Fast system and method for computing modulated lapped transforms

Claim 2

A computer-implemented method for processing a signal, said method comprising: receiving a first block of samples of the signal; applying butterfly coefficients to the first block of samples determined by a given window function to produce resulting vectors; storing a portion of the resulting vectors in a memory of a one block delay buffer and recovering current contents of the delay buffer; mapping the input signal with the applied butterfly coefficients to a cascade of butterflies using a first set of weights and reordering the cascade of butterflies; computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients; and combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 3

The method of claim 1, wherein the encoded output is produced as a vector with modulated lapped transform coefficients corresponding to the input block of samples.

Claim 4

The method of claim 2, wherein the modulated lapped transform has a sine window.

Claim 5

The method of claim 1, further comprising processing the encoded output by at least one of transmitting the output, storing the output, compressing the output, enhancing the output, and filtering the output.

Claim 6

The method of claim 4, wherein filtering the output comprises reducing interference within the input signal with a noise reducer.

Claim 7

The method of claim 4, wherein filtering the output comprises reducing interference within the input signal with an echo canceller.

Claim 8

The method of claim 4, wherein compressing the output is achieved by at least one of scalar and vector quantization.

Claim 9

The method of claim 1, wherein the window function is adapted for reducing blocking effects.

Claim 10

The method of claim 1, wherein the spatial transform is performed by a discrete cosine transform operation.

Claim 11

The method of claim 1, wherein the transform coefficients are combined by a coefficient combination operator.

Claim 12

The method of claim 1, wherein the encoded output is produced as a vector with biorthogonal modulated lapped transform coefficients corresponding to the input block of samples.

Claim 13

The method of claim 1, wherein a portion of the resulting vectors produced by applying the butterfly coefficients to the first block of samples is stored in a memory of a one block delay buffer.

Claim 14

A fast computation system stored on computer readable memory of a computer system for computing a spatial transform of an input signal, the fast computation system comprising: a window processor having a window function and an operator comprised of a first set of weights, wherein the window processor is preprogrammed for receiving the input signal as sample blocks and the operator is adapted to apply butterfly coefficients determined by the window function to produce resulting vectors, for mapping the input signal to a cascade of butterflies using the first set of weights, and for reordering the cascade of butterflies; and a transform processor having a transform module for computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients and a coefficient combination operator for combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 15

A fast computation system stored on computer readable memory of a computer system for computing a spatial transform of an input signal, the fast computation system comprising: a window processor having a window function and an operator comprised of a first set of weights, wherein the window processor is preprogrammed for receiving the input signal as sample blocks and the operator is adapted to apply butterfly coefficients determined by the window function to produce resulting vectors, for mapping the input signal to a cascade of butterflies using the first set of weights, and for reordering the cascade of butterflies; and a transform processor having a transform module for computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients and a coefficient combination operator for combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 16

The fast computation system of claim 13, wherein the encoded output is produced as a vector with modulated lapped transform coefficients corresponding to the input block of samples.

Claim 17

The fast computation system of claim 14, wherein the modulated lapped transform has a sine window.

Claim 18

The fast computation system of claim 13, further comprising an external module for processing the encoded output by at least one of transmitting the output, storing the output, compressing the output, enhancing the output, and filtering the output.

Claim 19

The fast computation system of claim 13, wherein the window operator is adapted for reducing blocking effects.

Claim 20

The fast computation system of claim 13, wherein the transform module is a discrete cosine transform operator.

Claim 21

The fast computation system of claim 13, wherein the encoded output is produced as a vector with biorthogonal modulated lapped transform coefficients corresponding to the input block of samples.

Claim 22

The fast computation system of claim 13, wherein the window processor further comprises a memory of a one block delay buffer for storing a portion of the resulting vectors in the memory of a one block delay buffer and for recovering current contents of the delay buffer.

Claim 23

A computer-readable medium having computer-executable components for causing a computer to function as a fast computation system for computing a spatial transform of an input signal, the system comprising: a computer-readable storage medium; and a computer computation system stored on the medium, the computer computation system having a window processor and a transform processor stored in the computer-readable storage medium; wherein the computer computation system is programmed so that the window processor receives the input signal as sample blocks by, applies butterfly coefficients determined by the window function to produce resulting vectors, maps the input signal to a cascade of butterflies with a first set of weights, and reorders the cascade of butterflies; and wherein the computer computation system is further preprogrammed so that the transform processor computes a spatial transform from the reordered cascade of butterflies to produce transform coefficients and combines the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 24

A computer-readable medium having computer-executable components for causing a computer to function as a fast computation system for computing a spatial transform of an input signal, the system comprising: a computer-readable storage medium; and a computer computation system stored on the medium, the computer computation system having a window processor and a transform processor stored in the computer-readable storage medium; wherein the computer computation system is programmed so that the window processor receives the input signal as sample blocks by, applies butterfly coefficients determined by the window function to produce resulting vectors, maps the input signal to a cascade of butterflies with a first set of weights, and reorders the cascade of butterflies; and wherein the computer computation system is further preprogrammed so that the transform processor computes a spatial transform from the reordered cascade of butterflies to produce transform coefficients and combines the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 25

The fast computation system of claim 21, wherein the encoded output is produced as a vector with modulated lapped transform coefficients corresponding to the input block of samples.

Claim 26

The fast computation system of claim 22, wherein the modulated lapped transform has a sine window.

Claim 27

The fast computation system of claim 21, further comprising an external module for processing the encoded output by at least one of transmitting the output, storing the output, compressing the output, enhancing the output, and filtering the output.

Claim 28

The fast computation system of claim 21, wherein the window operator is adapted for reducing blocking effects.

Claim 29

The fast computation system of claim 21, wherein the transform module is a discrete cosine transform operator.

Claim 30

The fast computation system of claim 21, wherein the encoded output is produced as a vector with biorthogonal modulated lapped transform coefficients corresponding to the input block of samples.

Claim 31

The fast computation system of claim 21, wherein the window processor further comprises a memory of a one block delay buffer for storing a portion of the resulting vectors in the memory of a one block delay buffer and for recovering current contents of the delay buffer.

Claim 32

A computer-readable medium having computer-executable instructions for performing a method for computing a spatial transform of an input signal, the method comprising: (a) applying butterfly coefficients to a first block of samples of the input signal; (b) mapping the input signal with the applied butterfly coefficients to a cascade of butterflies using a first set of weights and reordering the cascade of butterflies; (c) computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients; and (d) combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 33

A computer-readable medium having computer-executable instructions for performing a method for computing a spatial transform of an input signal, the method comprising: (a) applying butterfly coefficients to a first block of samples of the input signal; (b) mapping the input signal with the applied butterfly coefficients to a cascade of butterflies using a first set of weights and reordering the cascade of butterflies; (c) computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients; and (d) combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 34

The method of claim 29, wherein the encoded output is produced as a vector with modulated lapped transform coefficients corresponding to the block of samples.

Claim 35

The method of claim 30, wherein the modulated lapped transform has a sine window.

Claim 36

The method of claim 29, further comprising processing the encoded output by at least one of transmitting the output, storing the output, compressing the output, enhancing the output, and filtering the output.

Claim 37

The method of claim 32, wherein compressing the output is achieved by at least one of scalar and vector quantization.

Claim 38

The method of claim 29, wherein the window function is adapted for reducing blocking effects.

Claim 39

The method of claim 29, wherein the spatial transform is performed by a discrete cosine transform operation.

Claim 40

The method of claim 29, wherein the transform coefficients are combined by a coefficient combination operator.

Claim 41

The method of claim 29, wherein the encoded output is produced as a vector with biorthogonal modulated lapped transform coefficients corresponding to the input block of samples.

Claim 42

The method of claim 29, wherein, after (a) and before (b), further comprising producing resulting vectors by applying the butterfly coefficients to the block of samples, storing a portion of the resulting vectors in a memory of a one block delay buffer and recovering current contents of the delay buffer.

Claim 43

A computer-readable memory device encoded with a set of software instructions for performing a method for computing a spatial transform of an input signal, the set of software instructions having software routines, each routine containing: a first parameter value for receiving the input signal as sample blocks and applying butterfly coefficients determined by the first parameter value to the sample blocks to produce resulting vectors, for mapping the input signal with the applied butterfly coefficients to a cascade of butterflies using a first set of weights, and for reordering the cascade of butterflies; and a second parameter for computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients and for combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 44

A computer-readable memory device encoded with a set of software instructions for performing a method for computing a spatial transform of an input signal, the set of software instructions having software routines, each routine containing: a first parameter value for receiving the input signal as sample blocks and applying butterfly coefficients determined by the first parameter value to the sample blocks to produce resulting vectors, for mapping the input signal with the applied butterfly coefficients to a cascade of butterflies using a first set of weights, and for reordering the cascade of butterflies; and a second parameter for computing a spatial transform from the reordered cascade of butterflies to produce transform coefficients and for combining the transform coefficients to produce an encoded output corresponding to the input signal.

Claim 45

The method of claim 41, wherein the modulated lapped transform has a sine window.

Claim 46

The method of claim 39, wherein the encoded output is produced as a vector with modulated lapped transform coefficients corresponding to the input block of samples.

Claim 47

The method of claim 39, further comprising processing the encoded output by at least one of transmitting the output, storing the output, compressing the output, enhancing the output, and filtering the output.

Claim 48

The method of claim 42, wherein compressing the output is achieved by at least one of scalar and vector quantization.

Claim 49

The method of claim 39, wherein the spatial transform is performed by a discrete cosine transform operation.

Claim 50

The method of claim 39, wherein the transform coefficients are combined by a coefficient combination operator.

Claim 51

The method of claim 39, wherein the encoded output is produced as a vector with biorthogonal modulated lapped transform coefficients corresponding to the input block of samples.

Claim 52

The method of claim 39, wherein, after the butterfly coefficients are applied, further comprising storing a portion of the resulting vectors in a memory of a one block delay buffer and recovering current contents of the delay buffer.

Claim 53

A computer-readable medium having computer-executable instructions for performing a method for computing an inverse spatial transform of a preprocessed signal, the method comprising: (a) receiving the preprocessed signal which represents an encoded format of an original input signal, the encoded format containing a spatial transform with transform coefficients; (b) combining the transform coefficients to produce resulting vectors; (c) computing an inverse spatial transform from the resulting vectors to produce inverse transform coefficients; (d) mapping the inverse transform coefficients to a cascade of butterflies using a second set of weights and reordering the cascade of butterflies; and (e) applying butterfly coefficients determined by a given window function to the mapped inverse transform coefficients to produce a decoded output corresponding to the original input signal.

Claim 54

A computer-readable medium having computer-executable instructions for performing a method for computing an inverse spatial transform of a preprocessed signal, the method comprising: (a) receiving the preprocessed signal which represents an encoded format of an original input signal, the encoded format containing a spatial transform with transform coefficients; (b) combining the transform coefficients to produce resulting vectors; (c) computing an inverse spatial transform from the resulting vectors to produce inverse transform coefficients; (d) mapping the inverse transform coefficients to a cascade of butterflies using a second set of weights and reordering the cascade of butterflies; and (e) applying butterfly coefficients determined by a given window function to the mapped inverse transform coefficients to produce a decoded output corresponding to the original input signal.

Claim 55

The method of claim 50, wherein the modulated lapped transform has a sine window.

Claim 56

The method of claim 48, wherein the encoded format is a vector with modulated lapped transform coefficients corresponding to the preprocessed signal.

Claim 57

The method of claim 48, wherein the spatial transform is performed by an inverse discrete cosine transform operation.

Claim 58

The method of claim 48, wherein the transform coefficients are combined by an inverse coefficient combination operator.

Claim 59

The method of claim 48, wherein the encoded format is a vector with biorthogonal modulated lapped transform coefficients corresponding to the preprocessed signal.

Claim 60

The method of claim 48, wherein (d) further comprises storing a portion of the resulting vectors in a memory of a one block delay buffer and recovering current contents of the delay buffer.

Claim 61

A fast inverse computation system stored on computer readable memory of a computer system for computing an inverse spatial transform of a preprocessed signal, wherein the preprocessed signal represents an encoded format of an original input signal, the encoded format containing a spatial transform with transform coefficients, the fast inverse computation system comprising: an inverse transform processor having a coefficient combination operator for combining the transform coefficients to produce resulting vectors and an inverse transform module for computing an inverse spatial transform from the resulting vectors to produce inverse transform coefficients; and an inverse window processor having a window function and an operator comprised of a second set of weights, wherein the window processor receives the inverse transform coefficients and is preprogrammed for mapping the inverse transform coefficients to a cascade of butterflies using the second set of weights and reordering the cascade of butterflies and for applying butterfly coefficients determined by a given window function to the mapped inverse transform coefficients to produce a decoded output corresponding to the original input signal.

Claim 62

A fast inverse computation system stored on computer readable memory of a computer system for computing an inverse spatial transform of a preprocessed signal, wherein the preprocessed signal represents an encoded format of an original input signal, the encoded format containing a spatial transform with transform coefficients, the fast inverse computation system comprising: an inverse transform processor having a coefficient combination operator for combining the transform coefficients to produce resulting vectors and an inverse transform module for computing an inverse spatial transform from the resulting vectors to produce inverse transform coefficients; and an inverse window processor having a window function and an operator comprised of a second set of weights, wherein the window processor receives the inverse transform coefficients and is preprogrammed for mapping the inverse transform coefficients to a cascade of butterflies using the second set of weights and reordering the cascade of butterflies and for applying butterfly coefficients determined by a given window function to the mapped inverse transform coefficients to produce a decoded output corresponding to the original input signal.

Claim 63

The fast computation system of claim 55, wherein the encoded format is a vector with modulated lapped transform coefficients corresponding to the preprocessed signal.

Claim 64

The fast computation system of claim 56, wherein the modulated lapped transform has a sine window.

Claim 65

The fast computation system of claim 55, wherein the inverse transform module is an inverse discrete cosine transform operator.

Claim 66

The fast computation system of claim 55, wherein the encoded format is a vector with biorthogonal modulated lapped transform coefficients corresponding to the preprocessed signal.

Claim 67

The fast computation system of claim 55, wherein the inverse window processor further comprises a memory of a one block delay buffer for storing a portion of the resulting vectors in the memory of a one block delay buffer and for recovering current contents of the delay buffer.