

For two vectors of the same dimension p, of dedicated components ak and variables xk, the value …IMAGE… of the scalar product is reduced into partial sums fi of binary variables xki, which take one of the values of the fixed components ak according to the values of the xki according to m possibilities. A dedicated logical coding (1) makes it possible, on the basis of the variables xki, to generate for each bit of rank j of fi, a plurality of elementary partial sums, at the bit level, fij according to 2m possibilities by variation of the binary values akj of the bits of rank j. A bidimensional interconnection matrix (2) makes a single value of the elementary partial sums fij, whose set defines the corresponding partial sum fi, correspond to each bit of rank j akj. …??Application to image processing and data compression circuits by DCT transform. …IMAGE…