KR101539256B1 - Encoder and decoder for encoding/decoding location information about important spectral component of audio signal - Google Patents

Abstract

An audio signal encoder and decoder are provided. The audio signal encoder can efficiently encode position information of important frequency components in consideration of distribution of important frequency components. The audio signal encoder can encode position information using various presentation methods. Also, the audio signal decoder can easily locate the important frequency components using the position information.

Encoder, decoder, audio, important frequency components, selection, extraction, position

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to an encoder and a decoder for encoding / decoding location information on important frequency components of an audio signal,

Embodiments of the present invention relate to audio signal encoders and audio signal decoders that encode and encode audio signals.

There is an increasing interest in techniques for encoding and compressing audio signals. For example, Moving Picture Experts Group (MPEG) audio achieves high-quality, high-efficiency encoding. That is, the MPEG audio encodes the source signal in consideration of perceptual characteristics of a human being, and uses a perceptual coding technique in particular.

In encoding the audio signal, it is necessary to encode all the frequency components of the audio signal in order to maximize the quality. However, in order to improve the compression efficiency, only a part of the frequency components of the audio signal may be selected and encoded. Here, among the frequency components of the audio signal, the components selected as objects to be encoded are referred to as important frequency components.

A method may be proposed in which only important frequency components are selected for compressing an audio signal and only selected important frequency components are encoded. At this time, the method generally improves the compression efficiency. When the audio signal is encoded by applying the above method, the encoder must notify the decoder of information on which frequency bands the important frequency components are selected, and the decoder can determine the position of the important frequency components using the information . At this time, the encoder needs to efficiently inform the decoder of the information, and it is necessary to reduce the cost required to transfer the information to the decoder.

An audio signal encoder according to an embodiment of the present invention includes a component extraction unit for extracting important spectral components to be encoded among a plurality of frequency components of an audio signal, And a position information encoding unit for encoding the extracted position information of the important frequency components.

At this time, the position information encoding unit may encode the position information by setting the flag information to the first identification value that has been promised in advance when the important frequency components exist continuously or do not exist continuously.

In this case, if the important frequency components are irregularly present, the position information encoding unit may encode the position information by setting the flag information to a second identification value that is predetermined.

In this case, the position information encoding unit groups the plurality of frequency components into a plurality of groups, sets flag information to a predetermined third identification value, and distributes the distribution of the important frequency components to each of the plurality of groups The location information can be encoded by considering the location information.

In this case, the position information encoding unit may classify the position information of the important frequency components into a plurality of subsets, and apply the context arithmetic coding technique to each of the plurality of subsets to encode the position information.

In addition, an audio signal decoder according to an embodiment of the present invention uses a location information of coded important frequency components to recognize a location of important frequency components, Quantizing and encoding the frequency components and encoding the position information of the important frequency components in consideration of the distribution of the important frequency components and using the important frequency components based on the position of the recognized important frequency components, And a decoding unit for decoding the audio signal.

According to another aspect of the present invention, there is provided an audio signal encoding method comprising: extracting important spectral components to be encoded among a plurality of frequency components of an audio signal; And encoding the extracted position information of the important frequency components.

According to another aspect of the present invention, there is provided a method for decoding an audio signal, the method comprising the steps of: recognizing positions of important frequency components using positional information of encoded important frequency components; Quantizing and encoding the components and encoding the position information of the important frequency components in consideration of the distribution of the important frequency components, and encoding the position information of the important frequency components using the important frequency components, And recovering the signal.

The audio signal encoder and audio signal encoding method according to an embodiment of the present invention efficiently expresses positional information of important frequency components in consideration of distribution of important frequency components and encodes them, thereby reducing the cost required to express positional information .

The audio signal encoder and audio signal encoding method according to an embodiment of the present invention expresses and encodes position information of important frequency components in different ways according to distribution of important frequency components, It is possible to encode important frequency components.

In addition, the audio signal decoder and audio signal decoding method according to an embodiment of the present invention effectively detects the position of the important frequency components using the position information of the important frequency components, thereby easily restoring the audio signal using the important frequency components .

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a graph showing frequency components of an audio signal including important frequency components according to an embodiment of the present invention.

Referring to FIG. 1, an audio signal may be represented by frequency components. In Figure 1, twenty frequency components are shown.

People can hear notes with a magnitude greater than the minimum audible limit in a certain frequency band, but can not hear notes with a small magnitude. At this time, the minimum audible limit of a specific audio signal varies depending on other audio signals to be heard together, and this is called a masking effect. Perceptual coding can encode an audio signal in consideration of the masking effect. In compressing an audio signal according to a perceptual encoding technique, important frequency components, which are perceptually important components among all frequency components of an audio signal, are selected and encoded. In FIG. 1, the frequency components 1, 2, 3, 5, 6, 8, 9, 12, 13, 15, 16, 18 are the important frequency components, and 4, 7, 10, 11, 14, 17, 19, The 20th frequency components are not important frequency components.

At this time, the audio signal encoder can select important frequency components to be encoded out of the 20 frequency components shown in FIG. In particular, the audio signal encoder can calculate the perceptual importance of frequency components based on the psychoacoustic model and select important frequency components based on the computed perceptual importance. In addition, the audio signal encoder can extract the important frequency components considering the masking effect (for example, the signal to mask ratio (SMR) of each of the 20 frequency components), and the audio signal encoder In order to keep the signal-to-noise ratios of all the frequency bands substantially the same, it is possible to extract the important frequency components by further considering the signal-to-noise ratio of each of the plurality of frequency components.

As shown in FIG. 1, when twelve important frequency components out of the twenty frequency components of the audio signal are selected, the audio signal encoder quantizes and decodes the important frequency components. In this case, in order for the audio signal decoder to properly recover the audio signal using the quantized and decoded important frequency components, the audio signal encoder must inform the audio signal decoder of the position information of the selected important frequency components. In this case, the audio signal encoder can set the position information of the important frequency components to '1' if each of the 20 frequency components is an important frequency component and set to '0' if it is not an important frequency component, and transmit them to the audio signal decoder. Here, the position information can be expressed as "1 1 1 0 1 1 0 1 1 0 0 1 1 0 1 1 0 1 0 0 '.

As described above, the audio signal encoder can express the position information using an identification value (e.g., 0 or 1) indicating whether or not the frequency components are selected as important frequency components for each of the 20 frequency components. However, it may be inefficient to represent positions of 20-bit positions of the important frequency components with respect to 20 frequency components. For example, unlike that shown in FIG. 1, when each of the 20 frequency components is selected as the important frequency components, it may be wasteful to represent the position information in 20 bits. Particularly, when compressing or encoding an audio signal at a high bit rate, it is important to reduce the number of bits required to represent position information because all frequency components are highly likely to be selected as important frequency components.

In this case, the audio signal encoder according to the embodiment of the present invention can encode the position information using various methods according to the distribution of the important frequency components, and can reduce the number of bits necessary to express the position information . In addition, the audio signal decoder according to an embodiment of the present invention can recognize positions of important frequency components using the encoded position information, and decode important frequency components that are efficiently encoded.

FIG. 2 is a graph showing an example in which important frequency components are continuously present. FIG.

Referring to FIG. 2, it can be seen that all 20 frequency components are selected as important frequency components. Here, the position information of the 20 significant frequency components can be expressed as "1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ".

In this case, the audio signal encoder according to an embodiment of the present invention can encode position information using flag information and additional information when important frequency components are continuously present. The audio signal encoder may set the flag information (Location_Flag) to the first identification value (0) and the additional information to '1'. That is, when the significant frequency components are continuously present, the flag information (Location_Flag) is set to '0', and since each frequency component is an important frequency component, the additional information is set to '1'.

As a result, the audio signal encoder can reduce the number of bits required to represent position information by encoding 20-bit position information into 2-bit information. Further, the audio signal encoder can know that all the frequency components are selected as the important frequency components using the flag information (Location_Flag = 0) and the additional information (1).

FIG. 3 is a graph showing an example in which important frequency components are continuously absent.

Referring to FIG. 3, it can be seen that none of the 20 frequency components is selected as an important frequency component. Therefore, the positional information of the important frequency components can be represented as "0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0".

At this time, the audio signal encoder according to the embodiment of the present invention can set the flag information (Location_Flag) to the first identification value (0) since the important frequency components do not exist continuously. Further, since the audio signal encoder does not have the important frequency components, the additional information is set to one bit "0 ". As a result, the audio signal encoder can encode position information using 1-bit flag information and 1-bit additional information.

4 is a graph showing an example in which important frequency components are irregularly present.

Referring to FIG. 4, it can be seen that twelve important frequency components out of the twenty frequency components are irregularly selected. Since the 1, 2, 3, 5, 6, 8, 9, 12, 13, 15, 16 and 18 frequency components are selected as the important frequency components, the position information of the important frequency components is "1 1 1 0 1 1 0 1 1 0 0 1 1 0 1 1 0 1 0 0 ".

At this time, the audio signal encoder according to the embodiment of the present invention can encode the position information by setting the flag information (Location_Flag) to the second identification value (1) because the important frequency components are irregular. That is, the flag information (Location_Flag) is set to the second identification value (1) and the additional information is set to "1 1 1 0 1 1 0 1 1 0 1 1 0 1 1 0 1 0 0".

Also, the audio signal decoder according to an embodiment of the present invention can confirm that the location information (Location_Flag) is the second identification value (1), and that the important frequency components exist irregularly. Then, the audio signal decoder can recognize the position of the important frequency components using the additional information.

5 is a graph showing an example in which a plurality of frequency components are grouped into a plurality of groups and position information is set for each of the plurality of groups.

Referring to FIG. 5, it can be seen that twelve important frequency components out of twenty frequency components are selected. Since the 1, 2, 3, 5, 6, 8, 9, 12, 13, 14, 15 and 16 frequency components are selected as the important frequency components, the position information of the important frequency components is "1 1 1 0 1 1 0 1 1 0 0 1 1 1 1 1 0 0 0 0 ".

At this time, the audio signal encoder according to an embodiment of the present invention can group 20 frequency components into 5 groups. Then, the audio signal encoder sets the flag information (Location_Flag) to the third identification value (2), and encodes the position information in consideration of the distribution of important frequency components for each of the five groups.

That is, in the first group including the first, second and third frequency components, since the important frequency components are continuously present, the audio signal encoder can set the sub flag information to "0 ". Since all the frequency components are selected as the important frequency components, the audio signal encoder sets the additional information of 1 bit to "1 ". As a result, the position information "1 1 1" of the first group is encoded into the sub flag information (0) and the additional information (1). Here, the overall flag information is "2 ".

In addition, the second group includes the 4th, 5th, 6th, and 7th frequency components. In the second group, the important frequency components are the 5th and 6th frequency components. At this time, since the important frequency components are irregularly present in the audio signal encoder, the sub flag information is set to "1 ", and the additional information is set to" 0 1 1 0 " As a result, the audio signal decoder can encode the position information of the second group using the whole flag information 2, the sub flag information 1, and the additional information 0 1 1 0.

Similarly to the second group, the audio signal encoder sets the sub-flag information to "1" because important frequency components are irregularly present in the third group. The additional information is set to "1 1 0 0 1" in the same manner as the original position information. Therefore, the position information of the third group can be encoded using the flag information 2, the sub flag information 1, and the additional information 1 1 0 0 1.

Also, the position information of the fourth group is encoded using the flag information 2, the sub flag information 0, and the additional information 1 similarly to the position information of the first group.

Also, the 17th, 18th, 19th and 20th frequency components of the fifth group are not all important frequency components. Therefore, the audio signal encoder sets the sub flag information to "0 " and the additional information to" 0 " since the important frequency components are not consecutively present. Then, the audio signal decoder can recognize the original position information (positions of the important frequency components) using the flag information 2, the sub flag information 0 and the additional information 0.

FIG. 6 is a diagram illustrating an example of applying an implicit arithmetic coding technique to each of the subsets of location information in accordance with the present invention.

Referring to FIG. 6, the stream (1 0 0 1 1 1 0 1 0 1 0 1 0 1 0 0 1 1 0 0 1 0 1 1) described at the top shows the original location information of the important frequency components , And characters shown at the bottom indicate encoded location information.

The audio signal encoder according to an embodiment of the present invention may classify the position information into six subsets and apply context arithmetic coding to each of the six subsets. In particular, since each of the subsets includes position information of 4 bits, the audio signal encoder can apply a context based 16-ary arithmetic coding technique based on 2 ⁴ -ary.

The position information of the first subset is represented by 0x9, and the position information of the second subset is represented by 0xD. The third, fourth, and fifth positional information is also represented by 0x8, 0xC, and 0xB. Then, the audio signal encoder can losslessly encode the extracted characters.

In particular, the audio signal encoder can encode the audio signal in consideration of the stochastic occurrence frequency of the characters. That is, a small number of bits may be allocated for frequently occurring characters, and a large number of bits may be allocated for non-frequently occurring characters.

7 is a flowchart illustrating an audio signal encoding method according to an embodiment of the present invention.

Referring to FIG. 7, in an audio signal encoding method according to an embodiment of the present invention, an audio signal encoder extracts important spectral components to be encoded among a plurality of frequency components of an audio signal (S710).

At this time, the audio signal encoder can calculate the perceptual importance of a plurality of frequency components using a psychoacoustic model, and extract important frequency components based on the calculated perceptual importance. Here, the perceptual importance can be expressed as a signal-to-mask ratio. In addition, the audio signal encoder may further extract the important frequency components by considering the signal-to-noise ratio of each of the plurality of frequency components.

In addition, the audio signal encoder according to an embodiment of the present invention analyzes the distribution of the important frequency components, thereby determining the presentation methods of the position information of the important frequency components (S720).

At this time, the audio signal encoder can classify the presentation methods using the flag information (Location_Flag). That is, the audio signal encoder can determine the presentation method suitable for the distribution of the important frequency components by determining the flag information (Location_Flag) as 0, 1, or 2.

In particular, when the significant frequency components are consecutively present or continuously absent, the audio signal encoder sets the flag information to zero. In addition, when the important frequency components are irregularly present, the flag information is set to one. When a plurality of frequency components are grouped into a plurality of groups, the flag information is set to two.

In addition, the audio signal encoder according to an embodiment of the present invention checks the flag information through steps S731, S732, and 733 to understand the presentation methods of the position information.

If the flag information is 1, the audio signal encoder according to the embodiment of the present invention determines whether all the bits of the position information are 1 or 0, and expresses the additional information as 0 or 1 (S741). That is, when all the bits of the position information are 1, since the important frequency components are continuously present, the additional information is represented by 1 bit of 1 bit. Conversely, when all the bits of the position information are 0, the important frequency components are continuously non-existent, so that the additional information is represented by one bit of zero.

As a result, when the important frequency components are continuously present or non-existent, the position information can be simply encoded through the flag information and additional information.

In addition, when the flag information is set to 1, the audio signal encoder according to the embodiment of the present invention may set 1 for each frequency component selected as important frequency components, 1 for each frequency component not selected as important frequency components By setting 0, additional information is expressed (S742).

At this time, the location information of the important frequency components is encoded using the flag information (1) and the additional information.

In addition, when the flag information is set to 2, the audio signal encoder according to the embodiment of the present invention grasps that a plurality of frequency components are grouped into a plurality of groups, The sub flag information is set in consideration of the distribution (S743).

If, in each of the groups, the significant frequency components are consecutively present or continuously absent, the audio signal encoder sets the sub-flag information to zero, and if the significant frequency components are irregularly present, The flag information is set to one.

In addition, the audio signal encoder according to an embodiment of the present invention applies additional presentation methods corresponding to sub-flag information to each of the groups to determine additional information (S750).

When the sub flag information is 0, the additional information is set to 1 when the important frequency components are consecutively present. Conversely, if the significant frequency components are continuously absent, the additional information is set to zero.

When the sub flag information is 1, 1 is set for each of the frequency components selected as the important frequency components, and 0 is set as the additional information for each of the frequency components not selected as the important frequency components.

8 is a block diagram illustrating an audio signal encoder according to an embodiment of the present invention.

Referring to FIG. 8, an audio signal encoder according to an embodiment of the present invention includes a component extracting unit 810, a quantizer 820, and a lossless encoding unit 830.

The component extraction unit 810 extracts important spectral components to be encoded among a plurality of frequency components of the audio signal. At this time, position information related to the position of the extracted important frequency components is provided to the lossless coding unit 830. [

The quantizer 820 quantizes the significant frequency components. At this time, it can be adaptively adjusted or quantizes important frequency components according to a predetermined quantization step. In particular, quantizer 820 may be a well-known Max-Lloyd quantizer.

The lossless coding unit 830 generates a bitstream by encoding the data stream and the positional information generated from the significant frequency components. At this time, the lossless coding unit 830 includes a component coding unit 831 and a position information coding unit 832. [

The component encoding unit 831 encodes the quantized important frequency components. In this case, the component coding unit 831 can perform lossless coding, and in particular, can perform context arithmetic coding. The position information encoding unit 832 encodes position information of the extracted important frequency components in consideration of the distribution of the extracted important frequency components. With respect to the coding of the position information by the position information coding unit 832, the matters described with reference to Figs. 1 to 7 can be applied as they are, so that the following description will be omitted.

As a result, the bit stream generated from the component encoding unit 831 and the position information encoding unit 832 is provided to the audio signal decoder. Here, the bitstream includes a component related to the data stream generated from the component encoding unit 831 and a component related to the encoded positional information generated from the positional information encoding unit 832. [

9 is a flowchart illustrating an audio signal decoding method according to an embodiment of the present invention.

Referring to FIG. 9, in an audio signal decoding method according to an embodiment of the present invention, an audio signal decoder extracts flag information from encoded location information (S910). Then, the audio signal decoder can grasp the expression of the positional information used by the audio signal encoder on the basis of which identification value the extracted flag information has.

Also, the audio signal decoder according to an embodiment of the present invention extracts additional information from the encoded position information (S920).

In addition, the audio signal decoder according to an embodiment of the present invention decodes the encoded location information using the extracted flag information and the additional information (S930). Thus, the original position information is revealed.

In addition, the audio signal decoder according to an embodiment of the present invention recognizes positions of important frequency components based on the found original position information (S940). Accordingly, the audio signal decoder can recognize which frequency components among the plurality of frequency components are selected as the important frequency components.

In addition, the audio signal decoder according to an embodiment of the present invention reconstructs an audio signal using positions of recognized important frequency components (S950).

10 is a block diagram illustrating an audio signal decoder according to an embodiment of the present invention.

Referring to FIG. 10, an audio signal decoder according to an embodiment of the present invention includes a lossless decoding unit 1010, a position recognition unit 1020, and an audio signal restoration unit 1030.

The lossless decoding unit 1010 losslessly decodes a bitstream provided from an audio signal encoder. Here, the bitstream includes a data stream generated from the significant frequency components and position information of the encoded important frequency components. At this time, the lossless decoding unit 1010 decodes the position information of the important frequency components from the bitstream and decodes the bits of the important frequency components from the bitstream.
The position recognition unit 1020 recognizes the positions of the important frequency components based on the positional information of the encoded important frequency components. Here, when the important frequency components exist continuously or do not exist continuously, the encoded position information includes flag information having a predetermined first identification value, and when important frequency components are irregularly present, The location information may include flag information having a predetermined second identification value. At this time, the position recognition unit 1020 can grasp the distribution of the important frequency components, and further the manner of expressing the position information, using the flag information. The position recognition unit 1020 can grasp the accurate position of the important frequency components using the additional information included in the encoded position information.

The audio signal decompression unit 1030 includes an inverse quantizer 1031 and a spectrum decompression unit 1032. The inverse quantizer 1031 dequantizes the quantization values of the important frequency components, and the spectral decompression unit 1032 decompresses the quantized values of the important frequency components according to the positions of the recognized important frequency components using the values of the inversely quantized important frequency components. Restores the spectra. Then, the restored important frequency components can be finally converted into a time-domain audio signal. That is, since the restored important frequency components are values in the frequency domain, the values of the frequency domain are not explicitly shown in FIG. 10, but are converted into time domain audio signals through a subsequent signal processing process.

delete

The audio signal encoding / decoding method according to the present invention can be implemented in the form of a program command that can be executed through various computer means and recorded in a computer readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

While the invention has been shown and described with reference to certain preferred embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the equivalents of the claims, as well as the claims.

1 is a graph showing frequency components of an audio signal including important frequency components according to an embodiment of the present invention.

FIG. 2 is a graph showing an example in which important frequency components are continuously present. FIG.

FIG. 3 is a graph showing an example in which important frequency components are continuously absent.

4 is a graph showing an example in which important frequency components are irregularly present.

5 is a graph showing an example in which a plurality of frequency components are grouped into a plurality of groups and position information is set for each of the plurality of groups.

FIG. 6 is a diagram illustrating an example of applying an implicit arithmetic coding scheme to each of the subsets of location information in accordance with the present invention.

7 is a flowchart illustrating an audio signal encoding method according to an embodiment of the present invention.

8 is a block diagram illustrating an audio signal encoder according to an embodiment of the present invention.

9 is a flowchart illustrating an audio signal decoding method according to an embodiment of the present invention.

10 is a block diagram illustrating an audio signal decoder according to an embodiment of the present invention.

Claims (22)

A component extraction unit for extracting important spectral components to be encoded among a plurality of frequency components of an audio signal; And A position information encoding unit for encoding position information of the extracted important frequency components in consideration of the distribution of the extracted important frequency components, Lt; / RTI > The location information encoding unit Grouping the plurality of frequency components into a plurality of groups, setting flag information to a predetermined third identification value, encoding the position information by considering the distribution of the important frequency components for each of the plurality of groups, And outputs the audio signal. The method according to claim 1, The location information encoding unit Wherein the encoder encodes the position information by setting flag information to a first identification value that is predetermined when the important frequency components are present continuously or non-existently. The method according to claim 1, The location information encoding unit And encodes the position information using an identification value indicating whether each of the plurality of frequency components is included in the important frequency components. 3. The method of claim 2, The location information encoding unit Wherein the encoder encodes the position information by setting the flag information to a second identification value that is predetermined when the important frequency components are irregularly present. delete The method according to claim 1, The location information encoding unit And encodes the position information by setting sub-flag information for each of the plurality of groups based on distribution of the important frequency components belonging to each of the plurality of groups. The method according to claim 1, The location information encoding unit Wherein the location information of the important frequency components is classified into a plurality of subsets and the location information is encoded by applying a context arithmetic coding technique to each of the plurality of subsets. The method according to claim 1, A quantizer for quantizing the important frequency components; And A component encoding unit for encoding the quantized important frequency components, Further comprising: a decoder for decoding the audio signal. The method according to claim 1, The component extracting unit Wherein the important frequency components are extracted in consideration of perceptual importance of the plurality of frequency components calculated using a psychoacoustic model. 10. The method of claim 9, The component extracting unit And extracts the important frequency components based on a signal to mask ratio (SMR) of the plurality of frequency components. The method according to claim 1, The component extracting unit And extracts the important frequency components in consideration of a signal-to-noise ratio of each of the plurality of frequency components. The position-or sub-encoder, which recognizes the position of the important frequency components using the position information of the encoded important frequency components, quantizes and codes the important frequency components among the plurality of frequency components of the audio signal, And encodes the position information of the important frequency components. And And restoring the audio signal using the important frequency components based on the position of the recognized important frequency components, Lt; / RTI > The encoder comprising: Grouping the plurality of frequency components into a plurality of groups, setting flag information to a predetermined third identification value, encoding the position information by considering the distribution of the important frequency components for each of the plurality of groups, Wherein the audio signal decoder comprises: 13. The method of claim 12, Wherein the encoded position information includes flag information having a predetermined first identification value if the significant frequency components are present continuously or not continuously, Wherein when the important frequency components are irregularly present, the encoded location information includes the flag information having a predetermined second identification value. 13. The method of claim 12, The audio signal restoring unit Quantizes the quantized critical frequency components, and restores the audio signal using the inversely quantized critical frequency components based on the position of the recognized important frequency components. Extracting important spectral components to be encoded among a plurality of frequency components of an audio signal; And Encoding the location information of the extracted important frequency components in consideration of the distribution of the extracted important frequency components Lt; / RTI > The step of encoding the position information Grouping the plurality of frequency components into a plurality of groups, setting flag information to a predetermined third identification value, encoding the position information by considering the distribution of the important frequency components for each of the plurality of groups, And the audio signal is encoded. 16. The method of claim 15, The step of encoding the position information And encoding the position information by setting flag information to a first identification value that is predetermined when the important frequency components are present continuously or non-existently. 16. The method of claim 15, The step of encoding the position information And encoding the position information by setting the flag information to a second identification value that is predetermined when the important frequency components are irregularly present. delete 16. The method of claim 15, The step of encoding the position information And coding the position information by classifying the position information of the important frequency components into a plurality of subsets and applying a context arithmetic coding technique to each of the plurality of subsets. Way. Recognizing the position of the important frequency components using the position information of the encoded important frequency components, the encoder quantizing and encoding the important frequency components among the plurality of frequency components of the audio signal, and considering the distribution of the important frequency components And encoding the position information of the important frequency components. And And restoring the audio signal using the important frequency components based on the position of the recognized important frequency components Lt; / RTI > The encoder comprising: Grouping the plurality of frequency components into a plurality of groups, setting flag information to a predetermined third identification value, encoding the position information by considering the distribution of the important frequency components for each of the plurality of groups, And the audio signal is decoded. 21. The method of claim 20, Wherein the encoded position information includes flag information having a predetermined first identification value if the significant frequency components are present continuously or not continuously, Wherein when the important frequency components are irregularly present, the encoded position information includes the flag information having a predetermined second identification value. A computer-readable recording medium on which a program for performing the method of any one of claims 15 to 17 and 19 to 21 is recorded.

Images