Data encoding and decoding using Slepian-Wolf coded nested quantization to achieve Wyner-Ziv coding
A system and method for realizing a Wyner-Ziv encoder may involve the following steps: (a) applying nested quantization to input data from an information source in order to generate intermediate data; and (b) encoding the intermediate data using an asymmetric Slepian-Wolf encoder in order to generate compressed output data representing the input data. Similarly, a Wyner-Ziv decoder may be realized by applying an asymmetric Slepian-Wolf decoder to compressed input data (representing samples of a first source) to obtain intermediate values, and then, jointly decoding the intermediate values using side information (samples of a second source having known correlation with respect to the first source).
1. A method for realizing a Wyner-Ziv encoder, the method comprising:
applying a 2-dimensional nested lattice quantization to input data from an information source in order to generate intermediate data, wherein cell volume of a coarse lattice of the 2-dimensional nested lattice quantization is optimized to minimize a rate distortion; and
encoding the intermediate data using an asymmetric Slepian-Wolf encoder in order to generate compressed output data representing the input data.
2. The method of claim 1 , wherein said applying the 2-dimensional nested lattice quantization to the input data includes:
quantizing values of the input data with respect to a fine lattice to determine corresponding points of the fine lattice; and
computing values of the intermediate data based on the fine lattice points, respectively, wherein the values of the intermediate data designate cosets of the fine lattice modulo the coarse lattice, wherein the coarse lattice Λ2 is a sublattice of the fine lattice Λ1.
3. The method of claim 1 further comprising:
sending the compressed output data to a decode unit.
4. The method of claim 1 , wherein the information source is a source of audio information source, a source of video information, a source of image information, or a source of text information.
5. The method of claim 1 , wherein the information source is a continuous source.
6. The method of claim 1 , wherein the information source is a discrete source.
7. A method for realizing a Wyner-Ziv decoder, the method comprising:
(a) receiving compressed input data representing a block of samples of a first source X;
(b) receiving a block of samples of a second source Y;
(c) applying an asymmetric Slepian-Wolf decoder to the compressed input data, using the block of samples of the second source Y, in order to generate a block of intermediate values;
(d) performing joint decoding on each intermediate value and a corresponding sample of the second source block to obtain a corresponding decompressed output value, wherein said performing joint decoding includes computing a centroid of a conditional probability density of the first source X given said corresponding sample of the second source block restricted to a region of space corresponding to the intermediate value, wherein the centroid determines the decompressed output value.
8. The method of claim 7 , wherein the intermediate values specify cosets of a fine lattice with respect to a coarse lattice, wherein the coarse lattice is a sublattice of the fine lattice.
9. The method of claim 7 , wherein (a) and (b) are performed in parallel.