Lossless embedded additional data
OwnerPeter Graham Craven, Malcolm Lawinventors
Life of this patent
Estimating the term…
Abstract
Methods are disclosed for an encoder to embed a data stream into a quantized PCM digital audio signal and for a corresponding decoder to both retrieve the data stream and losslessly reconstruct the exact original audio. Some methods employ complimentary amplification and attenuation, while others employ gain redistribution. Pre-emphasis and soft clipping techniques are described as methods of losslessly reducing the peak excursion of the PCM audio signal. Also described is the lossless placing of data at predetermined positions within an audio stream.
Claims25
A method of losslessly burying binary data into a pulse code modulated “PCM” digital audio signal, the method comprising:
- receiving a PCM digital audio signal comprising samples having values that are quantised and define a range of values;
- receiving binary data;
- establishing a decoder gain g for a decoder, where 0.5<g, <1.0;
- taking a quantised value of a sample and choosing a replacement sample value from a set of quantised values that, when mapped by the decoder gain g and quantised, yields a value equal to said quantised value of the sample;
- and
- replacing the sample with a replacement sample having a value equal to the replacement sample value,
- wherein, conditionally on the set of quantised values containing more than one value, the step of choosing the replacement sample is performed in dependence on the binary data, and
- wherein the method provides a modified PCM signal from which a bit-exact replica of the PCM digital audio signal may be reconstructed.
A method of losslessly burying binary data into a set of signal samples representing a portion of a digital audio signal, the method comprising:
- dividing the set of signal samples into a first nonempty subset of signal samples and a second nonempty subset of signal samples;
- losslessly burying the binary data into at least some signal samples in the first nonempty subset;
- and
- decreasing a gain of signal samples in the first nonempty subset and increasing a gain of signal samples in the second nonempty subset so as to equalise the respective gains across both subsets by applying a lossless matrix transformation to the samples in each subset,
- wherein the method provides a modified PCM signal from which a bit-exact replica of the digital audio signal may be reconstructed.
A method of retrieving binary data from a set of signal samples representing a portion of a digital audio signal, the method comprising:
- dividing the set of signal samples into a first nonempty subset of signal samples and a second nonempty subset of signal samples;
- and,
- retrieving the binary data from signal samples in the first nonempty subset;
- and
- applying a lossless matrix transformation to the samples in the set,
- wherein the step of retrieving is performed using a lossless buried data method,
- wherein the lossless matrix transformation increases the gain of signal samples in the first nonempty subset and decreases the gain of signal samples in the second nonempty subset so as to equalise the respective gains across both subsets, and
- wherein the lossless matrix transformation enables the portion of the digital audio signal to be reconstructed such that the reconstructed audio data is bit-exact.
A method of losslessly placing binary data into a predetermined set of bit positions within a stream of audio data bits representing a PCM digital audio signal, the method comprising:
- retrieving signal bits from the predetermined set of bit positions;
- losslessly burying the retrieved signal bits into the stream of audio data bits representing the PCM digital audio signal;
- and
- placing the binary data into bit positions within the predetermined set of bit positions,
- wherein the losslessly burying of the retrieved signal bits enables the PCM digital audio signal to be reconstructed such that the reconstructed audio data is bit-exact.
A method of losslessly retrieving binary data from a modified stream of audio data bits representing an original PCM digital audio signal, the method comprising:
- establishing a set of bit positions within the modified stream that contain binary data;
- retrieving binary data bits from the set of bit positions;
- retrieving signal bits from the PCM digital audio signal modified stream using a method of lossless buried data;
- and
- placing the signal bits into the set of bit positions,
- wherein the method provides a bit-exact replica of the original PCM digital audio signal.
A method of losslessly burying binary data into a stream of audio data bits representing a PCM digital audio signal, the method comprising:
- losslessly pre-emphasising the PCM digital audio signal thus reducing an amplitude of lower frequency components and boosting an amplitude of higher frequency components;
- and
- losslessly burying the binary data into the PCM digital audio signal,
- wherein the step of losslessly burying the binary data into the PCM digital audio signal comprises applying a gain greater than 1.0 and less than 2.0,
- the low frequency gain of the pre-emphasis cancels the gain of the burying step, and
- the losslessly burying of the binary data enables the PCM digital audio signal to be reconstructed such that the reconstructed audio data is bit-exact.
An encoder for losslessly burying binary data stored in memory into a stream of audio data bits representing a PCM digital audio signal, wherein the encoder is adapted to:
- losslessly pre-emphasise, by a processing device of the encoder, the PCM digital audio signal thus reducing an amplitude of lower frequency components and boosting an amplitude of higher frequency components;
- and
- losslessly bury, by the processing device, the binary data stored in memory into the PCM digital audio signal,
- wherein the step of losslessly burying the binary data stored in memory into the PCM digital audio signal comprises applying a gain greater than 1.0 and less than 2.0,
- the low frequency gain of the pre-emphasis cancels the gain of the burying step, and
- the losslessly burying of the binary data stored in memory enables the PCM digital audio signal to be reconstructed such that the reconstructed audio data is bit-exact.
A decoder for losslessly retrieving a PCM digital audio signal from a set of signal samples stored in memory representing the PCM digital audio signal into which binary data has been losslessly buried, wherein the decoder is adapted to:
- losslessly retrieve, by a processing device of the decoder, the binary data from the set of signal samples stored in memory using a method of lossless buried data comprising applying a gain greater than 0.5 and less than 1.0;
- and
- losslessly de-emphasise, by the processing device, the set of signal samples stored in memory in order to reconstruct an amplitude of frequency components of the PCM digital audio signal that have been altered by lossless preemphasis,
- wherein the losslessly retrieving of the binary data enables the PCM digital audio signal to be reconstructed such that the reconstructed audio data is bit-exact.
A codec comprising:
- an encoder for losslessly burying binary data into a stream of audio data bits representing a PCM digital audio signal;
- and
- a decoder for losslessly retrieving the PCM digital audio signal from a set of signal samples representing the PCM digital audio signal into which binary data has been losslessly buried by the encoder,
- wherein the encoder is adapted to losslessly pre-emphasise the PCM digital audio signal thus reducing an amplitude of lower frequency components and boosting an amplitude of higher frequency components, and losslessly bury the binary data into the PCM digital audio signal, wherein the step of losslessly burying the binary data into the PCM digital audio signal comprises applying a gain greater than 1.0 and less than 2.0, and the low frequency gain of the pre-emphasis cancels the gain of the burying step,
- wherein the decoder is adapted to retrieve the binary data from the set of signal samples using a method of lossless buried data comprising applying a gain greater than 0.5 and less than 1.0, and losslessly de-emphasise the set of signal samples in order to reconstruct an amplitude of frequency components of the PCM digital audio signal that have been altered by lossless pre-emphasis, and
- wherein the losslessly burying of the binary data enables the PCM digital audio signal to be reconstructed such that the reconstructed audio data is bit-exact.
Dependencies are inferred from the claim text.