System and method for a dyadic-monotonic (DM) codec

1. A coder comprising a variable modulus, wherein the coder compresses data by mapping an input sequence of symbols into a coded sequence of the symbols, and the variable modulus is utilized during the data compression, where each symbol of the input sequence is drawn from an associated finite alphabet of input symbols and each symbol of the coded sequence is drawn from an associated finite alphabet of code symbols.

2. The coder as recited in claim 1 , wherein the modulus reflects a steepness of a probability distribution curve associated with a compression algorithm.

3. The coder as recited in claim 2 , wherein the modulus includes a negative exponential of the probability distribution.

4. The coder as recited in claim 2 , wherein the probability distribution is associated with a codec.

5. The coder as recited in claim 2 , wherein the codec is designed to utilize a minimal computational complexity given a predetermined performance level.

6. The coder as recited in claim 1 , wherein the modulus depends on a context of a previous set of data.

7. The coder as recited in claim 1 , wherein the modulus avoids increasing as a function of a run length.

8. The coder as recited in claim 7 , wherein the run length includes a plurality of identical bits in a sequence.

9. The coder as recited in claim 1 , wherein the coder includes an entropy coder.

10. A decoder comprising a variable modulus, wherein the decoder is a component of a codec that compresses data by mapping an input sequence of symbols into a coded sequence of the symbols, and the variable modulus is utilized during the data compression, where each symbol of the input sequence is drawn from an associated finite alphabet of input symbols and each symbol of the coded sequence is drawn from an associated finite alphabet of code symbols.

11. The decoder as recited in claim 10 , wherein the modulus reflects a steepness of a probability distribution curve associated with a compression algorithm.

12. The decoder as recited in claim 11 , wherein the modulus includes a negative exponential of the probability distribution.

13. The decoder as recited in claim 11 , wherein the probability distribution is associated with a codec.

14. The decoder as recited in claim 11 , wherein the codec is designed to utilize a minimal computational complexity given a predetermined performance level.

15. The decoder as recited in claim 10 , wherein the modulus depends on a context of a previous set of data.

16. The decoder as recited in claim 10 , wherein the modulus avoids increasing as a function of a run length.

17. The decoder as recited in claim 16 , wherein the run length includes a plurality of identical bits in a sequence.

18. The decoder as recited in claim 10 , wherein the decoder includes an entropy decoder.

19. A method for using a codec including a variable modulus that reflects a steepness of a probability distribution curve associated with a compression algorithm and does not increase with a run length, wherein the codec compresses data by mapping an input sequence of symbols into a coded sequence of the symbols, and the variable modulus is utilized during the data compression, where each symbol of the input sequence is drawn from an associated finite alphabet of input symbols and each symbol of the coded sequence is drawn from an associated finite alphabet of code symbols.

20. The method as recited in claim 19 , wherein the variable modulus includes -log2(delta), where delta is a negative integral power of two.

21. The method as recited in claim 19 , wherein the codec utilizes:

a) a dyadic constraint as follows:

delta(context)=2 −m 0<m≦n,

where m and n are integers, and delta includes a negative integral power of two, and

b) a monotonic constraint as follows:

(1-delta(context))≦(1-delta((2*context mod 2 k )+mps(context))),

where (1-delta((2*context mod 2 k )+mps(context))) is approximately a probability of reading a most probable symbol given that a previously read symbol was also most probable.