RU2484542C2 - Device for encoding stereophonic signals, device for decoding stereophonic signals and methods realised by said devices - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: device for encoding stereophonic signals disclosed in the present invention employs a configuration, having: a sum and difference calculator which generates a monaural signal related to a sum of a left and a right signal forming a stereo signal, and generates a side signal related to the difference between the left and right channel signals; a mode information generator which generates mode information per layer indicating a coding mode of one of monaural coding and stereophonic coding; and first to N-th layer coders which perform monaural coding in an i-th layer (i=1, 2 …, N, where N is an integer equal to or greater than 2) using information related to the monaural signal or performs stereophonic coding in the i-th layer using both the information related to the monaural signal and information related to the side signal, based on the mode information, and provide i-th layer encoded information.

EFFECT: providing scalable coding based on correlation between left and right channel signals and information value and setting layers for performing monaural and stereophonic coding.

13 cl, 20 dwg

Description

FIELD OF THE INVENTION

The present invention relates to a stereo signal encoding apparatus, stereo signal decoding apparatus, and encoding and decoding methods that are used to encode stereo speech.

State of the art

In mobile communications, compression coding for digital speech and image information is essential for the efficient use of transmission frequency bands. In particular, the technology of the speech codec (encoding and decoding), widely used in mobile phones, has high expectations, and the need for additionally increased sound quality is constantly growing with traditional high-performance encoding with high compression efficiency.

Recently, with the expansion of the bandwidth of communication networks, a need arises for realizing high sound quality in voice communication, and in order to satisfy this need, voice communication systems using stereo coding technologies have been developed.

As a method of encoding stereophonic speech, a conventional conventional method for detecting monophonic and side signals and encoding these signals is provided, wherein the monophonic signal is the sum of the left channel signal and the right channel signal, and the side signal is the difference between the left channel signal and the right signal channel (see patent document 1).

The left channel signal and the right channel signal represent the sound that the human left and right ears listen to, the monaural signal can represent common elements between the left channel signal and the right channel signal, and the side signal can represent the spatial difference between the left channel signal and the right channel signal.

There is a high correlation between the left channel signal and the right channel signal. Therefore, compared with the case where the right channel signal and the left channel signal are directly encoded, more appropriate coding can be performed in accordance with the characteristics of the monophonic signal and the side signal by converting the right channel signal and the left channel signal into a monophonic and side signal and then encoding these converted signals so that it is possible to implement coding with less redundancy, low bit rate (bit rate) and high quality.

Recently, the standardization of a scalable codec having a multilayer configuration has been studied, for example, in ITU-T (the telecommunication standardization sector of the International Telecommunication Union) and MPEG (motion picture expert group), and a more efficient and higher-quality speech codec is required.

For example, a scalable encoding device based on ITU-T G.729.1 performs encoding based on ITU-T G.729.1 recommendation at 8 kbps and, through additional encoding of the enhancement layer, can perform encoding on twelve types of bitrates, for example 8 kbps, 12 kbps, 14 kbps, 16 kbps, 18 kbps, 20 kbps, 22 kbps, 24 kbps, 26 kbps, 28 kbps, 30 kbps and 32 kbps This scalability is realized by sequentially coding distortion when coding the lower layer in the upper layer. Thus, the scalable G.729.1 coding device is formed using one base layer with a bit rate of 8 kbps, one enhancement layer with a bitrate of 4 kbps and ten enhancement layers with a bitrate of 2 kbps.

In addition, as a technology for performing scalable encoding of stereo signals, a stereo signal encoding device is disclosed in Patent Document 2. This stereo signal encoding device expresses additional information for each layer by a predetermined number of bits and, using a predetermined probabilistic model, performs arithmetic encoding of the bit sequences in order from highest bit sequence to low bit howling sequence. Here, this stereo signal encoding device has a sign of switching between the left channel signal and the right channel signal according to a predetermined rule and encoding these signals.

Patent Document 1. Patent Application Laid-Open (Japan) No. 2001-255892

Patent Document 2. Patent Application Laid-Open (Japan) No. HEI11-317672

SUMMARY OF THE INVENTION

Problems Resolved by the Invention

However, as described above, the stereo signal encoding apparatus disclosed in Patent Document 2 is configured to switch between the left channel signal and the right channel signal according to a predetermined rule and encode these signals, i.e. this coding is independent of the correlation between the left channel signal and the right channel signal and the significance of the information. In addition, there is a problem in that, although it is preferable to define a layer for performing monophonic encoding and a layer for performing stereo encoding by means of user operations in a stereo encoding apparatus that performs scalable encoding, the stereo encoding apparatus disclosed in Patent Document 2 is not can support this task.

Therefore, it is an object of the present invention to provide a stereo signal encoding apparatus, stereo signal decoding apparatus, and encoding and decoding methods for performing scalable encoding based on the correlation between the left channel signal and the right channel signal and the significance of the information and for defining a layer for performing monaural encoding and layer for performing stereo encoding.

Problem Solver

The stereo signal encoding apparatus of the present invention uses a configuration having: a sum and difference calculation module that generates a monophonic signal associated with the sum of the first channel signal and the second channel signal forming the stereo signal and generates a side signal associated with the difference between the signal of the first channel and the signal of the second channel; a mode information generating unit that generates mode information per layer indicating a coding mode of one of monophonic coding and stereo coding; and coding modules of the first through Nth layers, which perform monophonic coding in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) using information related to monophonic signal, or perform stereo coding in the i-th layer using both information related to the monaural signal and information associated with the side signal based on the mode information and provide encoded information of the i-th layer.

The stereo signal decoding apparatus of the present invention uses a configuration having: a receiving module that receives mode information and encoded layer 1 to N layers information detected by encoding processing in layers 1 to N of which the mode information indicates which monaural encoding and stereo encoding is performed during encoding processing in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) of the stereo encoding device fishing, which performs coding using a first channel signal and second channel signal forming a stereo signal; first to Nth layer decoding modules that perform monophonic decoding or stereo decoding using encoded information of the i-th layer based on mode information and provide the result of decoding the monaural signal in the i-th layer and the result of decoding the side signal in the i-th layer wherein the monaural signal is associated with the sum of the signal of the first channel and the signal of the second channel, and the side signal is associated with the difference between the signal of the first channel and the signal of the second channel; and a sum and difference calculator that calculates a decoded signal of the first channel and a decoded signal of the second channel using the result of decoding the monaural signal in the Nth layer and the result of decoding the side signal in the Nth layer.

The encoding method of the present invention includes the steps of: generating a monophonic signal associated with the sum of the signal of the first channel and the signal of the second channel forming the stereo signal, and generating a side signal associated with the difference between the signal of the first channel and the signal of the second channel; generating mode information per layer indicating a coding mode of one of monophonic coding and stereo coding; and performing monophonic coding in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) using information associated with the monophonic signal, or performing stereo coding in the i-th layer with using both information related to the monophonic signal and information related to the side signal based on the mode information and providing encoded information of the i-th layer.

The decoding method of the present invention includes the steps of: receiving mode information and encoded information of the first through Nth layers detected by encoding processing in the first through Nth layers, the mode information indicating which of the monophonic encoding and stereo encoding is performed when encoding processing in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) of the stereo signal encoding device, which performs encoding using the signal of the first the signal and the second channel signal forming a stereo signal; performing monophonic decoding or stereo decoding using the encoded information of the i-th layer based on the mode information and providing the result of decoding the monophonic signal in the i-th layer and the result of decoding the side signal in the i-th layer, the monaural signal being associated with the sum of the signal of the first channel and the signal of the second channel, and the side signal is associated with the difference between the signal of the first channel and the signal of the second channel; and computing a decoded signal of the first channel and a decoded signal of the second channel using the result of decoding the monaural signal in the Nth layer and the result of decoding the side signal in the Nth layer.

Advantages of the Invention

According to the present invention, by performing scalable encoding of a monaural signal (M signal) and side signal (S signal) calculated from the L signal and the R signal of the stereo signal, and setting the encoding mode for each layer in scalable encoding based on the mode information scalable coding can be performed according to the correlation between the left channel signal and the right channel signal and based on the significance of the information. Furthermore, according to the present invention, it is possible to define a layer for performing monophonic encoding and a layer for performing stereo encoding, so that the degree of freedom in controlling the accuracy of encoding can be increased.

Brief Description of the Drawings

1 is a block diagram showing the main components of a stereo signal encoding apparatus according to Embodiment 1 of the present invention;

FIG. 2 is a block diagram showing main components in a base layer encoding module according to Embodiment 1 of the present invention;

FIG. 3 illustrates operations in the case where a monaural encoding mode is set in a base layer encoding module according to Embodiment 1 of the present invention;

4 illustrates operations in the case where a stereo coding mode is set in a base layer coding unit according to Embodiment 1 of the present invention;

5 is a block diagram showing the main components in a monophonic coding module according to Embodiment 1 of the present invention;

6 is a flowchart showing a search algorithm in a zone search module according to Embodiment 1 of the present invention;

7 shows an example of a spectrum represented by pulses that are searched in the zone search module according to Embodiment 1 of the present invention;

FIG. 8 is a flowchart showing preprocessing a search algorithm in a full search module according to Embodiment 1 of the present invention;

FIG. 9 is a flowchart showing a search by the search algorithm of the full search module according to Embodiment 1 of the present invention;

10 illustrates an example of a spectrum represented by pulses that are searched in the zone search module and the full search module according to Embodiment 1 of the present invention;

11 is a block diagram showing main components in a monaural decoding module according to Embodiment 1 of the present invention;

12 is a flowchart showing a decoding algorithm of a spectrum decoding module according to Embodiment 1 of the present invention;

13 is a block diagram showing main components in a stereo coding unit according to Embodiment 1 of the present invention;

Fig. 14 illustrates a state where an M-signal spectrum and an S-signal spectrum are integrated in an integration module according to Embodiment 1 of the present invention;

Fig. 15 illustrates bit allocation in a spectrum encoding module according to Embodiment 1 of the present invention;

16 is a block diagram showing main components in a stereo decoding module according to Embodiment 1 of the present invention;

17 is a block diagram showing the main components of a stereo signal decoding apparatus according to Embodiment 1 of the present invention;

Fig. 18 is a block diagram showing main components in a base layer decoding module according to Embodiment 1 of the present invention;

Fig is a block diagram showing the main components in the decoding module of the second enhancement layer according to a variant implementation 1 of the present invention; and

20 is a block diagram showing the main components of a stereo signal encoding apparatus according to Embodiment 2 of the present invention.

The optimal mode of carrying out the invention

Embodiments of the present invention are explained in detail below with reference to the accompanying drawings.

First Embodiment

1 is a block diagram showing the main components of a stereo signal encoding apparatus 100 according to Embodiment 1 of the present invention. An example case is described in which the stereo signal encoding apparatus 100 according to Embodiment 1 of the present invention provides one base layer and three enhancement layers. The following illustrates an exemplary case in which a stereo signal consists of a left channel signal (hereinafter “L-signal”) and a right channel signal (hereinafter “R-signal”).

1, a stereo signal encoding apparatus 100 comprises a sum and difference calculator 101, a mode setting module 102, a base layer encoding module 103, a first enhancement layer encoding module 104, a second enhancement layer encoding module 105, a third enhancement layer encoding module 106 and a module 107 multiplexing.

The sum and difference calculating unit 101 calculates a sum signal (ie, a monophonic signal, hereinafter “M-signal”) and a difference signal (ie, side signal, hereinafter “S-signal”) using the L signal and R-signal according to the following equations 1 and 2 and outputs the results to the module 103 encoding the base layer. Here, the L signal and the R signal represent the sound that the human left and right ears are listening to, the M signal may represent common elements between the L signal and the R signal, and the S signal may represent the spatial difference between the L signal and R -signal.

M _i = L _i + R _i ... (equation 1)

S _i = L _i -R _i ... (equation 2)

In equations 1 and 2, the subscript “i” represents the sample number of each signal, but signals can be represented without “i”.

For example, an M _i signal may be recorded simply as an M signal.

The mode information for setting the encoding mode in the encoding units from the base layer encoding unit 103, the first enhancement layer encoding unit 104, the second enhancement layer encoding unit 105 and the third enhancement layer encoding unit 106 is received as input in the mode setting unit 102 by user operations and then output to these encoding modules and multiplexing module 107. Here, user operations include keyboard input, DIP switch and buttons, and booting from a PC (personal computer), etc.

The encoding mode in each encoding module corresponds to a monaural encoding mode for encoding only M-signal information or a stereo encoding mode for encoding both M-signal information and S-signal information. Here, “M-signal information” typically refers to the M-signal itself or coding distortion associated with the M-signal in each layer. In addition, “S-signal information” typically refers to the S-signal itself or coding distortion associated with the S-signal in each layer.

Further, the encoding mode in each layer is shown using each of the bits of the mode information. Thus, in bits, the value “0” represents the monaural encoding mode, and the value “1” represents the stereo encoding mode. More specifically, for example, each of the four bits of mode information is used to sequentially represent encoding modes in the base layer encoding unit 103, the first enhancement layer encoding unit 104, the second enhancement layer encoding unit 105 and the third enhancement layer encoding unit 106.

For example, the four-bit mode information “0000” means that monophonic coding is performed in all layers. In this case, the stereo signal encoding apparatus 100 can encode the M signal with maximum quality. In addition, for example, the “0011” mode information means that the encoding mode in the base layer encoding unit 103 and the first enhancement layer encoding unit 104 is a monophonic encoding mode, and the encoding mode in the second enhancement layer encoding unit 105 and the third enhancement layer encoding unit 106 is a stereo coding mode. In addition, for example, “1111” mode information means that stereo coding is performed in all layers. In this case, the stereo signal encoding apparatus 100 may encode the M signal and the S signal with equal weighting. Thus, using the information of the four-bit mode, sixteen types of encoding modes in four encoding modules can be represented.

In the present embodiment, the mode information output from the mode setting unit 102 is received in each coding unit and the multiplexing unit 107 as identical input information of the four-bit mode. Additionally, each encoding module checks only one bit of the four input bits required to set the encoding mode, and sets the encoding mode. Thus, in four bits of mode input information, the base layer encoding module 103 checks the first bit, the first enhancement layer encoding module 104 checks the second bit, the second enhancement layer encoding module 105 checks the third bit, and the third enhancement layer encoding module 106 checks the fourth bit.

However, instead of entering identical four-bit mode information in each encoding unit, the mode setting module 102 can pre-sort one bit required to set the encoding mode in each encoding unit and output one bit to each encoding unit. Thus, in the four-bit mode information, the mode setting module 102 can enter only the first bit in the base layer encoding unit 103, only the second bit in the first enhancement layer encoding unit 104, only the third bit in the second enhancement layer encoding unit 105, and only the fourth bit in module 106 encoding the third enhancement layer.

In addition, in any of the above cases, the mode information received as input from the mode setting unit 102 to the multiplexing unit 107 is referred to as four-bit mode information.

In the base layer encoding unit 103, the monaural encoding mode or the stereo encoding mode are set based on the mode information received as input from the mode setting unit 102. After setting the monophonic encoding mode in the base layer encoding unit 103, the base layer encoding unit 103 encodes only the M signal received as input from the sum and difference calculation unit 101 and outputs the resulting monophonic encoded information to the multiplexing unit 107 as encoded information of the base layer. Additionally, the base layer encoding module 103 finds and outputs distortion when encoding the base layer of the M signal received as input from the sum and difference calculation module 101 to the encoding module of the first enhancement layer as information of the M signal in the base layer and outputs S- a signal received as input from the sum and difference calculation unit 101, as is the encoding unit 104 of the first enhancement layer as S-signal information in the base layer. On the contrary, after setting the stereo coding mode in the base layer coding unit 103, the base layer coding unit 103 encodes both the M signal and the S signal received as input from the sum and difference calculation unit 101 and outputs the resulting stereo encoded information to the module 107 multiplexing as encoded information of the base layer. Additionally, the base layer encoding unit 103 finds distortions in encoding the base layer of M and S signals received as input from the sum and difference calculation unit 101, and outputs the results to the first enhancement layer encoding unit 104 as M signal information in the base layer and S-signal information in the base layer. In addition, the base layer encoding module 103 is explained in more detail below.

In the encoding module 104 of the first enhancement layer, the monaural encoding mode or the stereo encoding mode are set based on the mode information received as input from the mode setting module 102. After setting the monophonic encoding mode in the encoding module 104 of the first enhancement layer, the encoding module 104 of the first enhancement layer encodes the M-signal information in the base layer received as input from the base layer encoding module 103 and outputs the resulting monophonic encoded information to the multiplexing module 107 as encoded information of the first enhancement layer. Additionally, using the information of the M signal in the base layer received as input from the base layer encoding unit 103, the encoding module 104 of the first enhancement layer finds and outputs distortion when encoding the first enhancement layer associated with the M signal to encoding module 105 of the second the enhancement layer as M-signal information in the first enhancement layer and outputs S-signal information in the base layer, taken as input from the base layer encoding unit 103, as is the second enhancement layer encoding unit 105 as S-signal information in the first enhancement layer.

In contrast, after setting the stereo coding mode in the encoding module 104 of the first enhancement layer, the encoding module 104 of the first enhancement layer encodes both the M-signal information in the base layer and the S-signal information in the base layer received as input from module 103 encoding the base layer, and outputs the resulting stereo encoded information to the multiplexing unit 107 as encoded information of the first enhancement layer. Further, by using the information of the M signal in the base layer and the information of the S signal in the base layer received as input from the base layer encoding unit 103, the encoding module 104 of the first enhancement layer finds and outputs distortions when encoding the first enhancement layer associated with M - and S-signals, to the encoding module 105 of the second enhancement layer as the M-signal information in the first enhancement layer and the S-signal information in the first enhancement layer. In addition, the encoding module 104 of the first enhancement layer is explained in more detail below.

In the encoding unit 105 of the second enhancement layer, the monaural encoding mode or the stereo encoding mode are set based on the mode information received as input from the mode setting unit 102. After setting the monophonic encoding mode in the encoding module 105 of the second enhancement layer, the encoding module 105 of the second enhancement layer encodes the M signal information in the first enhancement layer received as input from the encoding module 104 of the first enhancement layer and outputs the resulting monophonic encoded information to the multiplexing module 107 as encoded information of the second enhancement layer. Additionally, using the information of the M signal in the first enhancement layer received as input from the encoding module 104 of the first enhancement layer, the encoding module 105 of the second enhancement layer finds and outputs distortion when encoding the second enhancement layer associated with the M signal to module 106 encoding the third enhancement layer as M-signal information in the second enhancement layer and outputs the S-signal information in the first enhancement layer, taken as input from the encoding module 104 of the first enhancement layer, as is the fashion s 106 of the third enhancement layer coding information as the S-signal in the second enhancement layer.

In contrast, after setting the stereo coding mode in the encoding module 105 of the second enhancement layer, the encoding module 105 of the second enhancement layer encodes both the M-signal information in the first enhancement layer and the S-signal information in the first enhancement layer, taken as input from a coding unit 104 of the first enhancement layer, and outputs the resulting stereo encoded information to the multiplexing unit 107 as encoded information of the second enhancement layer. Further, by using the M signal information in the first enhancement layer and the S signal information in the first enhancement layer, taken as input from the encoding module 104 of the first enhancement layer, the encoding module 105 of the second enhancement layer finds and outputs distortion when encoding the second enhancement layer, associated with the M and S signals, to the encoding module 106 of the third enhancement layer as the M signal information in the second enhancement layer and the S signal information in the second enhancement layer. In addition, the encoding module 105 of the second enhancement layer is explained in more detail below.

In the encoding module 106 of the third enhancement layer, the monophonic encoding mode or the stereo encoding mode are set based on the mode information received as input from the mode setting module 102. After setting the monophonic coding mode in the encoding module 106 of the third enhancement layer, the encoding module 106 of the third enhancement layer encodes the M-signal information in the second enhancement layer received as input from the encoding module 105 of the second enhancement layer and outputs the resulting monophonic encoded information to the multiplexing module 107 as encoded information of the third enhancement layer.

In contrast, after setting the stereo coding mode in the encoding module 106 of the third enhancement layer, the encoding module 106 of the third enhancement layer encodes both the M signal information in the second enhancement layer and the S signal information in the second enhancement layer, taken as input from a second enhancement layer encoding unit 105, and outputs the resulting stereo encoded information to the multiplexing unit 107 as encoded information of the third enhancement layer. In addition, the encoding module 106 of the third enhancement layer is explained in more detail below.

The multiplexing unit 107 multiplexes the mode information received as input from the mode setting unit 102, the encoded base layer information received as an input from the base layer encoding unit 103, the encoded information of the first enhancement layer received as input from the encoding unit 104 of the first enhancement layer , encoded information of the second enhancement layer received as input from the encoding module 105 of the second enhancement layer, and encoded information of the third enhancement layer Loy, received as input from encoding unit 106 of the third enhancement layer, and generates bit streams that are to be transmitted to stereo decoding apparatus.

In the stereo signal encoding apparatus 100, the base layer encoding module 103, the first enhancement layer encoding module 104 and the second enhancement layer encoding module 105 have the same configuration and therefore perform basically the same operations, and differ from each other only in their input signals and output signals . The encoding module 106 of the third enhancement layer does not require a configuration to find an encoding distortion and, therefore, differs from the above three encoding modules in terms of configuration. Thus, the third enhancement layer encoding unit 106 uses a configuration with the exception of monaural decoding unit 303, stereo decoding unit 306, switch 307, adder 308, adder 309, and switch 310 from the configuration shown in FIG. 2. Regarding the above three encoding modules having the same configuration, for example, the base layer encoding module 103 receives an M signal and an S signal as input; when performing monophonic encoding, outputs to the encoding module 104 of the first enhancement layer the distortion when encoding the base layer of the M-signal as M-signal information and the S-signal itself as S-signal information; and when performing stereo coding, outputs to the encoding unit 104 of the first enhancement layer a distortion when encoding the base layer of the M signal as M signal information and a distortion when encoding the base layer of the S signal as S signal information.

In addition, the encoding module 104 of the first enhancement layer and the encoding module 105 of the second enhancement layer: take as input the M-signal information in the previous layer and the S-signal information in the available layer; when performing monophonic coding, the coding distortion detected by additionally encoding the M-signal information in the previous layer and the S-signal information itself in the previous layer are output to the coding module in the subsequent layer; and when stereo coding is performed, coding distortion detected by additional encoding of the M-signal information in the previous layer and coding distortion detected by additional encoding of the S-signal in the previous layer are output to the encoding module in the subsequent layer. Further, the configurations and operations of the above encoding modules are explained using the base layer encoding module 103 as an example.

2 is a block diagram showing the main components in the base layer encoding unit 103.

2, the base layer encoding module 103 comprises a switch 301, a monophonic encoding module 302, a monophonic decoding module 303, a switch 304, a stereo encoding module 305, a stereo decoding module 306, a switch 307, an adder 308, an adder 309, a switch 310 and a switch 311 .

If the value of the first bit of the mode information received as input from the mode setting module 102 is “0”, the switch 301 outputs the M signal received as input from the sum and difference calculation module 101 to the monophonic encoding module 302, and if the value the first bit of the mode information received as input from the mode setting module 102 is “1”, outputs the M signal received as input from the sum and difference calculation unit 101 to the stereo coding unit 305.

The monophonic encoding unit 302 performs encoding (i.e., monophonic encoding) using the M signal received as input from the switch 301 and outputs the resulting monophonic encoded information to the monophonic decoding unit 303 and the switch 311. In addition, the monophonic encoding unit 302 explained in more detail below.

Monophonic decoding unit 303 decodes monophonic encoded information received as input from monophonic encoding unit 302 and outputs the resulting decoded signal (i.e., monophonic decoded M signal) to switch 307. In addition, monophonic decoding unit 303 is explained in more detail below.

If the value of the first bit of the mode information received as input from the mode setting module 102 is “1”, the switch 304 outputs the S signal received as input from the sum and difference calculation unit 101 to the stereo coding unit 305.

The stereo coding unit 305 performs encoding (i.e., stereo coding) using the M signal received as input from the switch 301 and the S signal received as input from the switch 304, and outputs the resulting stereo encoded information to the module 306 stereo decoding and a switch 311. In addition, the stereo encoding unit 305 is explained in more detail below.

The stereo decoding unit 306 decodes the stereo encoded information received as input from the stereo encoding unit 305 and outputs two resulting decoded signals, i.e. a stereo decoded M signal and a stereo decoded S signal to switch 307 and adder 309, respectively.

If the value of the first bit of the mode information received as input from the mode setting module 102 is “0”, the switch 307 outputs the monaural decoded M signal received as input from the monaural decoding unit 303 to the adder 308, or if the value of the first bit the mode information received as input from the mode setting module 102 is “1”, the stereo decoded M signal received as input from the stereo decoding unit 306 is output to the adder 308.

An adder 308 calculates the difference between the M signal received as input from the sum and difference calculation module 101 and one of the monophonic decoded M signal and the stereo decoded M signal received as input from the switch 307 as a distortion when encoding the base layer M-signal. Additionally, the adder 308 outputs this distortion when encoding the base layer of the M signal to the encoding module 104 of the first enhancement layer as information of the M signal in the base layer.

An adder 309 calculates the difference between the S signal received as input from the sum and difference calculation unit 101 and the stereo decoded S signal received as input from the stereo decoding unit 306 as a distortion in encoding the base layer of the S signal. Additionally, adder 309 outputs this distortion when encoding the base layer of the S signal to switch 310.

If the value of the first bit of the mode information received as input from the mode setting module 102 is “0”, the switch 310 outputs an S signal received as input from the sum and difference calculation unit 101, as is the first enhancement layer encoding unit 104 as S-signal information in the base layer. If the value of the first bit of the mode information received as input from the mode setting module 102 is “1”, the switch 310 outputs distortion when encoding the base layer of the S signal received as input from the adder 309 to the encoding module 104 of the first enhancement layer as S-signal information in the base layer.

If the value of the first bit of the mode information received as input from the mode setting module 102 is “0”, the switch 311 outputs the monophonic encoded information received as input from the monaural encoding unit 302 to the multiplexing unit 107 as encoded base layer information. If the value of the first bit of the mode information received as input from the mode setting unit 102 is “1”, the switch 311 outputs the stereo encoded information received as input from the stereo encoding unit 305 to the multiplexing unit 107 as encoded base layer information.

FIG. 3 illustrates operations in the case where the monaural encoding mode is set in the base layer encoding unit 103 based on the value “0” of the first bit of the mode information received as input from the mode setting unit 102.

As shown in FIG. 3, when the monaural encoding mode is set in the base layer encoding unit 103, the stereo encoding unit 305, the stereo decoding unit 306 and the adder 309 do not work, and the monaural encoding unit 302 and the monaural decoding unit 303 work. In addition, the adder 308 finds the residual signal between the monophonic decoded M-signal received as input from the monophonic decoding unit 303 via the switch 307, and the M-signal received as input from the sum and difference calculation unit 101, as a distortion when encoding the base M-signal layer. In addition, the switch 310 outputs an S signal received as input from the sum and difference calculator 101, as is the first enhancement layer encoding module 104. The switch 311 outputs the monophonic encoded information received as input from the monophonic encoding unit 302 to the multiplexing unit 107 as the encoded information of the base layer.

FIG. 4 illustrates operations in the case where the stereo coding mode is set in the base layer encoding unit 103 based on the value “1” of the first bit of the mode information received as input from the mode setting unit 102.

As shown in FIG. 4, when the stereo coding mode is set in the base layer encoding unit 103, the monaural encoding unit 302 and the monaural decoding unit 303 do not work, and the stereo encoding unit 305, the stereo decoding unit 306, and the adder 309 operate. In addition, the adder 308 finds a residual signal between the stereo decoded M-signal received as input from the stereo decoding unit 306 and the M-signal received as input from the sum and difference calculation unit 101, as distortion when encoding the base layer M- signal. In addition, the switch 310 outputs distortion when encoding the base layer of the S signal received as input from the adder 309 into the encoding module 104 of the first enhancement layer. The switch 311 outputs the stereo encoded information received as input from the stereo encoding unit 305 to the multiplexing unit 107 as encoded information of the base layer.

5 is a block diagram showing the main components in monophonic coding unit 302.

5, the monophonic coding module 302 comprises an LPC (linear prediction coefficient) analysis module 321, an LPC quantization module 322, an LPC dequantization module 323, an inverse filter 324, an MDCT (modified discrete cosine transform) module 325, 326 module spectrum coding and multiplexing module 327. The spectrum encoding module 326 includes a shape quantization module 111 and a gain quantization module 112, and a shape quantization module 111 includes a zone search module 121 and a full search module 122.

LPC analysis module 321 performs linear prediction analysis using an M signal received as input from sum and difference calculation module 101 via switch 301, and provides and outputs LPC parameters (i.e., linear prediction parameters) indicating the structure spectrum of the M-signal into the module 322 LPC quantization.

LPC quantization module 322 converts linear prediction parameters received as input from LPC analysis module 321 into parameters with good complementarity, such as LSP (pairs of spectral lines or pairs of spectral lines) and ISP (pairs of spectral immitances), and quantizes the transformed parameters through a quantization method such as VQ (vector quantization), predictive VQ, multi-stage VQ and separate VQ. LPC quantization module 322 outputs LPC quantized data obtained by quantization to LPC quantization module 323 and multiplexing module 327.

LPC quantization module 323 de-quantizes the LPC quantized data received as input from the LPC quantization module 322 and further inverts the resulting parameters, such as LSP and ISP, into LPC parameters.

The inverse filter 324 applies inverse filtering to the M signal received as input from the sum and difference calculation unit 101 through the switch 301 using the LPC parameters received as input from the LPC dequantization module 323 and outputs to the MDCT module 325 filtered M-signal in which the spectrum-specific structure is removed and changed to a flat shape. Here, the function of the inverse filter 324 is represented by the following equation 3.

[one]

… (уравнение 3)

... (equation 3)

In equation 3, the subscript i represents the sample number of each signal, x _i represents the input of the inverse filter 324, and y _i represents the output of the inverse filter 324. In addition, α _i represents the LPC parameters quantized and dequantized in the LPC quantization module 322 and LPC dequantization module 323, and J represents the linear prediction order.

The MDCT module 325 performs the MDCT of the back-filtered M-signal received as input from the inverse filter 324 and converts the M-signal of the time domain into the spectrum of the M-signal of the frequency domain. In addition, FFT (Fast Fourier Transform) can equally be used instead of MDCT. The MDCT module 325 outputs the spectrum of the M signal obtained by the MDCT to the spectrum encoding module 326.

The spectrum coding unit 326 receives the M-signal spectrum as input from the MDCT module 325, quantizes the spectral shape and gain of the input spectrum separately, and outputs the resulting pulse code and gain code to the multiplexing unit 327. The shape quantization module 111 quantizes the shape of the input spectrum in the positions and polarities of a small number of pulses, and the gain quantization module 112 calculates and quantizes the amplifications of the pulses that are searched in the shape quantization module 111 based on each frequency band. The spectrum coding unit 326 outputs a pulse code indicating the positions and polarities of the desired pulses and a gain code representing the amplification of the desired pulses to the multiplexing unit 327. In addition, the shape quantization module 111 and the gain quantization module 112 are explained in more detail below.

The multiplexing unit 327 provides monophonic encoded information by multiplexing the LPC quantized data received as input from the LPC quantization unit 322, and a pulse code and a gain code received as input from the spectrum encoding unit 326, and outputs the monophonic encoded information to the module 303 mono decoding and switch 311.

Next, a shape quantization module 111 and a gain quantization module 112 are explained in more detail. The shape quantization module 111 includes a zone search module 121 that searches for pulses in each of a plurality of frequency bands into which a predetermined search zone is divided, and a full search module 122 that searches for pulses throughout the search zone.

The following equation 4 provides the starting point of the search. Here, in equation 4, E represents the coding distortion, s _i represents the input spectrum, g represents the optimal gain, δ is the delta function, and p is the pulse position.

[2]

… (уравнение 4)

... (equation 4)

From Equation 4 above, the position of the momentum in order to minimize the cost function is the position in which the absolute value | sp | the input spectrum in each frequency band is maximum, and the polarity has the value of the input spectrum in this position of the pulse.

The following illustrates an exemplary case in which the input spectrum vector is eighty samples long, the number of frequency bands is five, and the spectrum is encoded using a total of eight pulses consisting of a single pulse per frequency band and three pulses in the entire zone. In this case, the length of each frequency band is sixteen samples. Additionally, the amplitude of the pulses for which to search is fixedly set to "1", and their polarity is "+" or "-".

The zonal search module 121 searches for the position of the maximum energy and its polarity (+/-) in each frequency band and allows one pulse to appear per frequency band. In this example, the number of frequency bands is five, and each frequency band requires four bits to indicate the position of the pulse (record positions: 16), and one bit to indicate polarity (+/-), requiring a total of 25 information bits.

The sequence of operations of the search algorithm module 121 zone search shown in Fig.6. Here, the symbols used in the flowchart of FIG. 6 mean the following:

i: position

b: frequency band number

max: maximum value

c: counter

pos [b]: search result (position)

pol [b]: search result (polarity)

s [i]: input spectrum

As shown in FIG. 6, the zonal search module 121 calculates an input spectrum s [i] of each sample (0≤c≤15) per frequency band (0≤b≤4) and calculates a maximum value of "max".

7 shows an example of a spectrum represented by pulses that are searched in the area search module 121. As shown in FIG. 7, one pulse having an amplitude of “1” and a polarity of “+” or “-” is located in each of the five frequency bands, each of which has a bandwidth of sixteen samples.

The full search module 122 searches for positions to accommodate three pulses throughout the search area and encodes the positions of the pulses and their polarity. In the full search module 122, a search is performed according to the following four conditions for encoding exact positions with a small number of information bits and a small amount of computation.

(1) Two or more pulses are not placed in one position. In this example, the pulses are not located at positions in which the pulse of each frequency band is located in the area search module 121. With the present invention, information bits are not used to represent amplitude components, so that information bits can be effectively used.

(2) The search for pulses is performed in order, one after another, in an open loop. During the search, according to rule (1), certain pulse positions cannot be searched.

(3) When searching for a position, the position in which the impulse should be less preferably placed is also encoded as position information alone.

(4) Given that the gain is encoded based on each frequency band, the pulse search is performed by coding distortion estimation relative to the ideal gain of each frequency band.

The full search module 122 performs the following two-step cost estimate in order to search for a single pulse across the entire input spectrum. First, in the first step, the full search module 122 estimates the costs in each frequency band and finds the position and polarity so as to minimize the cost function. Then, in the second stage, the full search module 122 estimates the total costs each time the above search is completed in the frequency band, and saves the position and polarity of the pulse in order to minimize costs as an end result. This search is performed per frequency band, in order. Additionally, this search is performed in order to satisfy the above conditions (1) to (4). Then, when the search for one pulse is completed, provided that this pulse is in the desired position, the search for the next pulse is performed. This search is performed until a predetermined number of pulses (three pulses in this example) is detected by repeating the above processing.

The flow of the search algorithm in the full search module 122 is shown in FIG.

FIG. 8 is a flowchart of a search preprocessing method, and FIG. 9 is a flowchart of a search method. Additionally, parts corresponding to the above conditions (1), (2) and (4) are shown in the flowchart of FIG. 9.

The symbols used in the flowchart of FIG. 8 mean the following:

c: counter

pf [*]: flag presence / absence of an impulse

b: frequency band number

pos [*]: search result (position)

n_s [*]: correlation value

n_max [*]: maximum correlation value

n2_s [*]: correlation squared value

n2_max [*]: maximum value of the correlation square

d_s [*]: power value

d_max [*]: maximum power value

s [*]: input spectrum

The symbols used in the flowchart of FIG. 9 mean the following:

i: pulse number

i0: pulse position

cmax: maximum value of the cost function

pf [*]: flag of presence / absence of an impulse (0: absence, 1: presence)

ii0: relative position of the pulse in the frequency band

nom: spectral amplitude

nom2: numerator term (spectral power)

den: denominator

n_s [*]: relative value

d_s [*]: power value

s [*]: input spectrum

n2_s [*]: correlation squared value

n_max [*]: maximum correlation value

n2_max [*]: maximum value of the correlation square

idx_max [*]: search result of each impulse (position) (here, idx_max [*] from 0 to 4 is equivalent to pos [b] in Fig.6)

fd0, fd1, fd2: buffer for temporary storage (type of real numbers)

id0, id1: buffer for temporary storage (integer type)

id0_s, id1_s: buffer for temporary storage (integer type)

>>: bit shift (right)

&: "and" as a bit sequence

Here, when searching in FIG. 8 and FIG. 9, the case in which idx_max [*] is “-1” corresponds to the case of the above condition (3), when the pulse is less preferably to be placed. A concrete example of this is when the spectrum is sufficiently approximated only by pulses for which a search is performed per frequency band, and pulses for which a search is performed over the entire zone, and when the additional addition of pulses with an identical value proportionally increases the coding distortion.

The polarities of the desired pulses correspond to the polarities of the input spectrum at these positions, and the full search module 122 encodes these polarities with 3 (pulses) × 1 = 3 bits. Here, when the position is "-1", i.e. when the pulse should not be placed, any polarity can be used. However, polarity can be used to detect a bit error, and, in general, is set fixed to "+" or "-".

Additionally, the full search module 122 encodes the pulse position information based on the number of pulse position combinations. In this example, since the input spectrum contains eighty samples and five pulses are already detected in five separate frequency bands, if cases in which pulses are not placed are also taken into account, position variations can be represented using seventeen bits by calculating the following equation 5.

[3]

75 + 1C3 = (75 + 1) * (74 + 1) * (73 + 1) / 3/2/1

= 70300

<131072

= 2 ^ 17 ... (equation 5)

Here, according to the rule prohibiting the placement of two or more pulses in one position, the number of combinations can be reduced, so that the effect of this rule becomes large when the number of pulses for which a full search is performed increases.

A method for encoding pulse positions for which a search is performed in the full search module 122 is described in detail below.

(1) Three pulse positions are sorted based on their magnitude and arranged in order from the smallest numerical value to the largest numerical value. Here, “-1” is left as it is.

(2) The pulse numbers are aligned to the left boundary by the number of pulses occurring in the individual frequency bands in order to reduce the numerical values of the pulse positions. The numerical values calculated in this way are referred to “as item numbers”. Here, “-1” is left as it is. For example, when referring to the position of the pulse "66", when one pulse is provided between 0 and 15, between 16 and 31, between 32 and 47 and between 48 and 64, the position number changes to "66-4 = 62".

(3) “-1” is set equal to the position number represented by “maximum pulse value + 1”. In this case, the order of values is adjusted and determined so that the specified position number is not confused with the position number in which the pulse is actually present. By this, the pulse number for pulse # 0 is limited to the range 0-73, the position number for pulse # 1 is limited to the range between the position number for pulse # 0-74, and the position number for pulse # 2 is limited to the range between the position number for pulse # 1-75 , i.e. the position number of the smaller pulse is configured not to exceed the position number of the larger pulse.

(4) Then, according to the integration processing shown in the following equation 6, to calculate the combined code, the position numbers (i0, i1, i2) are integrated to form the code (c). This integration processing is referred to as integration calculation processing of all combinations in the case where there is an order of magnitude.

[four]

c = ((76-0) * (77-0) * (153-2 * 0) / 3 + (74-0) * (75-0)) / 4

- ((76-i0) * (77-i0) * (153-2 * i0) / 3 + (74-i0) * (75-i0)) / 4;

c = c + (76-i0) * (77-i0) / 2- (76-i1) * (77-i1) / 2;

c = c + 75-i2; ... (equation 6)

(5) Then, by combining seventeen bits of this c and three bits for polarity, a code of twenty bits is generated.

Here, in the above position numbers, pulse # 0 at “73”, pulse # 1 at “74” and pulse # 2 at “75” are position numbers at which the pulses are not placed. For example, if there are three position numbers (73, -1, -1) according to the above relationship between position number one and the position number in which the impulse is not placed, these position numbers are reordered as (-1, 73, -1) and become the following (73, 73, 74).

Thus, using the model, in order to represent the input spectrum by means of a sequence of eight pulses (five pulses in separate frequency bands and three pulses in the entire zone), as shown in this example, encoding can be performed using 45 information bits.

10 illustrates an example of a spectrum represented by pulses that are searched in the area search module 121 and the full search module 122. In addition, in FIG. 10, pulses represented by bold lines are pulses that are searched in the full search module 122.

Gain quantization module 112 quantizes the gain of each frequency band. Eight pulses are arranged in frequency bands, and gain quantization module 112 calculates gains by analyzing the correlation between these pulses and the input spectrum.

If the gain quantization module 112 calculates ideal amplifications and then encodes using scalar or vector quantization, firstly, the gain quantization module 112 calculates ideal amplifications according to the following equation 7. Here, in equation 7, gn is the ideal gain of the frequency band n , s (i + 16n) is the input spectrum of the frequency band n, vn (i) is the vector detected by decoding the shape of the frequency band n.

[5]

… (уравнение 7)

... (equation 7)

Additionally, gain quantization module 112 performs encoding by jointly performing scalar quantization (SQ) of ideal amplifications or jointly performing vector quantization of these five amplifications. In the case of vector quantization, efficient coding can be performed by means of predictive quantization, multi-stage VQ, separate VQ, etc. Here, the gain can be perceptually audited based on a logarithmic scale, and therefore, by performing SQ or VQ after performing the logarithmic gain conversion, a perceptually good synthesizing sound can be provided.

Additionally, instead of calculating ideal gains, a method for directly assessing coding distortion is provided. For example, if VQ is performed for five amplifications, the coding distortion is calculated so as to minimize the following equation 8. Here, in equation 8, E _k is the distortion of the kth gain vector, s (i + 16n) is the input spectrum of the frequency band n, g (k) n is the nth element of the kth gain vector, and v ⁿ (i) is the shape vector detected by decoding the shape of the frequency band n.

[6]

… (уравнение 8)

... (equation 8)

11 is a block diagram showing the main components in the monophonic decoding unit 303. The monaural decoding module 303 shown in FIG. 11 includes a demultiplexing module 331, an LPC dequantization module 332, a spectrum decoding module 333, an IMDCT (inverse modified discrete cosine transform) module 334, and a synthesis filter 335.

11, the demultiplexing unit 331 demultiplexes the monophonic encoded information received as input from the monophonic encoding unit 302 into LPC quantized data, a pulse code, and a gain code, outputs the LPC quantized data to the LPC quantization module 332 and outputs a pulse code and gain code in the spectrum decoding module 333.

LPC quantization module 332 dequantizes the LPC quantized data received as input from demultiplexing module 331 and outputs the resulting LPC parameters to synthesis filter 335.

The spectrum decoding unit 333 decodes the shape vector and decoding gain by a method supporting the encoding method in the spectrum encoding unit 326 of FIG. 5 using a pulse code and a gain code received as input from the demultiplexing unit 331. Additionally, the spectrum decoding module 333 provides a decoded spectrum by multiplying the decoded shape vector by decoding gain and outputs this decoded spectrum to the IMDCT module 334.

The IMDCT module 334 converts the decoded spectrum received as input from the spectrum decoding module 333 in the opposite way with respect to the conversion in the MDCT module 325 shown in FIG. 5, and outputs the time series M signal detected by the conversion to a synthesis filter 335 .

Synthesizing filter 335 provides a monophonic decoded M signal by applying a synthesizing filter to the time series M signal received as input from IMDCT module 334 using LPC parameters received as input from LPC dequantization module 332.

Next, a method for decoding three pulses in a spectrum decoding unit 333 for which a full search is performed is explained.

In the full search module 122 from the spectrum encoding module 326, the position numbers (i0, i1, i2) are integrated into a single code using the above equation 5. In the spectrum decoding module 333, the opposite processing is performed. Thus, the spectrum decoding module 333 sequentially calculates the value of the integration equation when each position number changes, fixes the position number when the position number is less than the integration value, and performs decoding by performing this processing from the lower layer position number to the higher order position number one after the other . 12 is a flowchart showing a decoding algorithm of a spectrum decoding unit 333.

Additionally, in FIG. 12, when the integrated position code “k” entered contains an error due to a bit error, the process proceeds to the error processing step. Therefore, in this case, the position must be detected by the predetermined error handling.

Additionally, since the decoder performs loop processing, the amount of computation in the decoder is greater than in the encoder. Here, each loop is an open loop, and therefore, compared to the total processing amount in the encoding device, the amount of computation in the decoder is not so large.

13 is a block diagram showing major components in a stereo coding unit 305. The stereo coding unit 305 shown in FIG. 13 has a substantially identical configuration and performs basically the same operations as the monaural coding unit 302 shown in FIG. 5. Therefore, with regard to modules that perform identical operations for FIG. 5 and FIG. 13, “a” is assigned to the module number references in FIG. 13. For example, the module in FIG. 13 corresponding to the LPC analysis module 321 in FIG. 5 is expressed as the LPC analysis module 32a1. In addition, the stereo encoding module 305 in FIG. 13 differs from the monaural encoding module 302 in FIG. 5 by the inclusion of an additional inverse filter 351, an MDCT module 352, and an integration module 353. In addition, the spectrum encoding unit 356 from the stereo encoding unit 305 in FIG. 13 is different from the spectrum encoding unit 326 from the monaural encoding unit 302 in FIG. 5 with input signals, and therefore is assigned another reference number.

The inverse filter 351 applies inverse filtering to the S signal received as input from the sum and difference calculation module 101, using LPC parameters received as input from the LPC dequantization module 323a to smooth the spectrum-specific structure, and outputs the filtered S-signal to MDCT module 352. Here, the function of the inverse filter 324a is represented by the above equation 3. Strictly speaking, although the LPC coefficients obtained from the M-signal do not coincide with the spectral structure of the S-signal, given that M -signal and S-signal, in general, have similar spectral structures and that the amount of computation and the amount of ROM required for LPC analysis, quantization and dequantization of the S-signal is saved, LPC parameters taken as input from the LPC- module 323a dequantization are used in reverse filtering processing in the inverse filter 351.

The MDCT module 352 performs the MDCT of the back-filtered S signal received as input from the inverse filter 351, and converts the S-signal of the time domain into the spectrum of the S-signal of the frequency domain. Here, instead of MDCT, FFT can equally be used. The MDCT module 352 outputs the spectrum of the S signal detected by the MDCT to the integration module 353.

The integration module 353 integrates the spectrum of the M signal received as input from the MDCT module 325a and the spectrum of the S signal received as input from the MDCT module 352 so that the spectra of the same frequency are adjacent to each other and outputs the resulting integrated spectrum to spectrum encoding module 356.

FIG. 14 illustrates a state where an M-signal spectrum and an S-signal spectrum are integrated in an integration unit 353. The spectrum encoding unit 356 uses the integrated spectrum detected by integrating the two spectra, as shown in FIG. 14, as one target encoding spectrum, and therefore allocates a larger number of bits to the important parts when encoding the spectrum of the M signal and the spectrum of the S signal.

Returning to FIG. 13, the spectrum encoding module 356 differs from the spectrum encoding module 326 by using the integrated spectrum received as input from the integration module 353 as an input spectrum. In addition, the spectrum encoding module 356 differs from the spectrum encoding module 326 in the number of pulses that are searched across the entire input spectrum.

In association with the number of pulses for which a full search is performed, bit allocation in the spectrum encoding unit 356 is explained with respect to FIG.

The spectrum coding unit 356 uses the integrated spectrum as an input spectrum, and therefore, the number of samples in the input spectrum is two times the number of samples in the input spectrum in the spectrum coding unit 326, and the number of samples in each of the five frequency bands detected by dividing the input spectrum is twice the number of samples in the spectrum coding unit 326. Considering that the total number of bits of the shape code is 45 bits in the monophonic coding unit 302, the spectrum coding unit 356 performs bit allocation as shown in FIG. As shown in FIG. 15, the number of pulses for which a full search is performed is “2” in the spectrum coding unit 356, which is different from the spectrum coding unit 326 in which the number of pulses for which a full search is performed is “3”.

In addition, as shown in FIG. 15, the number of bits to be used in spectrum coding is “46” in total in the spectrum coding unit 356, which is different from the spectrum coding unit 326 in which the number of bits to be used in spectrum coding is "45" in total.

Here, it is equally possible to fully compare the total number of bits that should be used when coding the spectrum in spectrum coding unit 356 with the total number of bits that should be used when coding the spectrum in spectrum coding unit 326. For example, the search range for one of two pulses for which a full search is performed in the spectrum coding unit 356 may be limited from 0-159 samples to 0-50 samples. By this means, 160 × 51 <8192 kinds of search results can be expressed by 13 bits, so that the total number of bits that should be used in spectrum coding can be suppressed beyond 45 bits. Alternatively, for example, after searching for an impulse per frequency band, by limiting the search range of the fifth frequency band (i.e., the highest frequency band) from 0-31 samples to 0-15 samples, the total number of bits can be compared equally, which should be used when coding the spectrum in the spectrum coding unit 356, with the total number of bits to be used when coding the spectrum in the spectrum coding unit 326. This is because in this case it is possible to represent the positions of the pulses of the frequency band in the five frequency bands by 5 × 4 + 4 = 24 bits.

If the spectrum encoding unit 356 encodes an integrated spectrum integrating the spectrum of the M signal and the spectrum of the S signal, bit allocation is automatically performed based on the characteristics of the M signal and the S signal, so that efficient encoding can be performed according to the significance of the information.

For example, if the L-signal and the R-signal are completely identical, the spectrum of the S-signal is "0", and the pulses are placed only in the positions of the spectrum of the M-signal in the integrated spectrum. Therefore, the spectrum of the M signal is encoded accurately.

In contrast, if the phase of the L-signal and the phase of the R-signal are approximately opposite, the spectrum of the S-signal becomes significant, and more pulses are placed in the positions of the spectrum of the S-signal in the integrated spectrum. Therefore, the spectrum of the S signal is encoded accurately. Thus, without special classification based on decisions or cases, bit allocation is performed automatically, and the spectrum of the M signal and the spectrum of the S signal are encoded efficiently.

In addition, if there are large elements at a certain frequency, and the phase of the L signal and the phase of the R signal are not approximately opposite, then one of the spectrum of the M signal and the spectrum of the S signal is likely to have large elements. Here, the M-signal spectrum and the S-signal spectrum of elements with the same frequency are integrated side by side in the integrated spectrum, and the integrated spectrum is divided into a plurality of frequency bands and encoded in the spectrum encoding unit 356, so that only one of the M-signal spectrum is searched and encoded and the spectrum of the S-signal frequency with significant elements. Through this, it is possible to prevent the coding of two pulses of the element with the same frequency and to implement effective coding.

Fig is a block diagram showing the main components in the module 306 stereo decoding. The stereo decoding module 306 comprises a demultiplexing module 331a, an LPC dequantization module 332a, a spectrum decoding module 333a, an IMDCT module 334a and a synthesis filter 335a that perform operations identical to those of the demultiplexing module 331, the LPC dequantization module 332, the spectrum decoding module 333, IMDCT module 334 and synthesizing filter 335 from monophonic decoding module 303 shown in FIG. 11. Further, the stereo decoding module 306 includes a decomposition module 361, an IMDCT module 362, and a synthesis filter 363. In addition, in FIG. 16, the output of the synthesis filter 335a is a stereo decoded M signal, and the output of the synthesis filter 363 is stereo decoded S- signal.

The decomposition unit 361 decomposes the decoded spectrum received as input from the spectrum decoding unit 333a into the decoded spectrum of the M signal and the decoded spectrum of the S signal by opposite processing with respect to the processing in the integration module 353 in FIG. 13. Additionally, decomposition module 361 outputs the decoded spectrum of the M signal to IMDCT module 334a and outputs the decoded spectrum of the S signal to IMDCT module 362.

The IMDCT module 362 converts the decoded spectrum of the S signal received as input from the decomposition module 361 in the opposite way to the MDCT module 352 shown in FIG. 13 and outputs the time series S signal detected by the conversion to a synthesis filter 363 .

Synthesizing filter 363 provides a stereo decoded S signal by applying a synthesizing filter to an S-signal of a time series received as input from IMDCT module 362 using LPC parameters received as input from LPC dequantization module 332a.

Next, the configuration and operations of the stereo signal decoding apparatus supporting the stereo signal encoding apparatus 100 shown in FIG. 1 are explained.

17 is a block diagram showing the main components of a stereo signal decoding apparatus 200 supporting a stereo signal encoding apparatus 100.

17, the stereo signal decoding apparatus 200 comprises a demultiplexing module 201, a mode setting module 202, a base layer decoding module 203, a first enhancement layer decoding module 204, a second enhancement layer decoding module 205, a third enhancement layer decoding module 206, and a sum calculation module 207 and difference.

The demultiplexing unit 201 demultiplexes the bit streams received as input from the stereo signal encoding apparatus 100 into mode information, encoded information of the base layer, encoded information of the first enhancement layer, encoded information of the second enhancement layer and encoded information of the third enhancement layer, and outputs them to the module 202 mode settings, base layer decoding module 203, first enhancement layer decoding module 204, second enhancement layer decoding module 205 I and decoding module 206 of the third enhancement layer, respectively.

The mode setting module 202 outputs mode information for setting the decoding modes in the base layer decoding module 203, the first enhancement layer decoding module 204, the second enhancement layer decoding module 205 and the third enhancement layer decoding module 206, received as input from the demultiplexing unit 201, to these decoding modules.

The decoding mode in each decoding unit is referred to as a monaural decoding mode for decoding only M-signal information or a stereo decoding mode for decoding both M-signal information and S-signal information. Here, the information of the M signal typically corresponds to the M signal itself or coding distortion associated with the M signal in each layer. In addition, S-signal information typically corresponds to the S-signal itself or coding distortion associated with the S-signal in each layer.

Further, the decoding mode in each layer is shown using each of the mode information bits. Thus, in bits, the value “0” represents the mono decoding mode, and the value “1” represents the stereo decoding mode. More specifically, for example, each of the four bits of mode information is used to sequentially represent decoding modes in the base layer decoding module 203, the first enhancement layer decoding module 204, the second enhancement layer decoding module 205, and the third enhancement layer decoding module 206. For example, the four-bit mode information “0000” means that monaural decoding is performed in all layers. In addition, for example, the “0011” mode information means that the base layer decoding module 203 and the first enhancement layer decoding module 204 perform monophonic decoding, and the second enhancement layer decoding module 205 and the third enhancement layer decoding module 206 perform stereo decoding. Thus, using the information of the four-bit mode, sixteen types of decoding modes in four decoding modules can be represented.

In the present embodiment, the mode information output from the mode setting unit 202 is received in each decoding unit as identical input information of the four-bit mode. Additionally, each decoding module checks only one bit of the four input bits required to set the decoding mode, and sets the decoding mode. Thus, in the four-bit mode input, the base layer decoding module 203 checks the first bit, the first enhancement layer decoding module 204 checks the second bit, the second enhancement layer decoding module 205 checks the third bit, and the third enhancement layer decoding module 206 checks the fourth bit.

However, instead of entering identical four-bit mode information in each decoding unit, the mode setting unit 202 can pre-sort one bit required to set the decoding mode in each decoding unit and output one bit to each decoding unit. Thus, in four bits of mode information, the mode setting module 202 can only enter the first bit in the base layer decoding module 203, only the second bit in the first enhancement layer decoding module 204, only the third bit in the second enhancement layer decoding module 205, and only the fourth bit in decoding module 206 of the third enhancement layer.

In addition, in any of the above cases, the mode information received as input from the demultiplexing unit 201 to the mode setting unit 202 is referred to as four-bit mode information.

In the base layer decoding unit 203, a monaural decoding mode or a stereo decoding mode is set based on the mode information received as input from the mode setting unit 202. More specifically, after setting the monophonic decoding mode, the base layer decoding module 203 decodes the monophonic encoded information received from the demultiplexing unit 201 as input encoded information of the base layer and outputs the resulting decoded M-signal of the base layer to the first enhancement layer decoding module 204. In this case, the S-signal information is not decoded, and therefore, the zero signal is apparently output to the first enhancement layer decoding unit 204 as a decoded S-signal of the base layer.

On the contrary, after setting the stereo decoding mode, the base layer decoding unit 203 decodes the stereo encoded information received from the demultiplexing unit 201 as input encoded information of the base layer and outputs the resulting decoded M-signal of the base layer and the decoded S-signal of the base layer to the decoding unit 204 of the first enhancement layer. Here, the base layer decoding unit 203 completely resets the M signal and the S signal (i.e., sets the values 0 for these signals) before decoding. In addition, the base layer decoding module 203 is explained in more detail below.

In the decoding module 204 of the first enhancement layer, the monaural encoding mode or the stereo encoding mode are set based on the mode information received as input from the mode setting unit 202. More specifically, after setting the monophonic decoding mode, the first enhancement layer decoding unit 204 decodes the monophonic encoded information received from the demultiplexing unit 201 as input encoded information of the first enhancement layer, and detects an encoding distortion of the base layer of the M signal. The first enhancement layer decoding unit 204 sums the distortion in encoding the base layer of the M signal and the decoded base layer M signal received as input from the base layer decoding unit 203, and outputs the summation result to the second enhancement layer decoding unit 205 as a decoded M signal first enhancement layer. The decoded S-signal of the base layer, taken as input from the base layer decoding unit 203, is output as is to the second enhancement layer decoding unit 205 as a decoded S-signal of the first enhancement layer.

On the contrary, after setting the stereo decoding mode, the first enhancement layer decoding unit 204 decodes the stereo encoded information received from the demultiplexing unit 201 as input encoded information of the first enhancement layer, and detects distortions in encoding the base layer of the M and S signals. The first enhancement layer decoding unit 204 sums the distortion in encoding the base layer of the M signal and the decoded base layer M signal received as input from the base layer decoding unit 203, and outputs the summation result to the second enhancement layer decoding unit 205 as a decoded M signal first enhancement layer. In addition, the decoding module 204 of the first enhancement layer summarizes the distortion when encoding the base layer of the S signal and the decoded S signal of the base layer, taken as input from the decoding module 203 of the base layer, and outputs the result of the summation to the decoding module 205 of the second enhancement layer as decoded S signal of the first enhancement layer. In addition, the first enhancement layer decoding unit 204 is explained in more detail below.

In the second enhancement layer decoding unit 205, the monaural encoding mode or the stereo encoding mode are set based on the mode information received as input from the mode setting unit 202. More specifically, after setting the monophonic decoding mode, the second enhancement layer decoding unit 205 decodes the monophonic encoded information received from the demultiplexing unit 201 as input encoded information of the second enhancement layer, and detects an encoding distortion of the first enhancement layer associated with the M signal. The second enhancement layer decoding unit 205 summarizes the coding distortion of the first enhancement layer associated with the M signal and the decoded M-signal of the first enhancement layer, taken as input from the decoding unit 204 of the first enhancement layer, and outputs the summation result to the third decoding unit 206 enhancement layer as a decoded M signal of the second enhancement layer. The decoded S signal of the first enhancement layer, received as input from the decoding module 204 of the first enhancement layer, is output as it is to the decoder module 206 of the third enhancement layer as the decoded S signal of the second enhancement layer.

In contrast, after setting the stereo decoding mode, the second enhancement layer decoding unit 205 decodes the stereo encoded information received from the demultiplexing unit 201 as input encoded information of the second enhancement layer, and detects distortions in the encoding of the first enhancement layer associated with the M and S signals. The second enhancement layer decoding unit 205 summarizes the coding distortion of the first enhancement layer associated with the M signal and the decoded M-signal of the first enhancement layer, taken as input from the decoding unit 204 of the first enhancement layer, and outputs the summation result to the third decoding unit 206 enhancement layer as a decoded M signal of the second enhancement layer. In addition, the second enhancement layer decoding unit 205 summarizes the encoding distortion of the first enhancement layer associated with the S signal and the decoded S signal of the first enhancement layer received as input from the first enhancement layer decoding unit 204, and outputs the summing result to the module 206 decoding the third enhancement layer as a decoded S signal of the second enhancement layer. In addition, the decoding module 205 of the second enhancement layer is explained in more detail below.

In the third enhancement layer decoding unit 206, a monaural encoding mode or a stereo encoding mode are set based on the mode information received as input from the mode setting unit 202. More specifically, after setting the monophonic decoding mode, the third enhancement layer decoding unit 206 decodes the monophonic encoded information received from the demultiplexing unit 201 as input encoded information of the third enhancement layer, and detects an encoding distortion of the second enhancement layer associated with the M signal. The third enhancement layer decoding unit 206 sums the second enhancement layer encoding distortion associated with the M signal and the second enhancement layer decoded M signal received as input from the second enhancement layer decoding unit 205, and outputs the summation result to the sum calculation unit 207 and differences as a decoded M signal of the third enhancement layer. The decoded S signal of the second enhancement layer, received as input from the decoding module 205 of the second enhancement layer, is output as is to the sum and difference calculation unit 207 as the decoded S signal of the third enhancement layer.

On the contrary, after setting the stereo decoding mode, the third enhancement layer decoding unit 206 decodes the stereo encoded information received from the demultiplexing unit 201 as input encoded information of the third enhancement layer and detects distortions in the encoding of the second enhancement layer associated with the M and S signals. The third enhancement layer decoding unit 206 sums the second enhancement layer encoding distortion associated with the M signal and the second enhancement layer decoded M signal received as input from the second enhancement layer decoding unit 205, and outputs the summation result to the sum calculation unit 207 and differences as a decoded M signal of the third enhancement layer. In addition, the third enhancement layer decoding unit 206 summarizes the second enhancement layer encoding distortion associated with the S signal and the second enhancement layer decoded S signal received as input from the second enhancement layer decoding unit 205, and outputs the summation result to the module 207 calculating the sum and difference as the decoded S signal of the third enhancement layer. In addition, the third enhancement layer decoding unit 206 is explained in more detail below.

The sum and difference calculation unit 207 calculates the decoded L signal and the decoded R signal according to the following equations 9 and 10 using the decoded M signal of the third enhancement layer and the decoded S signal of the third enhancement layer, taken as input from the third enhancement decoding unit 206 layer.

Li '= (Mi' + Si ') / 2 ... (equation 9)

Ri '= (Mi'-Si') / 2 ... (equation 10)

In equations 9 and 10, Mi 'represents the decoded M signal of the third enhancement layer, si' represents the decoded S signal of the third enhancement layer, Li 'represents the decoded L signal, and Ri' represents the decoded R signal.

Fig is a block diagram showing the main components in the module 203 decoding the base layer.

The base layer decoding module 203 shown in FIG. 18 comprises a switch 231, a monaural decoding module 232, a stereo decoding module 233, a switch 234, and a switch 235.

If the value of the first bit of the mode information received as input from the mode setting module 202 is “0”, the switch 231 outputs the monophonic encoded information received from the demultiplexing unit 201 as input encoded information of the base layer to the monophonic decoding module 232, and if the value the first bit of mode information received as input from mode setting module 202 is “1”, outputs stereo encoded information received from demultiplexing module 201 as input th coded information to base layer 233 in the stereo decoding unit.

Monophonic decoding module 232 performs monophonic decoding using monophonic encoded information received as input from switch 231, and outputs the resulting decoded base layer M signal to switch 234. In addition, the configuration and operations in monophonic decoding module 232 are identical to the configuration and operations in the monophonic decoding unit 303 shown in FIG. 11, and therefore, their specific explanation is omitted.

The stereo decoding unit 233 performs stereo decoding using stereo encoded information received as input from the switch 231, outputs the resulting decoded M-signal of the base layer and the decoded S-signal of the base layer to the switch 234 and the switch 235, respectively. In addition, the configuration and operations in the stereo decoding unit 233 are identical to the configurations and operations in the stereo decoding unit 306 shown in FIG. 16, and therefore, their specific explanation is omitted.

If the value of the first bit of the mode information received as input from the mode setting unit 202 is “0”, the switch 234 outputs the decoded base layer M signal received as input from the monaural decoding unit 232 to the first enhancement layer decoding unit 204. If the value of the first bit of the mode information received as input from the mode setting unit 202 is “1”, the switch 234 outputs the decoded base layer M signal received as input from the stereo decoding unit 233 to the decoding unit 204 of the first enhancement layer.

If the value of the first bit of the mode information received as input from the mode setting module 202 is “0”, the switch 235 is turned off and does not output a signal. Here, as equivalent processing, in fact, a signal from all zero values (i.e., a zero signal) is output to the first enhancement layer decoding unit 204 as a decoded S-signal of the base layer. If the value of the first bit of the mode information received as input from the mode setting unit 202 is “1”, the received decoded S-signal of the base layer as input from the stereo decoding unit 233 is output to the decoding unit 204 of the first enhancement layer.

19 is a block diagram showing the main components in the decoding module 205 of the second enhancement layer. Here, the first enhancement layer decoding module 204, the second enhancement layer decoding module 205, and the third enhancement layer decoding module 206 shown in FIG. 17 have the same internal configuration and operations, but differ in input and output signals. Therefore, an exemplary case is illustrated with only the decoding module 205 of the second enhancement layer.

19, the decoding module 205 of the second enhancement layer comprises a switch 251, a mono decoding module 252, a stereo decoding module 253, a switch 254, an adder 255, a switch 256, and an adder 257.

If the value of the third bit of the mode information received as input from the mode setting unit 202 is “0”, the switch 251 outputs the monophonic encoded information received from the demultiplexing unit 201 as input encoded information of the second enhancement layer to the monaural decoding unit 252. In addition, if the value of the third bit of the mode information received as input from the mode setting unit 202 is “1”, the switch 251 outputs the stereo encoded information received from the demultiplexing unit 201 as input encoded information of the second enhancement layer to the stereo decoding unit 253 .

Monophonic decoding module 252 performs monophonic decoding using monophonic encoded information received as input from switch 251, and outputs the resulting distortion when encoding the first enhancement layer associated with the M signal to switch 254. In addition, the configuration and operations in module 252 the monophonic decoding shown in FIG. 11 are identical to the configuration and operations of the monophonic decoding unit 303, and therefore, their specific explanation is omitted .

The stereo decoding unit 253 performs stereo decoding using the stereo encoded information received as input from the switch 251, and outputs the resulting distortion when encoding the first enhancement layer associated with the M signal and distortion when encoding the first enhancement layer associated with the S signal , into switch 254 and switch 257, respectively. In addition, the configuration and operations in the stereo decoding unit 253 are identical to the configuration and operations in the stereo decoding unit 306 shown in FIG. 16, and therefore, their specific explanation is omitted.

If the value of the third bit of the mode information received as input from the mode setting module 202 is “0”, the switch 254 outputs a distortion when encoding the first enhancement layer associated with the M signal received as input from the monaural decoding unit 252 to the adder 255. In addition, if the value of the third bit of the mode information received as input from the mode setting module 202 is “1”, the switch 254 outputs distortion when encoding the first enhancement layer associated with the M-signal received as stve input from stereo decoding module 253, an adder 255.

An adder 255 summarizes the distortion in encoding the first enhancement layer associated with the M signal received as input from the switch 254, and the decoded M signal of the first enhancement layer received as input from the decoding unit 204 of the first enhancement layer, and outputs the summation result to a third enhancement layer decoding unit 206 as a decoded M signal of the second enhancement layer.

An adder 257 summarizes the distortion in encoding the first enhancement layer associated with the S signal received as input from the stereo decoding unit 253 and the decoded S signal of the first enhancement layer received as input from the decoding unit 204 of the first enhancement layer, and outputs the result into switch 256.

If the value of the second bit of the mode information received as input from the mode setting module 202 is “0”, the switch 256 outputs a decoded S-signal of the first enhancement layer, which is received as input from the decoding module 204 of the first enhancement layer, as is in the module 206 decoding the third enhancement layer. In addition, if the value of the second bit of the mode information received as input from the mode setting unit 202 is “1”, the switch 256 outputs the summation result received as input from the adder 257 to the third enhancement layer decoding unit 206 as decoded S- signal of the second enhancement layer.

Thus, according to the present embodiment, scalable encoding is performed for the mono signal (i.e., the M signal) and the side signal (i.e., the S signal) calculated from the L signal and the R signal of the stereo signal, so that scalable coding can be performed using the correlation between the L signal and the R signal. Further, according to the present embodiment, the encoding mode in each layer in scalable encoding is set based on the mode information, so that a layer for performing monaural encoding and a layer for performing stereo encoding can be defined and the degree of freedom in controlling the encoding accuracy can be increased.

Furthermore, according to the present embodiment, the M-signal spectrum and the S-signal spectrum are integrated and encoded so that the spectra of the same frequency are adjacent to each other, so that it is possible to perform automatic bit allocation without special classification based on decisions or cases in stereo encoding and performing efficient encoding according to the significance of the L-signal and R-signal information.

Second Embodiment

FIG. 20 is a block diagram showing the main components of a stereo signal encoding apparatus 110 according to Embodiment 2 of the present invention. The stereo signal encoding device 110 shown in FIG. 20 has a substantially identical configuration and performs substantially identical operations to the stereo signal encoding device 100 shown in FIG. Therefore, with regard to modules that perform identical operations for FIG. 1 and FIG. 20, “a” is assigned to the module numbers in FIG. 20. For example, the module in FIG. 20 corresponding to the sum and difference calculation module 101 in FIG. 1 is expressed as the sum and difference calculation module 101a. In addition, the stereo signal encoding device 110 in FIG. 20 is different from the stereo signal encoding device 100 in FIG. 1 by the inclusion of additional mode setting modules 112-114. In addition, the mode setting module 111 of the stereo signal encoding device 110 in FIG. 20 is different from the mode setting module 102 of the stereo signal encoding device 100 in FIG. 1 by input signals, and therefore, another reference number is assigned to it. Here, the mode setting modules 111-114 shown in FIG. 20 have identical internal configuration and operations, but differ in input signals and output signals. Therefore, an exemplary case is illustrated with only the mode setting module 111.

The mode setting module 111 calculates the power of the M-signal and the S-signal received as input from the sum and difference calculation module 101a, and, based on the calculated power and predetermined conditional equations, sets the monophonic coding mode for encoding only the M-signal information or the stereo mode encoding for encoding both the M-signal information and the S-signal information. For example, the stereo coding mode is set if the power of the S signal exceeds the power of the M signal, or the monaural coding mode is set if the power of the S signal is lower than the power of the M signal. In addition, if the power of the M-signal and the power of the S-signal is low, the monaural encoding mode is set. This allows for the fact that when encoders are developed, a stereo encoder that processes two types of signals provides a higher bit rate than a monophonic encoder that processes one type of signal. In addition, information about a predetermined mode is output to the base layer encoding unit 103a and the multiplexing unit 107a.

The power calculation in the mode setting module 111 is performed according to the following equations 11 and 12.

[7]

… (уравнение 11)

... (equation 11)

… (уравнение 12)

... (equation 12)

In equations 11 and 12, i represents the sample number, PowM represents the power of the M signal, and M _i represents the M signal. In addition, PowS represents the power of the S signal, and s _i represents the S signal.

A predetermined conditional equation in the mode setting module 111 is shown in the following equation 13.

[8]

In equation 13, α represents a constant for estimating the total power and can fit the upper limit value of the power of the signal, which is not perceived. In addition, β represents a constant for estimating the power of the S signal. A method for calculating the constant β for estimating the power of an S signal is described below. In addition, m represents a mode. Here, for example, the constant α for estimating the total power and the constant β for estimating the power of the S-signal are stored in ROM.

As for the constant β for estimating the power of the S-signal, if the signal of less distortion during encoding is selected from the L-signal and the R-signal, a method for calculating and storing corresponding β in the mode setting modules 111-114 is possible. A specific method for calculating the constant for estimating the power of the S-signal β is explained below.

Here, a method for calculating the constant β for estimating the power of the S signal in the mode setting unit 111 is explained. First, a large amount of stereo speech data is received as input in the learning mode setting unit 111, and the relationship between the power of the M signal and the power of the S signal is calculated according to the following equation 14.

[9]

… (уравнение 14)

... (equation 14)

In equation 14, i represents the sample number of each signal, and j represents the number of training stereo speech data. In addition, M _i represents the M signal, and s _i represents the S signal. In addition, PowM _j represents the power of the M-signal of the J-x training stereo speech data, and PowS _j represents the power of the S-signal of the J-x training stereo speech data.

Then, the opposite processing with respect to the downmix is performed for the decoded M signal and the decoded S signal detected by two-mode encoding and decoding in the base layer encoding unit 103a to find the decoded L signal and the decoded R signal. The sums of the signal-to-noise ratios of the resulting decoded L-signal and the decoded R-signal are calculated (i.e., the signal-to-noise ratios in the case where the distortion in the coding of the L-signal and R-signal, taken as input to the device 110 encoding stereo signals, is regarded as noise), i.e. E0j and E1j.

Then, by gradually changing the value of β between 0 and 1.0, the total signal-to-noise ratio E _β shown in the following equation 15 is calculated.

[10]

The value of β in order to maximize the above E _β is calculated. This value is stored in the mode setting module 111 and is used as a constant for estimating the power of the S-signal β. Similar to the mode setting module 111, the mode setting modules 112-114 calculate and store a constant for estimating the power of the S signal β.

In addition, the stereo signal decoding apparatus according to Embodiment 2 of the present invention has a configuration identical to that of FIG. 17 of Embodiment 1, and therefore, the explanation is omitted.

Thus, according to the present embodiment, as encoding processing in each layer proceeds, the encoding mode in each layer for scalable encoding is set based on local speech features, so that a layer for performing monophonic encoding and a layer for performing stereo encoding can be automatically set. and provide high quality decoded signals. In addition, if the bit rate varies between modes, the transmission rate is controlled automatically, so that the number of information bits can be saved.

Embodiments of the present invention are described above.

In addition, although the embodiments described above describe cases in which stereo signals are mainly used as speech signals, it is understood that stereo signals can be used as audio signals.

Furthermore, although exemplary cases have been described in the embodiments above, in which the integration module 353 integrates the M-signal spectrum and the S-signal spectrum so that the spectra of the same frequency are adjacent to each other, the present invention is not limited to this, and it is equally possible integrate these spectra into the integration module 353 so that the spectrum of the S signal is easily placed adjacent to before or after the spectrum of the M signal.

In addition, although the embodiments described above describe cases in which two types of stereo signals are represented by the names “left channel signal” and “right channel signal”, more general names such as “first channel signal” and "second channel signal". In addition, the association between the bit values “0” and “1” and the encoding modes “monaural encoding mode” and “stereo encoding mode” is not limited.

In addition, although the embodiments described above exemplary cases in which the present invention is applied to technical requirements in which the sampling frequency is 16 kHz and the frame length is 20 ms, the present invention is not limited to this, and the present invention can be equally applied. other technical requirements in which the sampling frequency is 8 kHz, 24 kHz, 32 kHz, 44.1 kHz, 48 kHz, etc., and the frame length is 10 ms, 30 ms, 40 ms, etc. The present invention is independent of sample rate or frame length.

In addition, although the embodiments described above describe cases in which the four-layer configuration is used in scalable coding, the present invention is not limited to this, and other numbers of layers other than four can equally be used. The present invention is independent of the number of layers.

In addition, although the embodiments described above exemplary cases in which pulse coding is used to encode the spectrum of the excitation signal, the present invention is not limited to this, and in order to encode the spectrum of the excitation signal, predicting VQ can equally be used VQ, split VQ, multi-stage VQ, bandwidth extension technologies, inter-channel predictive decoding, etc. The present invention is independent of spectrum coding schemes.

In addition, although exemplary cases have been described in the embodiments above in which stereo signals are encoded to transmit encoded information, the present invention is not limited to this, and encoded information can equally be stored on a storage medium. For example, although encoded information of audio signals is often stored in a storage device or on a disk and used, the present invention is equally effective in this case. The present invention does not depend on whether encoded information is transmitted or stored.

In addition, although exemplary cases have been described in the embodiments above in which a stereo signal is generated using two channels, the present invention is not limited thereto, and a stereo signal can be generated equally using a plurality of channels, for example, 5.1 channels.

In addition, although the embodiments described above describe cases in which encoding is performed using only the size of the spectra of the M signal and S signal as a measure of distance, the present invention is not limited thereto, and encoding using a phase difference or energy ratio between the M-signal and the S-signal as a measure of distance. The present invention is independent of the distance measure that should be used when coding the spectrum.

In addition, although the embodiments described above describe cases in which a stereo signal decoding apparatus receives and processes bit streams transmitted from a stereo signal encoding apparatus, the present invention is not limited thereto, and a stereo signal decoding apparatus can receive and process bit streams until while these bitstreams are transmitted from an encoding device that can generate bitstreams that can be processed in this decoding device vania.

In addition, the stereo signal encoding device and stereo signal decoding device according to the present invention can be installed on the communication terminal device and the base station device in the mobile communication system, so that it is possible to provide the communication terminal device, the base station device and the mobile communication system having functional effects identical to the above.

Although the above embodiments describe exemplary cases in which the present invention is implemented using hardware, the present invention can be implemented using software. For example, by describing the algorithm according to the present invention in a programming language — storing this program in a memory and instructing the information processing module to execute this program, it is possible to implement a function identical to that in the stereo signal encoding device according to the present invention.

Moreover, each function block used in the explanation of each of the above embodiments may typically be implemented as an LSI consisting of an integrated circuit. It can be separate microcircuits, or they can be partially or completely contained on one microcircuit.

The term LSI is used throughout this document, but it may also be referred to as IC, “system LSI”, “super-LSI” or “ultra-LSI”, depending on the varying degree of integration.

Moreover, the method of integrating microcircuits is not limited to LSI, and implementation using specialized circuits or a general-purpose processor is also possible. After manufacturing an LSI, the use of an FPGA (Field Programmable Gate Array) or reconfigurable processor where the connections or connectors of the circuit cells in LSI can be reconfigured is also possible.

In addition, if integrated circuit technology appears to replace LSI, as a result of improvements in semiconductor technology or other derivative technology, it is of course also possible to integrate function blocks using this technology. The use of biotechnology is also allowed.

Disclosure of the essence of patent application (Japan) No. 2008-72497, filed March 19, 2008, and patent application (Japan) No. 2008-274536, filed October 24, 2008, including a detailed description, drawings and abstract, are fully contained in this document by reference.

Industrial applicability

The present invention is suitable for use, for example, in an encoding device that encodes speech signals and audio signals, and in a decoding device that decodes encoded signals.

Claims (13)

1. A device for encoding stereo signals, comprising:
- a module for calculating the sum and difference, which generates a monophonic signal associated with the sum of the signal of the first channel and the signal of the second channel forming the stereo signal, and generates a side signal associated with the difference between the signal of the first channel and the signal of the second channel;
a mode information generating unit that generates mode information per layer indicating a coding mode of one of monophonic coding and stereo coding; and
coding modules of layers 1 through N, which perform monophonic coding in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) using information related to the monophonic signal, or perform stereo coding in the i-th layer using both information related to the monaural signal and information related to the side signal based on the mode information and provide encoded information of the i-th layer. 2. The stereo coding device according to claim 1, in which:
a mode information generating unit generates mode information from N bits indicating an encoding mode using each of the bits; and
the encoding module of the i-th layer performs monophonic encoding in the i-th layer or performs stereo coding in the i-th layer based on the value of the i-th bit of mode information. 3. The stereo signal encoding device according to claim 2, wherein the first layer encoding module comprises:
a monophonic coding module of the first layer, which, when the value of the first bit of mode information indicates monophonic coding, performs monophonic coding in the first layer using a monophonic signal and outputs coding distortion associated with the monophonic signal in the first layer and a side signal to the coding module of the second layer ; and
a stereo coding module of the first layer, which, when the value of the first bit of mode information indicates stereo coding, performs stereo coding in the first layer using both a monaural signal and a side signal and outputs coding distortion associated with the monaural signal in the first layer and distortion when encoding, associated with a side signal in the first layer, to the encoding module of the second layer. 4. The encoding device for stereo signals according to claim 3, in which the encoding module of the nth (n = 2, 3, ..., N-1) layer contains:
a monophonic encoding module of the nth layer, which, when the value of the nth bit of the mode information indicates monophonic encoding, performs monophonic encoding in the nth layer using information associated with the monophonic signal and outputs coding distortion associated with the monophonic signal in the nth layer, and the information associated with the side signal, taken as input from the (n-1) th layer, into the coding module of the (n + 1) th layer; and
a stereo coding unit of the nth layer, which, when the value of the nth bit of the mode information indicates stereo coding, performs stereo coding in the nth layer using both information related to the monaural signal and information related to the side signal, and outputs the coding distortion associated with the monophonic signal in the nth layer, and the coding distortion associated with the side signal in the nth layer to the coding module of the (n + 1) th layer. 5. The stereo signal encoding device according to claim 4, wherein the Nth layer encoding module comprises:
a monophonic coding module of the Nth layer, which, when the value of the Nth bit of mode information indicates monophonic coding, performs monophonic coding in the Nth layer using information associated with the monophonic signal; and a stereo coding unit of the Nth layer, which, when the value of the Nth bit of mode information indicates stereo coding, performs stereo coding in the Nth layer using both information related to the monaural signal and information associated with the side signal. 6. The stereo signal encoding device according to claim 5, in which the stereo encoding module of the i-th layer contains:
a first conversion module that converts information associated with the monophonic signal into the frequency domain and provides a first spectrum;
a second conversion module that converts information associated with the side signal into the frequency domain and provides a second spectrum;
an integration module that integrates the first spectrum and the second spectrum to provide an integrated spectrum; and a spectrum coding unit that performs spectrum coding with respect to the integrated spectrum. 7. The stereo signal encoding device according to claim 6, in which the integration module integrates the first spectrum and the second spectrum so that the spectra of the same frequency are adjacent to each other. 8. The stereo signal encoding device according to claim 6, wherein the integration module integrates the first spectrum and the second spectrum so that the first spectrum is adjacent before or after the second spectrum. 9. The stereo signal encoding device according to claim 1, wherein the mode information generating unit generates mode information to be applied to the (i + 1) th layer using a monaural signal and a side signal received as input in the encoding unit i th layer. 10. The stereo signal encoding device according to claim 9, in which the mode information generating unit calculates the power of the monaural signal and side signal received as input in the i-layer layer encoding unit, and generates mode information based on the relative relationship between the calculated powers. 11. A device for decoding stereo signals, comprising:
a receiving module that receives mode information and encoded information of the first through Nth layers detected by encoding processing in the first through Nth layers, the mode information indicating which of monophonic encoding and stereo encoding is performed when encoding is processed in i- m layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) of the stereo signal encoding device, which performs encoding using the signal of the first channel and the signal of the second channel, forming x stereo signal;
first to Nth layer decoding modules that perform monophonic decoding or stereo decoding using encoded information of the i-th layer based on mode information and provide the result of decoding the monaural signal in the i-th layer and the result of decoding the side signal in the i-th layer wherein the monaural signal is associated with the sum of the signal of the first channel and the signal of the second channel, and the side signal is associated with the difference between the signal of the first channel and the signal of the second channel; and
a sum and difference calculator that calculates a decoded signal of the first channel and a decoded signal of the second channel using the result of decoding the monaural signal in the Nth layer and the result of decoding the side signal in the Nth layer. 12. A method for encoding stereo signals, comprising the steps of:
form a monaural signal associated with the sum of the signal of the first channel and the signal of the second channel forming the stereo signal, and form a side signal associated with the difference between the signal of the first channel and the signal of the second channel;
generating mode information per layer indicating a coding mode of one of monophonic coding and stereo coding; and
perform monophonic coding in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) using the information associated with the monophonic signal, or perform stereo coding in the i-th layer with both the information associated with the monophonic signal and the information associated with the side signal based on the mode information provide encoded information of the ith layer. 13. A method for decoding stereo signals, comprising the steps of:
receive mode information and encoded information of the first through Nth layers detected by encoding processing in the first through Nth layers, the mode information indicating which of the monophonic encoding and stereo encoding is performed in the encoding processing in the i-th layer (i = 1, 2, ..., N, where N is an integer equal to or greater than 2) a stereo signal encoding device that performs encoding using the signal of the first channel and the signal of the second channel forming a stereo signal l;
perform monophonic decoding or stereo decoding using the encoded information of the i-th layer based on the mode information and provide the result of decoding the monophonic signal in the i-th layer and the result of decoding the side signal in the i-th layer, the monaural signal being associated with the sum of the signal of the first channel and the signal of the second channel, and the side signal is associated with the difference between the signal of the first channel and the signal of the second channel; and
the decoded signal of the first channel and the decoded signal of the second channel are calculated using the result of decoding the monaural signal in the Nth layer and the result of decoding the side signal in the Nth layer.

Images