Method of compression of binary data with a random number generator

The method of claim 1 further comprising

• a) assigning predetermined upper and lower levels for selection of said least distances;
• b) retrieving a sequence of samples from said input numerical sequence corresponding to said least distances are in a range of said upper and bottom levels;
• c) generating a sequence of indexes having a value of positions of said selected samples in said input numerical sequence;
• d) substituting output of operation step a) of claim 1 with said sequence of said selected samples claim 11, step b);
• e) assigning a unique base value and a range for said index of said consequent dictionary sequence;
• f) repeating steps from b) through g) of claim 1;
• g) assigning a dictionary number to of said consequent dictionary sequence;
• h) accumulating a table of correspondence between said dictionary number said base value and said range for said index of said dictionary sequence in said protocol;
• i) comparing said least distances produced by said consequent dictionaries for corresponding samples of said input numerical sequence;
• j) selecting a smallest distance from a plurality of said least distances for corresponding samples of said input numerical sequence;
• whereby providing a losseless consequent refinement of the results of an approximation of said input numerical sequence by said multiple random dictionaries.

An apparatus for binary data compression of a sequence of data values for transmission or storage, using a dictionary sequence comprising:

• a) computer readable program code means for causing a computer to create a sequence of numerical equivalents of said input data with a predetermined peak amplitude and a predetermined number of said numerical equivalents;
• b) computer readable program code means for causing a computer to calculate a frequency distribution of said sequence of numerical equivalents of said input data;
• c) computer readable program code means for causing a computer to generate a dictionary sequence utilizing a random number generator having said frequency distribution similar for said dictionary sequence and said sequence of numerical equivalents of said input data;
• d) computer readable program code means for causing a computer to search for a least distance from a plurality of distances between every said numerical equivalent of said input data and said sequence of dictionary samples;
• e) computer readable program code means for causing a computer to search for an index of a best dictionary sample, corresponding to every said least distance;
• f) computer readable program code means for causing a computer outputting said sequence of least distances, sequence of said indexes of said best dictionary samples;
• thereby providing transforming the random binary data to the sequence, comprising the sequence of the small samples with many repetitions;
• g) computer readable program code means for causing a computer outputting a protocol with a sequence of predetermined parameters and a sequence of results of calculations, comprising the length of said sequence of numerical equivalents of said input data, said peak amplitude, a parameters of said frequency distribution for the dictionary;
• thereby providing losseless decompression of said input data;
• h) computer readable program code means for causing a computer performing compression for said substituted sequence and said protocol, selecting a method of said compression from a list, including:
• Lempel-Ziv type dictionary compression, statistical coding, decorrelation, memory cells packing, according to one or more criteria selected based on said particular application, for said substituted sequence and said protocol;
• whereby providing losseless compression of the random input binary sequence with said calculated random dictionary, approximately matched said input sequence.