Lossless coding method for waveform data
AssigneeDolby Laboratories Licensing Corporation
InventorsPeter G Craven, Michael A. Gerzon
Life of this patent
Estimating the term…
Abstract
In a method of lossless processing of an integer value signal in a prediction filter which includes a quantiser, a numerator of the prediction filter is implemented prior to the quantiser and a denominator of the prediction filter is implemented recursively around the quantiser to reduce the peak data rate of an output signal. In the lossless processor, at each sample instant, an input to the quantiser is jointly responsive to a first sample value of a signal input to the prediction filter, a second sample value of a signal input to the prediction filter at a previous sample instant, and an output value of the quantiser at a previous sample incident. In a preferred embodiment, the prediction filter includes noise shaping for affecting the output of the quantiser.
Claims33
A lossless processor for integer-valued signals comprising a prediction filter, wherein the prediction filter includes a quantiser where, at each sample instant, an input to the quantiser is jointly responsive to:
- (a) a first sample value of a signal input to the prediction filter;
- (b) a second sample value of a signal input to the prediction filter at a previous sample instant;
- and
- (c) an output value of the quantiser at a previous sample instant.
A system for transmitting or storage of integer-value signals in one or more channels, said system comprising:
- (a) a lossless encoder providing a prediction encoded signal, wherein said lossless encoder comprises a first lossless processor for integer-valued signals, said first lossless processor comprising a first prediction filter that includes a first quantiser where, at each sample instant, an input to the first quantiser is jointly responsive to:
- (1) a first sample value of a signal input to the first prediction filter;
- (2) a second sample value of a signal input to the first prediction filter at a previous sample instant;
- and
- (3) an output value of the first quantiser at a previous sample instant;
- and
- (b) a lossless decoder that inverts a transformation performed by the lossless encoder, wherein the lossless decoder comprises a second lossless processor for integer-valued signals, said second lossless processor comprising a second prediction filter that includes a second quantiser where, at each sample instant, an input to the second quantiser is jointly responsive to:
- (1) a first sample value of a signal input to the second prediction filter;
- (2) a second sample value of a signal input to the second prediction filter at a previous sample instant;
- and
- (3) an output value of the second quantiser at a previous sample instant.
A method of lossless processing of an integer-valued signal in a prediction filter comprising a quantiser, wherein the prediction filter has a transfer function in which a numerator is implemented prior to the quantiser and a denominator is implemented recursively around the quantiser.
A signal processor in a decoder for lossless gain control comprising an input terminal, an output terminal and a signal processing path coupled between the input terminal and the output terminal, wherein the signal processing path comprises:
- an autodither generator having an output;
- a linear combining circuit having an output, having a first input coupled to the input terminal, and having a second input coupled to the output of the autodither generator, wherein the output of the linear combining circuit provides a signal representing a linear combination of signals received at the first and second inputs of the linear combining circuit;
- and
- a quantiser having an input coupled to an output of the multiplier, and having an output coupled to the output terminal.
An encoding system in which a lossless encoder is preceded by a rounding quantiser that quantises an input signal to a reduced quantising precision such that a first data rate resulting from lossless encoding of the reduced quantising-precision signal is less than a second data rate that would result from lossless encoding of the input signal.
Dependencies are inferred from the claim text.