Granted Patent B2

US 9,224,398 · App. 13/807,728 · Granted Dec 29, 2015

Abstract

Claims (107)

1. Apparatus comprising at least one processor and at least one memory including computer code, the at least one memory and the computer code configured to with the at least one processor cause the apparatus to at least perform:

transforming, via a first domain transformation module implemented in the at least one memory, an audio signal into a sparse domain signal, the sparse domain signal representing the audio signal;

transforming, via a second domain transformation module implemented in the at least one memory, the sparse domain signal into a measurement domain signal;

determining, via a sampling pattern generation module implemented in the at least one memory, a sampling pattern dependent on the measurement domain signal, wherein determining the sampling pattern comprises:

determining, via a main lobe generation module implemented in the at least one memory, a main lobe sampling pattern;

determining, via a side lobe generation module implemented in the at least one memory, a side lobe sampling pattern;

determining, via a vertical sampling generation module implemented in the at least one memory, a vertical sampling bar sampling pattern; and

determining, via a star shaped sampling generation module implemented in the at least one memory, a star sampling pattern; and

measuring, via a sparse measurement determination module implemented in the at least one memory, the measurement domain signal dependent on the sampling pattern.

2. The apparatus as claimed in claim 1 , wherein transforming the audio signal into a sparse domain signal further causes the apparatus to perform at least one of: modified discrete cosine transforming the audio signal; and

discrete wavelet transforming the audio signal.

3. The apparatus as claimed in claim 1 , wherein transforming the sparse domain signal into a measurement domain signal further causes the apparatus to perform:

a first transformation; and

a further transformation comprising fourier transforming the first transformation, wherein the first transformation causes the apparatus to perform at least one of:

discrete cosine transforming the sparse domain signal;

discrete sine transforming the sparse domain signal; and

Karhunen-Loeve transforming the sparse domain signal.

4. The apparatus as claimed in claim 3 , wherein the audio signal comprises at least two channel audio signals, the sparse domain signal comprising at least two channel sparse domain signals each representing an associated channel audio signal, wherein transforming the first sparse domain signal into a first measurement domain signal further causes the apparatus to perform:

aligning each channel sparse domain signal to form a three dimensional sparse domain signal;

and three dimensional transforming of the three dimensional sparse domain signal.

5. The apparatus as claimed in claim 3 , wherein fourier transforming on the first transformation causes the apparatus to perform two dimensional transforming on each channel of the first transformation.

6. The apparatus as claimed in claim 1 , wherein determining a sampling pattern dependent on the measurement domain signal causes the apparatus to perform:

combining the main lobe sampling pattern, the side lobe sampling pattern, the vertical sampling bar sampling pattern, and the star sampling pattern.

7. The apparatus as claimed in claim 6 , wherein determining the main lobe sampling pattern causes the apparatus to perform determining a main lobe width, wherein the main lobe width is the width of one dimension of the measurement domain signal within which an average of signal energy envelope for the width is greater than a predefined value.

8. The apparatus as claimed in claim 7 , wherein determining the vertical sampling bar sampling pattern causes the apparatus to perform determining a first bar sampling first dimension value adjacent to the main lobe width value; and

determining further bar sampling first dimension values by incrementing the first bar sampling first dimension value by a predefined function value.

9. The apparatus as claimed in claim 6 , wherein determining the side lobe sampling pattern causes the apparatus to perform determining the width of one dimension of the measurement domain signal within which the signal energy envelope is greater than a second predefined value for at least one further dimension value of the measurement domain signal.

10. The apparatus as claimed in claim 6 , wherein determining the star sampling pattern causes the apparatus to perform:

determining a plurality of radial sampling patterns; and

extracting a sub-part of the radial sampling patterns.

11. An apparatus comprising at least one processor and at least one memory including computer code, the at least one memory and the computer code configured to with the at least one processor cause the apparatus to at least perform:

determining, via a sparse measurement determination module implemented in the at least one memory, a measurement domain signal dependent on a sampling pattern and a measured measurement domain signal, wherein determining the measurement domain signal comprises:

determining, via a main lobe generation module implemented in the at least one memory, a main lobe sampling pattern;

determining, via a side lobe generation module implemented in the at least one memory, a side lobe sampling pattern;

determining, via a vertical sampling generation module implemented in the at least one memory, a vertical sampling bar sampling pattern; and

determining, via a star shaped sampling generation module implemented in the at least one memory, a star sampling pattern;

transforming, via a reconstruction module implemented in the at least one memory, the measurement domain signal into a sparse domain signal; and

transforming, via the reconstruction module implemented in the at least one memory, the sparse domain signal into an audio signal.

12. The apparatus as claimed in claim 11 , wherein transforming the sparse domain signal into an audio signal further causes the apparatus to perform at least one of:

inverse modified discrete cosine transforming the audio signal; and inverse discrete wavelet transforming the audio signal.

13. The apparatus as claimed in claim 11 , wherein transforming the measurement domain signal into the sparse domain signal further causes the apparatus to perform:

a first transformation causing the apparatus to perform fourier transforming on the measurement domain signal; and

a further transformation causing the apparatus to perform at least one of;

inverse discrete cosine transforming the first transformation;

inverse discrete sine transforming the first transformation; and

inverse Karhunen-Loeve transforming the first transformation.

14. The apparatus as claimed in claim 13 , wherein fourier transforming on the first transformation causes the apparatus to perform two dimensional transforming on each channel of the measurement domain signal.

15. The apparatus as claimed in claim 11 , further caused to perform:

combining the main lobe sampling pattern, the side lobe sampling pattern, the vertical sampling bar sampling pattern, and the star sampling pattern to form the sampling pattern.

16. The apparatus as claimed in claim 15 wherein determining the vertical sampling bar sampling pattern causes the apparatus to perform determining a first bar sampling first dimension value adjacent to a main lobe sampling pattern width value; and determining further bar sampling first dimension values by incrementing the first bar sampling first dimension value by a predefined function value.

17. The apparatus as claimed in claim 15 , wherein determining the star sampling pattern causes the apparatus to perform:

determining a plurality of radial sampling patterns; and

extracting a sub-part of the radial sampling patterns.

18. A method comprising:

transforming an audio signal into a sparse domain signal, the sparse domain signal representing the audio signal;

transforming the sparse domain signal into a measurement domain signal;

determining, via a sampling pattern generation module implemented in the at least one memory, a sampling pattern dependent on the measurement domain signal, wherein determining the sampling pattern comprises:

determining, via a main lobe generation module implemented in the at least one memory, a main lobe sampling pattern;

determining, via a side lobe generation module implemented in the at least one memory, a side lobe sampling pattern;

determining, via a vertical sampling generation module implemented in the at least one memory, a vertical sampling bar sampling pattern; and

determining, via a star shaped sampling generation module implemented in the at least one memory, a star sampling pattern; and

measuring, via a sparse measurement determination module implemented in the at least one memory, the measurement domain signal dependent on the sampling pattern.

19. The method as claimed in claim 18 , wherein transforming the audio signal into a sparse domain signal comprises at least one of:

modified discrete cosine transforming the audio signal; and

discrete wavelet transforming the audio signal.

20. The method as claimed in claim 18 , wherein transforming the sparse domain signal into a measurement domain signal comprises:

a first transformation; and

a further transformation comprising fourier transforming the first transformation, wherein first transformation comprises at least one of:

discrete cosine transforming the sparse domain signal;

discrete sine transforming the sparse domain signal; and

Karhunen-Loeve transforming the sparse domain signal.

21. The method as claimed in claim 20 , wherein the audio signal comprises at least two channel audio signals, the sparse domain signal comprising at least two channel sparse domain signals each representing an associated channel audio signal, wherein transforming the first sparse domain signal into a first measurement domain signal comprises:

aligning each channel sparse domain signal to form a three dimensional sparse domain signal; and

three dimensional transforming of the three dimensional sparse domain signal.

22. The method as claimed in claim 20 , wherein fourier transforming on the first transformation comprises two dimensional transforming on each channel of the first transformation.

23. The method as claimed in claim 18 , wherein determining a sampling pattern dependent on the measurement domain signal comprises:

combining the main lobe sampling pattern, the side lobe sampling pattern, the vertical sampling bar sampling pattern, and the star sampling pattern.

24. The method as claimed in claim 23 , wherein determining the main lobe sampling pattern comprises determining a main lobe width, wherein the main lobe width is the width of one dimension of the measurement domain signal within which an average of signal energy envelope for the width is greater than a predefined value.

25. The method as claimed in claim 24 , wherein determining the vertical sampling bar sampling pattern comprises determining a first bar sampling first dimension value adjacent to the main lobe width value; and determining further bar sampling first dimension values by incrementing the first bar sampling first dimension value by a predefined function value.

26. The method as claimed in claim 23 , wherein determining the side lobe sampling pattern comprises determining the width of one dimension of the measurement domain signal within which the signal energy envelope is greater than a second predefined value for at least one further dimension value of the measurement domain signal.

27. The method as claimed in claim 23 , wherein determining the star sampling pattern comprises:

determining a plurality of radial sampling patterns; and

extracting a sub-part of the radial sampling patterns.

28. A method comprising:

determining, via a sparse measurement determination module implemented in the at least one memory, a measurement domain signal dependent on a sampling pattern and a measured measurement domain signal, wherein determining the measurement domain signal comprises:

determining, via a main lobe generation module implemented in the at least one memory, a main lobe sampling pattern;

determining, via a side lobe generation module implemented in the at least one memory, a side lobe sampling pattern;

determining, via a vertical sampling generation module implemented in the at least one memory, a vertical sampling bar sampling pattern; and

determining, via a star shaped sampling generation module implemented in the at least one memory, a star sampling pattern;

transforming, via a reconstruction module implemented in the at least one memory, the measurement domain signal into a sparse domain signal; and

transforming, via the reconstruction module implemented in the at least one memory, the sparse domain signal into an audio signal.

29. The method as claimed in claim 28 , wherein transforming the sparse domain signal into an audio signal comprises at least one of:

inverse modified discrete cosine transforming the audio signal; and

inverse discrete wavelet transforming the audio signal.

30. The method as claimed in claim 28 , wherein transforming the measurement domain signal into the sparse domain signal comprises:

a first transformation comprising fourier transforming on the measurement domain signal; and

a further transformation comprising at least one of:

inverse discrete cosine transforming the first transformation;

inverse discrete sine transforming the first transformation; and

inverse Karhunen-Loeve transforming the first transformation.

31. The method as claimed in claim 30 , wherein fourier transforming on the first transformation comprises two dimensional transforming on each channel of the measurement domain signal.

32. The method as claimed in claim 28 , further comprising:

combining the main lobe sampling pattern, the side lobe sampling pattern, the vertical sampling bar sampling pattern, and the star sampling pattern to form the sampling pattern.

33. The method as claimed in claim 32 wherein determining the vertical sampling bar sampling pattern comprises determining a first bar sampling first dimension value adjacent to a main lobe sampling pattern width value; and determining further bar sampling first dimension values by incrementing the first bar sampling first dimension value by a predefined function value.

34. The method as claimed in claim 32 , wherein determining the star sampling pattern comprises:

determining a plurality of radial sampling patterns; and

extracting a sub-part of the radial sampling patterns.

Assignments (8)