A method of decoding of syntactic elements of a data stream is disclosed where, before beginning of decoding of the data stream, cells of all context models are initialized with predefined values, so that each context model contains in each cell data on a probability and a counter of a context occurrence number. A number of cells stored in each context model is selected to be not less than a number of all possible states of context elements associated with a respective context model. The process of decoding of at least a portion of bits of the data stream includes, among other steps: selecting a group of context models; calculating values of at least two context elements; extracting data on the probability from the selected cell of the selected context module; updating data in the selected cell; and carrying out a procedure of probability inheritance.

A method for encoding and decoding a javascript object notation (JSON) document utilizing a statistical tree representing a JSON Schema. The encoded statistical tree may be optimized.

An information processing device includes: a memory configured to store data concerning a write access; and a processor coupled to the memory, the processor being configured to: record, for each data, time of a write access of the data to management information, when writing out data from the memory to a storage, determine a plurality of data as a group of overall compression based on the management information, the plurality of data having a difference of time of write accesses being equal to or less than a threshold value, and compress the plurality of data corresponding to the determined group by an overall compression, and write compressed data obtained through the overall compression to the storage.

A recovery mechanism for robust header compression (ROHC) is disclosed for wireless communication systems. The ROHC recovery mechanism may allow a receiver and/or transmitter in the wireless systems to establish or reestablish a context of a packet transmission session when an initialization and refresh message is lost. In the ROHC recovery mechanism, upon receiving a compressed packet that is not associated with a context, a receiver sends a message to a transmitter suggesting the transmitter to transition to another mode. Upon receiving a subsequent packet transmission that is not associated with a context, the receiver sends another message indicating that a context has not been established or has been lost. The transmitter may then send the receiver necessary information to establish a context for the packet transmission session.

One aspect of the disclosure provides a device, comprising: an allocation module, for determining one or more metrics of each of a plurality of data streams; a compression module, for compressing each of the plurality of data streams and generating a plurality of compressed data streams, the compression module applying a compression ratio that varies as a function of the metrics determined by the allocation module; and a buffer memory, for storing the plurality of compressed data streams.

According to one embodiment, a code table generation device includes a table generation unit, a merge unit and a tree generation unit. The table generation unit generates a frequency table including symbols and frequencies of occurrence respectively associated with the symbols, based on a frequency of occurrence for each symbol of input symbols. The merge unit acquires top K symbols in descending order of the frequencies of occurrence and remaining symbols from the symbols, divides the remaining symbols into one or more symbol sets, and determines a frequency of occurrence associated with a root node of each of subtrees correspond to the respective symbol sets. The tree generation unit generates a Huffman tree using the K symbols and the root node of each of the subtrees.

A method of neural network decoding includes receiving a first syntax element in a model parameter set from a bitstream of a compressed neural network representation (NNR) of a neural network. The first syntax element indicates whether a coding tree unit (CTU) block partitioning is enabled for a tensor in an NNR aggregate unit. The method also includes reconstructing the tensor in the NNR aggregate unit based on the first syntax element.

A method for subdividing an interval during entropy decoding for a bitstream representing a set of video pictures is provided. A sub-interval value is computed by (i) performing an initial right bit-shifting operation on a probability estimator value to reduce a length in bits of the probability estimator value, (ii) multiplying the right bit-shifted probability estimator value by a range value representing the interval, (iii) performing another right bit-shifting operation on a result of the multiplication, and (iv) adding a constant value to a result of the other right bit-shifting operation, wherein the probability estimator value is associated with a probability of a bin having a particular value. The sub-interval value computed based on the right bit-shifted probability estimator value is used to update the interval.

A binary data compression/decompression method is disclosed, where any input binary data string (IFDS) is uniquely and reversibly compressed/decompressed without any data loss by first uniquely formatting and fully describing the IFDS using a set of well defined binary constructs, followed by creating complex structures from custom combinations of said binary constructs that occur within the arbitrary IFDS content, wherein the choice of the said custom combinations depend on the said IFDS content in term of binary constructs therefore creating IFDS content variations and distributions from an expected nominal base wherein said variations and distributions reflect the actual content of the arbitrary IFDS, followed by uniquely processing these variations and distributions in content using several schemes where each scheme brings a unique compression feature, and wherein once this processing completes (i.e. the end of the arbitrary IFDS is reached), it is called that the end of one compression cycle is reached, and wherein another compression cycle can be applied to the data by repeating the cycle steps, and where such compression cycles are repeated until the desired compressed file is reached or until a file floor size limit is reached, floor size below which the disclosed compression has limitations.

A memory device includes a memory; and at least one processor configured to: obtain a symbol stream including a plurality of symbols; determine a Huffman tree corresponding to the symbol stream, wherein each symbol of the plurality of symbols is assigned a corresponding prefix code from among a plurality of prefix codes based on the Huffman tree; generate a prefix length table based on the Huffman tree, wherein the prefix length table indicates a length of the corresponding prefix code for each symbol; generate a logarithm frequency table based on the prefix length table, wherein the logarithm frequency table indicates a logarithm of a frequency count for each symbol, generate a cumulative frequency table which indicates a cumulative frequency count corresponding to each symbol; generate a compressed bitstream by iteratively applying an encoding function to the plurality of symbols based on the logarithm frequency table and the cumulative frequency table; and store the compressed bitstream in the memory.