Abstract: The present invention comprises a distributed data processing system including a plurality of data processing elements for expeditiously performing an encoding or prediction function pursuant to a context-based model in an adaptive, optimal and time-progressive manner. The distributed data processing system, having access to each symbol of an input data string at each clock cycle, adaptively generates context-relevant data sets which provide the best model for coding or prediction based on the input symbols. Each symbol and its best model for encoding or prediction emerge concurrently from the system, resulting in a favorable time complexity of O(n) for an n-symbol input data string.

Abstract: A method and apparatus are provided for adaptively and predictively determining probabilities of occurrence for each symbol of a finite alphabet within a symbol string. A plurality of intervals are defined over a portion of the symbol string. As successive new symbols are added to the string, they enter the intervals, and old symbols pass out of the intervals. A probability for each symbol of the alphabet is maintained and updated by the following process. For each new symbol which enters the intervals, it is determined whether the new symbol is a given character of the alphabet, and whether each old symbol leaving each interval is the given character. Accordingly, the number of occurrences of the given character within each interval may change. A probability update value is determined, having a component from each interval determined by whether the number of occurrences of the given character in that interval changed.

Abstract: A system and method for maximizing data compression by optimizing model selection during coding of an input stream of data symbols, wherein at least two models are run and compared, and the model with the best coding performance for a given-size segment or block of compressed data is selected such that only its block is used in an output data stream. The best performance is determined by 1) respectively producing comparable-size blocks of compressed data from the input stream with the use of the two, or more, models and 2) selecting the model which compresses the most input data. In the preferred embodiment, respective strings of data are produced with each model from the symbol data and are coded with an adaptive arithmetic coder into the compressed data. Each block of compressed data is started by coding the decision to use the model currently being run and all models start with the arithmetic coder parameters established at the end of the preceding block.

Abstract: Apparatus and method for compressing and de-compressing binary decision data by arithmetic coding and decoding wherein the estimated probability Qe of the less probable of the two decision events, or outcomes, adapts as decisions are successively encoded. To facilitate coding computations, an augend value A for the current number line interval is held to approximate one by renormalizing A whenever it becomes less than a prescribed minimum AMIN. When A is renormalized, the value of Qe is up-dated. The renormalization of A and up-dating of Qe are preferably based on a single-bit test. Also, each Qe value is preferably specified as a 12-bit value having the least significant bit set to 1 and having no more than four other bits set to 1. The number of Qe values in the 1/4 to 1/2 probability range is enhanced to improve coding efficiency.

Abstract: Method and apparatus which cyclically generate a compressed, arithmetically-coded binary stream in response to binary occurrence counts of symbols in an uncoded string. The symbols in the uncoded string are drawn from a multi-character alphabet which is not necessarily a binary one. Coding operations and hardware are simplified by deriving from the binary occurrence counts an estimate of the probability of each unencoded symbol at its precise lexical location. The probability estimation eliminates any requirement for division or multiplication by employing magnitude-shifting of the binary occurrence counts. The encoded stream is augmented by the estimated symbol probability at the same time that an internal variable is updated with an estimate of the portion of a probability interval remaining after coding the current symbol, the interval estimate being obtained from the left-shifted occurrence counts. Decoding is the dual of encoding.

Abstract: A two-stage single pass adaptive modeling method and means for a finite alphabet first order MARKOV symbol source where the model is used to control an encoder on a per symbol basis thereby enabling efficient compression within a fixed preselected implementation complexity.

Abstract: A method and apparatus for recursively generating an arithmetically compressed binary number stream responsive to the binary string from conditional sources. Throughput is increased by reducing the number of operations required to encode each binary symbol so that only a single shift of k bits is required upon receipt of each least-probable symbol or an "add time", followed by a decision and a one-bit shift in response to each most-probable symbol encoding. The concurrent augmentation of the compressed stream and an internal variable involves only the function of a probability interval estimate of the most-probable symbol, and not upon the past encoding state of either variable (2.sup.-k, 49, 63, C, T). Each binary symbol may be recovered by subtracting 2.sup.-k from the q-most-significant bits of the compressed stream and testing the leading bit of the difference.

Abstract: Carry-over control in strings resulting from the high to low order combining of two binary number strings is obtained through the insertion of a control character within the resultant string after detecting a run of consecutive 1's. Upon subsequent accessing and decomposition of the resultant string, the control character causes string decomposition to operate for a number of cycles in a carry correction mode. If the control character indicates that a carry has rippled through the n lesser significant positions of the resultant string, then upon decomposition, those "n" consecutive 1's are changed to 0's, and a 1 is added to the least significant position in the string preceding the control character. If the control character indicates no carry occurrence, then it is merely deleted from the string. The control of carries in this manner permits the generation of arithmetic string compression code sequences in an instantaneous FIFO pattern with only a modest reduction of compression efficiency.

Abstract: A method and means of arithmetic coding of conditional binary sources permitting instantaneous decoding and minimizing the number of encoding operations per iteration. A single shift and subtract operation for each encoding cycle can be achieved if an integer valued parameter representative of a probability interval embracing each source symbol relative frequency is used for string encoding and control. If the symbol being encoded is the most probable, then nothing is added to the arithmetic code string. However, an internal variable is updated by replacing it with an augend amount. If the updated internal variable has a leading zero, then both it and the code string are shifted left by one position. If the symbol being encoded is the least probable, then a computed augend is added to the code string and the code string is shifted by an amount equal to the integer valued parameter.