Apparatus and method for coding signal in a communication system

• US 8,924,203 B2
• Filed: 10/26/2012
• Issued: 12/30/2014
• Est. Priority Date: 10/28/2011
• Status: Expired due to Fees

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Kim, Hyun-Woo
• Primary Examiner(s): AZAD, ABUL K
• Application Number:
  US13/661,945
• Publication Number:
  US 20130117028A1
• Time in Patent Office: 795 Days
• Field of Search: 704219-230
• US Class Current: 704/223
• CPC Class Codes: G10L 19/038 Vector quantisation, e.g. TwinVQ audio
• CPC Class Codes: G10L 19/04 using predictive techniques
• CPC Class Codes: G10L 19/12 the excitation function being a code excitation, e.g. in code excited linear prediction [CELP] vocoders
• CPC Class Codes: G10L 19/24 Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding

Abstract

Disclosed are an apparatus for coding a signal in a communication system including: a coding unit configured to code voice and audio signals based on a code excited linear prediction (CELP) coding method; a residual signal calculation unit configured to calculate residual signals of the voice and audio signals; a frequency transform unit configured to transform the residual signal into a signal in a frequency domain; an energy calculation unit configured to use frequency coefficients of the residual signals to calculate frequency energy of the residual signals; an energy concentration calculation unit configured to calculate energy concentrations of each vector dimension of the residual signals from the frequency energy of the residual signals; and a vector dimension determination unit configured to compare the energy concentrations of each vector dimension to determine targeted vector dimensions of the residual signals.

Claims

1. An apparatus for coding a signal in a communication system, comprising:

a coding unit configured to code voice and audio signals based on a code excited linear prediction (CELP) coding method;

a residual signal calculation unit configured to calculate residual signals of the voice and audio signals;

a frequency transform unit configured to transform the residual signal into a signal in a frequency domain;

an energy calculation unit configured to use frequency coefficients of the residual signals to calculate frequency energy of the residual signals;

an energy concentration calculation unit configured to calculate energy concentrations of each vector dimension of the residual signals from the frequency energy of the residual signals; and

a vector dimension determination unit configured to compare the energy concentrations of each vector dimension to determine targeted vector dimensions of the residual signals.

2. The apparatus of claim 1, wherein the vector dimension determination unit determines the vector dimensions having a maximum value as the targeted vector dimensions in the energy concentrations of each vector dimension.

3. The apparatus of claim 1, wherein the residual signal calculation unit calculates a difference between the voice and audio signals and a signal synthesized by a CELP codec by a CELP coding method.

4. The apparatus of claim 1, wherein the frequency transform unit transforms the residual signals from a domain time into a frequency domain by a modified discrete cosine transform (MDCT) or a discrete Fourier transform (DFT).

5. The apparatus of claim 1, wherein a residual signal weighting unit configured to apply a perceptual weighting filter to frequency coefficients of the residual signals to acquire weighting signals of the residual signals.

6. The apparatus of claim 1, wherein the energy calculation unit uses the frequency coefficients of the residual signals to calculate energy in a sub-band of the residual signals.

7. The apparatus of claim 1, wherein the energy concentration calculation unit arranges the frequency energies of the residual signals in an energy size sequence and calculates the frequency energies of each vector dimension according to the size sequence to calculate the energy concentrations of each vector dimension.

8. The apparatus of claim 1, further comprising:

a position determination unit configured to allocate the frequency coefficients to the targeted vectors of the residual signals as much as the targeted vector dimension in a sequence that absolute values of the frequency coefficients are large to store the position of the targeted vector to which the frequency coefficients are allocated; and

a quantization unit configured to calculate the position of the frequency coefficients allocated to the targeted vector to quantize the position of the targeted vector.

9. The apparatus of claim 8, further comprising:

a gain quantization unit configured to quantize a gain of the targeted vector;

a normalization unit configured to normalize the targeted vector with the gain of the quantized targeted vector;

a shape quantization unit configured to quantize the normalized targeted vector; and

a code quantization unit configured to quantize a position code of the targeted vector.

10. The apparatus of claim 9, wherein the gain quantization unit uses a training data to quantize the gain of the targeted vector with a value most approximating a previously generated codebook, and

the shape quantization unit performs quantization by applying Algebraic vector quantization to the normalized targeted vector or quantizes the normalized targeted vector with the value most approaching the codebook.

11. A method for coding a signal in a communication system, comprising:

coding voice and audio signals based on a code excited linear prediction (CELP) coding method;

calculating residual signals of the voice and audio signals;

transforming the residual signal into a signal in a frequency domain;

using frequency coefficients of the residual signals to calculate frequency energy of the residual signals;

calculating energy concentrations of each vector dimension of the residual signals from the frequency energy of the residual signals; and

comparing the energy concentrations of each vector dimension to determine targeted vector dimensions of the residual signals.

12. The method of claim 11, wherein in the determining of the targeted vector dimension, the vector dimensions having a maximum value is determined as the targeted vector dimensions in the energy concentrations of each vector dimension.

13. The method of claim 11, wherein in the calculating of the residual signal, a difference between the voice and audio signals and a signal synthesized by a CELP codec is calculated by a CELP coding method.

14. The method of claim 11, wherein in the transforming into the frequency domain, the residual signals is transformed from a domain time into a frequency domain by a modified discrete cosine transform (MDCT) or a discrete Fourier transform (DFT).

15. The method of claim 11, further comprising:

applying a perceptual weighting filter to frequency coefficients of the residual signals to acquire weighting signals of the residual signals.

16. The method of claim 11, wherein in the calculating of the frequency energy, energy in a sub-band of the residual signals is calculated using the frequency coefficients of the residual signals.

17. The method of claim 11, wherein in the calculating of the energy concentrations, the frequency energies of the residual signals are arranged in an energy size sequence and then the frequency energy of each vector dimension is each calculated according to the size sequence to calculate the energy concentrations of each vector dimension.

18. The method of claim 11, further comprising:

allocating the frequency coefficients to the targeted vectors of the residual signals as much as the targeted vector dimension in a sequence that absolute values of the frequency coefficients are large to store the position of the targeted vector to which the frequency coefficients are allocated; and

calculating the position of the frequency coefficients allocated to the targeted vector to quantize the position of the targeted vector.

19. The method of claim 18, further comprising:

quantizing a gain of the targeted vector;

normalizing the targeted vector with the gain of the quantized targeted vector;

quantizing the normalized targeted vector; and

quantizing a position code of the targeted vector.

20. The method of claim 19, wherein in the quantizing of the gain, the gain of the targeted vector is quantized with a value most approximating a previously generated codebook using a training data, and

in the shape quantizing, Algebraic vector quantization is applied to the normalized targeted vector to perform quantization or the normalized targeted vector is quantized with the value most approaching the codebook.