KR20140122990A - Apparatus and method for encoding/decoding multichannel audio signal - Google Patents

Abstract

A multi-channel audio signal encoding / decoding apparatus and method for up-mixing a single non-correlation signal for each audio channel signal.
The multi-channel audio signal encoding apparatus comprises: an N-channel encoding unit for encoding N audio channel signals to generate M audio channel signals and additional information; And an M channel encoder for encoding the M audio channel signals and outputting a bit stream.

Description

[0001] APPARATUS AND METHOD FOR ENCODING / DECODING MULTICHANNEL AUDIO SIGNAL [0002]

The present invention relates to an apparatus and method for encoding / decoding multi-channel audio signals, and more particularly, to a multi-channel audio signal encoding / decoding apparatus and method for applying up-mixing only one non-correlation signal for each audio channel signal.

MPS (MPEG Surround) is an audio codec for multi-channel audio coding, and is a coding / decoding technology capable of high-compression transmission of multi-channel audio signals.

The MPS may convert the input signal to a mono or stereo signal and transmit the additional information for upmixing the converted signal together with the converted signal to restore the multi-channel audio signal even if the number of channels of the input signal increases have. That is, since the signal transmitted by the encoding apparatus using the MPS is a mono or stereo signal, if the MPS encoding unit is not provided in the decoding apparatus, the signal converted using the general audio encoder can be decoded and reproduced. Further, when the decoding apparatus has an MPS decoding unit, it is also possible to decode a multi-channel audio signal.

However, the conventional MPS decoding apparatus can upmix the converted signal using the upmixing matrix as shown in Equation (1) in the process of decoding the multi-channel audio signal.

At this time, the mono DMX signal generated from m ₀ {L, R, Ls, Rs, C} and received from the encoding apparatus, and dm ⁱ ₀ may be an uncorrelated signal having mismatch with m ₀ . Further, the uncorrelated signal dm ⁱ ₀ may be a signal arbitrarily generated from the mono signal m ₀ .

That is, the decoding apparatus can recover {Lsynth, Rsynth, LSsynth, RSsynth} signals by scaling and summing m ₀ and dm ⁱ ₀ through an upmixing matrix.

However, as shown in Equation (2), the conventional upmixing matrix uses a plurality of uncorrelated signals for one channel of a mono signal, resulting in deterioration of sound quality.

Therefore, there is a demand for an apparatus and a method capable of minimizing deterioration of sound quality in a multi-channel audio encoding / decoding apparatus.

The present invention can provide an apparatus and method for generating a sound field effect close to an original sound by minimizing distortion of the sound quality by upmixing only one non-correlation signal for each audio channel signal.

In addition, according to an embodiment of the present invention, an apparatus and method for applying only one non-correlation signal to each audio channel signal by determining the number of channels to be downmixed according to the number of channels of the input signal can be provided.

An apparatus for encoding a multi-channel audio signal according to an exemplary embodiment of the present invention includes: an N-channel encoder for encoding N audio channel signals to generate M audio channel signals and additional information and outputting additional information; And an M channel encoder for encoding the M audio channel signals and outputting a bit stream.

The N-channel encoding unit of the multi-channel audio signal encoding apparatus according to an embodiment of the present invention includes: a signal grouping unit for grouping N audio channel signals into two and outputting N audio channel signals when N is an even number; And a signal processing unit for extracting and outputting the additional information from the two grouped audio channel signals, downmixing the grouped two audio channel signals into one signal, and outputting the M audio channel signals. have.

The N-channel encoding unit of the multi-channel audio signal encoding apparatus according to an embodiment of the present invention includes a signal grouping unit for grouping N audio channel signals into two and outputting N audio channel signals when N is an odd number; (N-1) / 2 channel audio channel signals by downmixing the two grouped audio channel signals into one signal, and outputs the audio channel signals A processor; And a signal delay unit for delaying and outputting an audio channel signal of a non-grouped channel among the N audio channel signals.

The signal delay unit of the multi-channel audio signal encoding apparatus according to an embodiment of the present invention includes a signal delay unit for extracting the additional information from the signal processing unit and for downmixing the received two signals into one signal, Can be delayed.

An apparatus for decoding a multi-channel audio signal according to an exemplary embodiment of the present invention includes an M-channel decoding unit decoding M audio channel signals in a received bitstream; And an N-channel decoding unit decoding the N audio channel signals using the M audio channel signals and the additional information.

An N-channel decoding unit of a multi-channel audio signal decoding apparatus according to an exemplary embodiment of the present invention includes an uncorrelated signal generating unit for generating M non-correlation signals using M audio channel signals; And an upmixing unit for upmixing the M audio channel signals and the M non-inertial signals with additional information to output N audio channel signals.

An N-channel decoding unit of a multi-channel audio signal decoding apparatus according to an exemplary embodiment of the present invention includes: a signal delay unit for delaying and outputting one audio channel signal among the M audio channel signals; An uncorrelated signal generating unit for generating (M-1) non-inferiority signals using (M-1) audio channel signals that are not delayed among the M audio channel signals; And an upmixing unit for upmixing the (M-1) audio channel signals and the (M-1) non-inertial signals with additional information to output (M-1) * 2 audio channel signals .

The signal delay unit of the multi-channel audio signal decoding apparatus according to an embodiment of the present invention may delay one of the M audio channel signals based on the operation time of the uncorrelated signal generating unit and the upmixing unit have.

According to another aspect of the present invention, there is provided a multi-channel audio signal encoding method, comprising: generating M audio channel signals and additional information by encoding N audio channel signals; Outputting the additional information; And outputting a bit stream by encoding the M audio channel signals.

According to an embodiment of the present invention, there is provided a method of decoding a multi-channel audio signal, comprising: decoding M audio channel signals in a received bitstream; And decoding the N audio channel signals using the M audio channel signals and the additional information.

According to an embodiment of the present invention, only one non-correlated signal is applied to each audio channel signal and upmixed, thereby minimizing the distortion of the sound quality and generating the sound field effect close to the original sound.

According to an embodiment of the present invention, only one non-correlation signal may be applied to each audio channel signal by determining the number of channels to be downmixed according to the number of channels of the input signal.

1 is a block diagram of a multi-channel audio signal encoding apparatus and a multi-channel audio signal decoding apparatus according to an embodiment of the present invention.
2 is a diagram illustrating an N-channel encoding unit according to an embodiment of the present invention.
3 is a block diagram of the N-channel encoding unit according to the first embodiment of the present invention.
4 is a block diagram of an N-channel encoding unit according to a second embodiment of the present invention.
FIG. 5 shows a configuration of an N-channel encoding unit according to a third embodiment of the present invention.
FIG. 6 shows a configuration of an N-channel encoding unit according to a fourth embodiment of the present invention.
7 is a block diagram illustrating an N-channel decoding unit according to an embodiment of the present invention.
FIG. 8 shows a configuration of an N-channel decoding unit according to the first embodiment of the present invention.
FIG. 9 shows a configuration of an N-channel decoding unit according to a second embodiment of the present invention.
10 is a flowchart illustrating a multi-channel audio signal encoding method according to an embodiment of the present invention.
11 is a flowchart illustrating a multi-channel audio signal encoding method according to an embodiment of the present invention.
12 is a flowchart illustrating a multi-channel audio signal decoding method according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The multi-channel audio signal encoding method and the multi-channel audio signal decoding method according to an embodiment of the present invention can be performed by a multi-channel audio signal encoding apparatus and a multi-channel audio signal decoding apparatus, respectively.

1 is a block diagram of a multi-channel audio signal encoding apparatus and a multi-channel audio signal decoding apparatus according to an embodiment of the present invention.

Referring to FIG. 1, a multi-channel audio signal encoding apparatus 110 according to an embodiment of the present invention may include an N-channel encoding unit 111 and an M-channel encoding unit 112.

The N-channel encoding unit 111 may encode N audio channel signals to generate M audio channel signals and additional information. At this time, M may be determined differently depending on whether N is an even number or an odd number. For example, M may be determined according to Equation (2).

That is, when N is an even number, M is N / 2, and when N is an odd number, M may be (N-1) / 2 + 1.

Further, the additional information may include at least one of a channel level difference (CLD), an inter-channel correlation / coherence (ICC), and an inter-channel phase difference (IPD).

The N-channel encoding unit 111 may transmit the generated M audio channel signals to the M-channel encoding unit 112, and may transmit the generated additional information to the multi-channel audio signal decoding apparatus 120.

The detailed configuration and operation of the N-channel encoding unit 111 will be described below with reference to FIG. 2 to FIG.

The M-channel encoding unit 112 may encode M audio channel signals generated by the N-channel encoding unit 111 and output a bit stream. At this time, the M-channel encoding unit 112 may be one of the encoders capable of encoding M channels. For example, the M-channel encoding unit 112 may be a USAC coder, which is an Extended HE-AAC capable of encoding up to 24 channels.

Referring to FIG. 1, a multi-channel audio signal decoding apparatus 120 according to an embodiment of the present invention may include an M-channel decoding unit 121 and an N-channel decoding unit 122.

The M-channel decoding unit 121 may receive the bitstream from the multi-channel audio signal encoding apparatus 110 and may decode M audio channel signals in the received bitstream. At this time, the M-channel decoding unit 121 may be one of decoders capable of decoding M channels. For example, the M-channel decoding unit 121 may be a USAC decoder that is an Extended HE-AAC capable of decoding up to 24 channels.

The N-channel decoding unit 122 receives the additional information from the M-channel encoding unit 112 and generates N audio channel signals using the received additional information and the M audio channel signals decoded by the M-channel decoding unit 121. [ Can be decoded.

The detailed configuration and operation of the N-channel decoding unit 122 will be described below with reference to FIG. 7 to FIG.

2 is a diagram illustrating an N-channel encoding unit according to an embodiment of the present invention.

2, an N-channel encoding unit 111 according to an exemplary embodiment of the present invention may include a signal grouping unit 210, a signal processing unit 220, and a signal delaying unit 230. Referring to FIG.

The signal grouping unit 210 may group the N audio channel signals into two and output them.

The signal processing unit 220 extracts and outputs the additional information from the two audio channel signals grouped by the signal grouping unit 210 and outputs the two audio channel signals grouped by the signal grouping unit 210 as one signal And mix and output M audio channel signals.

When N, which is the number of channels of the audio channel signals input to the multi-channel audio signal encoding apparatus, is an even number, the signal processing unit 220 may include M signal processors in parallel. The signal processing unit 220 may be configured such that the two audio channel signals grouped by the signal grouping unit 210 are input to one signal processor. At this time, each of the signal processors can receive the two audio channel signals grouped by the signal grouping unit 210, and output the additional information and one audio channel signal. Also, since the number of signal processors is M, the audio channel signal outputted by the signal processors can be M audio channel signals as a result.

Specifically, the signal processor extracts the spatial cues CLD, ICC, and IPD from the received two audio channel signals, outputs them as additional information, and outputs one audio channel signal that downmixes the received two audio channel signals . For example, the signal processor may be an encoding unit such as a Two-To-One (TTO) Box of the MPS.

When N, which is the number of channels of audio channel signals input to the multi-channel audio signal encoding apparatus, is an even number, the specific configuration and operation of the N-channel encoding unit 122 will be described in detail with reference to FIG.

When N, which is the number of channels of the audio channel signal input to the multi-channel audio signal encoding apparatus, is odd, the signal processing unit 220 extracts additional information from the two audio channel signals grouped by the signal grouping unit 210 And the signal grouping unit 210 downmixes the two audio channel signals grouped by the signal grouping unit 210 into one signal to output (N-1) / 2 channel audio channel signals. At this time, the signal processing unit 220 may include (N-1) / 2 signal processors in a parallel structure.

If N, which is the number of channels of the audio channel signal input to the multi-channel audio signal encoding apparatus, is an odd number, M may be (N-1) / 2 + 1. That is, the audio channel signal of the (N-1) / 2 channel output by the signal processing unit 220 may be less than one channel of the M audio channel signals. The N channel encoding unit 122 outputs the audio channel signal output from the signal delay unit 230 together with the (N-1) / 2 channel audio channel signal output from the signal processing unit 220, thereby generating M It is possible to output an audio channel signal.

When the number N of channels of the audio channel signal input to the multi-channel audio signal encoding apparatus is an odd number, the specific configuration and operation of the N-channel encoding unit 122 will be described in detail with reference to FIG.

The number of channels N 'of the audio channel signals input to the multi-channel audio signal encoding apparatus 110 is N', which is the number of channels processed by the signal processing unit 220 and N ' The number of channels, and the number of channels. That is, N may be N '+ K.

At this time, M may be determined differently depending on whether N 'is an even number or an odd number. For example, M may be determined according to Equation (3).

That is, when N 'is an even number, M may be N' / 2 + K, and when N 'is an odd number, M may be (N'-1) / 2 + 1 + K.

The signal delay unit 230 may delay and output the audio channel signals of the channels not grouped by the signal grouping unit 210 among the N audio channel signals.

When N, which is the number of channels of the audio channel signals input to the multi-channel audio signal encoding apparatus, is odd, an audio channel signal that is not grouped by the signal grouping unit 210 for grouping the audio channel signals into two can be generated. At this time, if the audio channel signal that is not grouped is directly output, the synchronization with the audio channel signal output from the signal processing unit 220 may be different.

Accordingly, the signal delay unit 230 may output the audio channel signal that is not grouped by delaying the audio channel signal to synchronize with the audio channel signal output from the signal processing unit 220. At this time, the signal delay unit 230 delays the audio channel signal of the ungrouped channel based on the time for the signal processing unit 220 to extract the additional information and downmix the received two signals to one signal, Time can be determined.

For example, when M audio channel signals are PCM signals, the signal delay unit 230 may determine Enc_Delay, which is a delay time, using Equation (4).

In this case, Delay 1 (QMF Analysis) may be a delay time occurring in 64 band QMF analysis of MPS, and Delay 2 (Hybrid QMF Analysis) may be a delay time occurring in Hybrid Analysis using a filter of 13 tap. In addition, Delay 3 (QMF Synthesis) may be a delay time occurring in MPS 64-band QMF synthesis. At this time, Delay 3 (QMF Synthesis) may be the same time as Delay 1 (QMF Analysis).

For example, Delay 1 (QMF Analysis) may be 288 and Delay 2 (Hybrid QMF Analysis) may be 6 x 64 = 384. Since the Hybrid QMF is performed after the 64-band QMF is performed, the signal delay unit 230 can determine Delay 2 (Hybrid QMF Analysis) by multiplying 6 by 64.

For example, when M audio channel signals are QMF signals, the signal delay unit 230 can determine Enc_Delay, which is a delay time, using Equation (5).

3 is a block diagram of the N-channel encoding unit according to the first embodiment of the present invention.

3 is an example of an N-channel encoding unit 111 when N, which is the number of channels of audio channel signals input to the multi-channel audio signal encoding apparatus 110, is an even number.

When the number N of channel numbers of audio channel signals input to the multi-channel audio signal encoding apparatus 110 is an even number, the signal processing unit 220 includes M signal processors 300 in a parallel structure as shown in FIG. can do. At this time, the signal processor 300 may be a two-to-one (TTO) box.

3, the signal grouping unit 210 groups the N audio channel signals 310 into two, and transmits the grouped audio channel signals 320 to one signal processor 300 have. At this time, the number of the audio channel signals 320 in which the N audio channel signals 310 are grouped by two may be N / 2. Also, when N is an even number, M is N / 2, so that the audio channel signals 320 in which the N audio channel signals 310 are grouped into two can correspond to one signal processor 300, respectively.

Next, the signal processors 300 can receive the two audio channel signals 320 grouped by the signal grouping unit 210, and output the additional information and one audio channel signal. In addition, since the number of the signal processors 300 is M, the signal processor 220 including the signal processors 300 can output M audio channel signals 430.

4 is a block diagram of an N-channel encoding unit according to a second embodiment of the present invention.

4 is an example of an N-channel encoding unit 111 when N, which is the number of channels of an audio channel signal input to the multi-channel audio signal encoding apparatus 110, is an odd number.

When the number N of the channel of the audio channel signal input to the multi-channel audio signal encoding apparatus 110 is an odd number, the signal processing unit 220 may include (N-1) / 2 signal processors 400 ) Can be included in a parallel structure.

3, the signal grouping unit 210 groups the N audio channel signals 410 into two and transmits the grouped audio channel signals 420 to one signal processor 400 have.

N is an odd number, the signal grouping unit 210 may not group one audio channel signal 421 among the N audio channel signals 410 by grouping the N audio channel signals 410 into two. At this time, the signal grouping unit 210 may transmit the audio channel signal 421 that is not grouped to the signal delay unit 230.

Since the number of the audio channel signals 420 grouped by the signal grouping unit 210 is the number excluding the audio channel signal 421 transmitted to the signal delay unit 230 among the N audio channel signals, ). Since the signal grouping unit 210 groups the two audio channel signals into one group, the number of the audio channel signals 420 grouped by the signal grouping unit 210 may be (N-1) / 2 . Accordingly, the audio channel signals 420 grouped by the signal grouping unit 210 may be transmitted in a one-to-one correspondence with (N-1) / 2 signal processors 400 of the parallel structure.

Next, each of the (N-1) / 2 signal processors 400 receives the two audio channel signals 420 grouped by the signal grouping unit 210, and outputs the additional information and one audio channel signal . In this case, since the number of signal processors 400 is (N-1) / 2, the signal processor 220 including the signal processors 400 has a number of channels equal to (M-1) / The audio channel signal 431 can be output.

At this time, the signal delay unit 230 may delay the audio channel signal 421 received from the signal grouping unit 210 by a time period for processing the two audio channel signals grouped by the signal processor 400. [

The signal output from the N-channel encoding unit 111 is supplied to the M-1 channel audio channel signal 431 output by the signal processing unit 220 and the 1-channel audio channel signal 431 output from the signal delay unit 230 432 are included, it can be M audio channel signals 430.

FIG. 5 shows a configuration of an N-channel encoding unit according to a third embodiment of the present invention.

5 is an example of the N-channel encoding unit 111 when N, which is the number of channels of the audio channel signal input to the multi-channel audio signal encoding apparatus 110 is N '+ K and N' is an even number.

When the number N of channels of the audio channel signals input to the multi-channel audio signal encoding apparatus 110 is N '+ K and N' is an even number, the signal processing unit 220 performs N / Two signal processors 500 may be included in parallel.

5, the signal grouping unit 210 includes N 'audio channel signals 520 for transmitting N audio channel signals 510 to the signal processing unit 200 and N channel audio signals 520 for transmitting to the signal processing unit 200 And may be grouped into K audio channel signals 530 to be transmitted to the signal delay unit 230 without being transmitted.

The signal grouping unit 210 may group the N 'audio channel signals 520 into two and transmit the grouped audio channel signals 520 to one signal processor 500.

Also, the signal grouping unit 210 may transmit the K audio channel signals 530 to the signal delay unit 230, respectively. At this time, the N-channel encoding unit 111 may be one-to-one matched with K audio channel signals 530 by K signal delay units 230 as shown in FIG. The N channel encoder 111 encodes the K channel audio signals 530 and 530 into one signal delay unit 230 in the case of a signal delay unit in which the signal delay unit 230 can simultaneously receive and delay a plurality of signals, ). ≪ / RTI >

Next, each of the N '/ 2 signal processors 500 receives the two audio channel signals grouped by the signal grouping unit 210, and outputs the additional information and one audio channel signal. At this time, since the number of the signal processors 500 is N '/ 2, the signal processing unit 220 including the signal processors 500 can output N' / 2 audio channel signals 541.

At this time, the signal delay unit 230 may delay and output the K audio channel signals 530 received from the signal grouping unit 210 by the time for processing the two audio channel signals grouped by the signal processor 500 have.

The signal 540 output from the N channel encoding unit 111 is supplied to the N '/ 2 audio channel signals 541 output from the signal processing unit 220 and the K audio channels Signal 542 may all be included. At this time, since M is N / 2 + K, the signal 540 output from the N-channel encoding unit 111 may be M audio channel signals.

FIG. 6 shows a configuration of an N-channel encoding unit according to a fourth embodiment of the present invention.

6 is an example of the N-channel encoding unit 111 when N, which is the number of channels of the audio channel signal input to the multi-channel audio signal encoding apparatus 110, is N '+ K and N' is an odd number.

When the number N of channels of the audio channel signal input to the multi-channel audio signal encoding apparatus 110 is N '+ K and N' is an odd number, the signal processing unit 220 outputs N ' -1) / two signal processors 600 in a parallel structure.

6, N 'audio channel signals 620 for transmitting N audio channel signals 610 to the signal processing unit 200 and N channel audio signals 620 for transmitting the N audio channel signals 610 to the signal processing unit 200 And may be grouped into K audio channel signals 630 to be transmitted to the signal delay unit 230 without being transmitted.

The signal grouping unit 210 may group the N 'audio channel signals 620 into two, and transmit the grouped audio channel signals 620 to one signal processor 600. Since N 'is an odd number, when the signal grouping unit 210 groups the N' audio channel signals 620 two by two, one of the N 'audio channel signals 620 is grouped I can not. At this time, the signal grouping unit 210 can transmit the audio channel signal 621 that can not be grouped to the signal delay unit 230.

In addition, the signal grouping unit 210 may transmit the K audio channel signals 630 to the signal delay unit 230, respectively. At this time, the N-channel encoding unit 111 may be one-to-one matched with the K audio channel signals 630 by the K signal delay units 230 as shown in FIG. The N channel encoding unit 111 encodes the K audio channel signals 630 and 630 into one signal delay unit 230 in the case of a signal delay unit in which the signal delay unit 230 can simultaneously receive and delay a plurality of signals, ). ≪ / RTI >

Next, each of the (N'-1) / 2 signal processors 600 receives the two audio channel signals grouped by the signal grouping unit 210, and outputs the additional information and one audio channel signal . At this time, since the number of the signal processors 600 is (N'-1) / 2, the signal processor 220 including the signal processors 500 receives (N'-1) / 2 audio channel signals 541 Can be output. In addition, the signal delay unit 230 may delay the audio channel signal 621 received from the signal grouping unit 210 by a time required for processing the two audio channel signals grouped by the signal processor 600.

That is, the number of audio channel signals 641 processed by the N channel encoding unit 111 and processed by the N channel audio signal 620 in the signal processing unit 220 and the signal delay unit 230 is (N'- (N-1) / 2 + 1) / 2, since the sum of the audio channel signals output one by one from the audio signal processor 600 and the audio channel signals 642 delayed by the signal delay unit 230, have.

The signal delay unit 230 may delay and output the K audio channel signals 630 received from the signal grouping unit 210 by the time for processing the two audio channel signals grouped by the signal processor 600 have.

The signal 640 output from the N channel encoding unit 111 is supplied to the (N-1) / N channel audio signal processing unit 220 and the signal delay unit 230, A signal including both 2 + 1 audio channel signals 641 and K audio channel signals 642 output from the signal delay unit 230 may be included. At this time, since M is N / 2 + K, the signal 640 output from the N-channel encoding unit 111 may be M audio channel signals.

FIG. 7 illustrates an N-channel decoding unit according to an embodiment of the present invention. Referring to FIG.

7, an N-channel decoding unit 122 according to an embodiment of the present invention may include an uncorrelated signal generating unit 710, an upmixing unit 720, and a signal delaying unit 730 .

The noncorrelated signal generator 710 can generate a plurality of emergency signals using M audio channel signals decoded by the M channel decoder 121. [

If N is an even number, the uncorrelated signal generator 710 may generate an M-channel audio channel signal using the M signal generators.

Also, when N is an odd number, the non-correlation signal generator 710 can generate an M-1 channel audio channel signal using the M-1 signal generators. At this time, an audio channel signal that does not generate an uncorrelated signal among the M audio channel signals may be input to the signal delay unit 730.

The upmixing unit 720 upmixes the M audio channel signals and the M non-coherent signals generated by the non-correlation signal generating unit 710 with the additional information received from the N channel encoding unit 111, A signal can be output.

For example, the upmixing unit 720 of the N-channel audio signal using the equation (6) column vector _{y (n) = [y 0} (n), y 1 (n), ..., y M - ₁ (n)] ^T can be determined.

In this case, the vector sequence m (n) = [m ₀ (n), m ₁ (n), ..., m _{M -1} (n)] ^T is M audio channel signals,

May be M non-coherent channel signals generated by the uncorrelated signal generator 710. [

Also, the M (n) matrix may be a matrix that performs upmixing on the input signal at n sample times. For example, the matrix M (n) can be constructed as shown in Equation (7).

In this case, 0 is a 2x2 zero matrix, and R _i (n) can be a 2x2 matrix defined as Equation (8).

At this time, the component of R _i (n)

Can be extracted from the additional information received from the N-channel encoding unit 111. [

The additional information can be received for each b index, which is a frame unit received from the N-channel encoding unit 111, and R _i (n) applied in units of samples can be obtained by interpolation between neighboring frames.

More specifically, the upmixing unit 720 performs upmixing using the MPS method

. ≪ / RTI >

At this time, the upmixing unit 720 may extract C _{L , R} from the CLD included in the side information, and extract α (b) and β (b) from the CLD and ICC included in the side information. And, C _{L , R} , α (b), and β (b) can correspond to the MPS standard.

The operator □ in Equation (6) may be an operator that performs an interlace operation on each element of a vector to generate a new vector string. For example, the operator < RTI ID = 0.0 > D < / RTI >

Equation (6) may be expressed by Equation (11) using Equations (7) to (10).

The matrix element calculation process according to Equation (11) may be expressed as Equation (12).

The upmixing unit 720 multiplies M non-inertial signals d (n) and m (n) corresponding to M audio channel signals m (n) by 2x2 elements of the upmixing matrix Matrix, it is possible to output the N audio channel signals y (n).

The N channel decoding unit 122 according to the embodiment of the present invention generates N audio channel signals by applying only one uncorrelated signal d (n) for each audio channel signal m (n), thereby minimizing distortion of the sound quality And the effect of the sound field can be generated close to the original sound.

If N is odd, there may be one audio channel signal that does not require upmixing. Therefore, the uncorrelated signal generator 710 may not generate an uncorrelated signal corresponding to an audio channel signal that does not require upmixing. The signal delay unit 730 may delay and output an audio channel signal that does not generate an uncorrelated signal.

If an audio channel signal for which an uncorrelated signal is not generated is directly output, the synchronization with the audio channel signal output from the upmixing unit 720 may be different.

Therefore, the signal delay unit 730 can output the audio channel signal that does not generate the decorrelation signal and outputs the audio channel signal in synchronization with the audio channel signal output from the upmixing unit 720. At this time, the signal delay unit 730 may delay the up-mixer 720 to generate the non-correlation signal based on the time for upmixing the (M-1) audio channel signals and (M-1) The delay time for delaying the audio channel signal can be determined.

Further, an additional delay may occur in the process of the M-channel decoding unit 121 performing the decorrelator.

Therefore, the Dec_Delay, which is the delay time for delaying the audio channel signal by the signal delay unit 730, may be different from the delay time determined by the time delay unit 230 of the N-channel encoding unit 111. [ At this time, the delay time may be a value set in a time delay unit (Delay Unit) used by the signal delay unit 730 to delay the audio channel signal.

For example, when the signal decoded by the M-channel decoding unit 121 is a PCM (Pulse Code Modulation) signal, the signal delay unit 730 can determine Dec_Delay, which is a delay time, using Equation (13).

At this time, Delay 4 (Decorrelator filtering delay) may be a delay time occurring in the process of performing the decorrelator.

If the signal decoded by the M-channel decoding unit 121 is a QMF signal, the signal delay unit 730 can determine Dec_Delay, which is a delay time, using Equation (14).

FIG. 8 shows a configuration of an N-channel decoding unit according to the first embodiment of the present invention.

8 is a configuration of an N-channel decoding unit when N, which is the number of channels of an audio channel signal input to the multi-channel audio signal encoding apparatus 110, is not defined as N '+ K.

The non-correlation signal generator 710 may include a plurality of signal generators 800 as shown in FIG.

For example, when the number N of channels to be decoded by the N-channel decoding unit 122 is an even number, N may be M * 2. That is, the N-channel decoding unit 122 can output N audio channel signals 830 by upmixing M audio channel signals 810 and adding M audio channel signals.

The noncorrelated signal generator 710 includes M signal generators 800 and inputs M audio channel signals 810 to M signal generators 800 to generate M noncorrelated signals 821. [ Can be generated.

The upmixing unit 720 may upmix the M audio channel signals 710 and the M non-correlation signals 821 to output N audio channel signals 830 of M * 2.

If the number N of channels to be decoded by the N-channel decoding unit 122 is an even number, the signal delay unit 730 may not operate.

For example, when the number N of channels to be decoded by the N-channel decoding unit 122 is an odd number, N may be (M-1) * 2 + 1. That is, the N channel decoding unit 122 can output N audio channel signals 830 by upmixing the M audio channel signals 810 and adding (M-1) audio channel signals.

At this time, the decorrelation signal generator 710 includes (M-1) signal generators 800 and (M-1) signal generator 800 outputs (M-1) (M-1) pieces of non-correlation signals 821 can be generated.

Of the M audio channel signals, the audio channel signal 811 not input to the signal generator 800 may be input to the signal delay unit 730.

The upmixing unit 720 upmixes (M-1) audio channel signals 710 and (M-1) non-correlation signals 821 to generate (M-1) (831).

At this time, the signal delay unit 730 receives the (M-1) number of audio channel signals and the (M-1) number of the non-inertial signals after the upmixing unit 720 receives the audio channel signals. It is possible to delay the audio channel signal 811 not inputted to the signal generator 800 based on the time required for outputting the (M-1) * 2 audio channel signals 831.

The signals output from the N channel decoding unit 122 are supplied to the (M-1) * 2 audio channel signals 831 output from the upmixing unit 720 and one audio (M-1) * 2 + 1, since it includes all of the channel signals 832 and 832.

FIG. 9 shows a configuration of an N-channel decoding unit according to a second embodiment of the present invention.

9 shows a configuration of the N-channel decoding unit 122 when N, which is the number of channels of audio channel signals input to the multi-channel audio signal encoding apparatus 110, is defined as N '+ K.

If N is defined as N '+ K and N' is an even number, M may be N '/ 2 + K. Further, when N 'is an odd number, M may be (N'-1) / 2 + 1 + K.

That is, if N is defined as N '+ K, the M audio channel signals 910 may include the K audio channel signals 930 and the N' audio channel signals 941 as the remaining audio channel signals 920). Accordingly, the M-channel decoding unit 120 groups the M audio channel signals decoded as shown in FIG. 9 into K audio channel signals 930 and remaining audio channel signals 920 except for K channels, To the decryption unit 122.

At this time, the remaining audio channel signal 920 may be N '/ 2 when N' is an even number and (N'-1) / 2 + 1 when N 'is an odd number. The number of signal generators 900 included in the emergency line signal generator 710 may be determined according to the number of the remaining audio channel signals 920.

For example, if N 'is an even number, N' may be (M-K) * 2 and the number of remaining audio channel signals 920 may be (M-K). That is, the N-channel decoding unit 122 can up-mix and amplify the remaining audio channel signals 920 to output N 'audio channel signals 941.

The noncorrelated signal generator 710 includes (MK) signal generators 900 and inputs the remaining audio channel signals 920 to the (MK) signal generators 900 to generate (MK) The correlation signal 921 can be generated.

Next, the upmixing unit 720 upmixes the remaining audio channel signals 920 and (MK) non-correlation signals 921 to output N 'number of audio channel signals 941 of (MK) * 2 .

At this time, the signal delay unit 730 receives the remaining audio channel signals 920 and (M-K) mismatch signals 921 from the upmixing unit 720. The K audio channel signals 930 can be delayed based on the time taken to output the N 'audio channel signals 941. [

The signal 940 output from the N channel decoding unit 122 is supplied to the N channel audio signal 841 output from the upmixing unit 720 and the K channel audio channel signal 842 output from the signal delay unit 730, (N '+ K), since it is a signal including all of the N channel audio signal 940 and the N channel audio signal 940.

For example, if N 'is an odd number, N' may be (M-K-1) * 2 + 1. That is, the N-channel decoding unit 122 adds (MK-1) audio channel signals by upmixing (MK-1) audio channel signals out of (MK) remaining audio channel signals 920, It is possible to output a plurality of audio channel signals 941.

At this time, the decorrelation signal generator 710 includes (MK-1) signal generators 900 and outputs one audio channel signal to the (MK-1) signal generators 900 in the remaining signal 920 (MK-1) pieces of non-correlation signals 941 can be generated by inputting the (MK-1) audio channel signals excluding the (MK-1)

Of the remaining audio channel signals 920, an audio channel signal 921 not input to the signal generator 900 and K audio channel signals 930 received from the M channel decoding unit 121 are input to the signal delay unit 730 As shown in FIG.

Next, the upmixing unit 720 upmixes (MK-1) audio channel signals and (MK-1) non-correlation signals 921 to output (MK-1) * 2 audio channel signals .

At this time, the signal delay unit 730 receives the (M-K-1) number of audio channel signals and the (M-K-1) number of the non-inertia signals 921 after the upmixing unit 720 receives the audio channel signals. It is possible to delay the audio channel signal 921 and the K audio channel signals 930 that have not been input to the signal generator 800 based on the time required to output the two audio channel signals MK-1 * MK-1 * .

The remaining audio channel signals 941 excluding the K audio channel signals 943 delayed from the signal 940 output from the N channel decoding unit 122 are input to the up mixer 720 as shown in FIG. (MK-1) * two audio channel signals 831 to be output and an audio channel signal 942 to be output by the signal delay unit 730. That is, the number of remaining audio channel signals 941 may be N 'with (M-K-1) * 2 + 1.

Since the signal 940 output from the N-channel decoding unit 122 is a signal including both the K audio channel signals 943 and the N 'audio channel signals 941, Channel signal 940. < / RTI >

10 is a flowchart illustrating a multi-channel audio signal encoding method according to an embodiment of the present invention.

10 is a flowchart illustrating a method of encoding a multi-channel audio signal when the number N of channel numbers of audio channel signals input to the multi-channel audio signal encoding apparatus 110 is an even number.

In step 1010, the signal processing unit 220 may group the N audio channel signals into two. At this time, the number of audio channel signals obtained by grouping the N audio channel signals into two can be N / 2.

In step 1020, the signal processing unit 220 may extract additional information from the two audio channel signals grouped in step 1010. At this time, the signal processing unit 220 includes N / 2 signal processors, and the signal processor can extract additional information from each grouped audio channel signal.

In step 1030, the signal processors of the signal processing unit 220 may downmix the two audio channel signals grouped in step 1010 to one signal to output M audio channel signals. At this time, the signal processing unit 220 includes M signal processors of N / 2, and the signal processor down-mixes the grouped audio channel signals into one audio channel signal to output M audio channel signals .

In step 1040, the M-channel encoding unit 112 may encode the M audio channel signals output in step 1030 to generate a bit stream.

In step 1050, the M-channel coding unit 112 may transmit the bitstream generated in step 1040 to the M-channel decoding unit 121 of the multichannel decoding apparatus 120. In addition, the signal processing unit 220 may transmit the additional information extracted in operation 1020 to the upmixing unit 720 of the multi-channel decoding apparatus 120.

11 is a flowchart illustrating a multi-channel audio signal encoding method according to an embodiment of the present invention.

11 is a flowchart illustrating a multi-channel audio signal encoding method when N, which is the number of channels of an audio channel signal input to the multi-channel audio signal encoding apparatus 110, is an odd number.

In step 1110, the signal processing unit 220 may group the N audio channel signals into two. At this time, since N is an odd number, the number of audio channel signals obtained by grouping N audio channel signals into two can be (N-1) / 2.

In addition, when the signal grouping unit 210 groups the N audio channel signals into two, it may not be possible to group one audio channel signal among the N audio channel signals. At this time, the signal grouping unit may transmit an audio channel signal that can not be grouped to the signal delay unit 230.

In step 1120, the signal processing unit 220 may extract additional information from the two audio channel signals grouped in step 1110. At this time, the signal processing unit 220 includes (N-1) / 2 signal processors, and the signal processor can extract additional information from each grouped audio channel signal.

In step 1130, the signal processors of the signal processing unit 220 down mix the two audio channel signals grouped in step 1110 into one signal to output (N-1) / two audio channel signals have. In this case, the signal processing unit 220 includes (N-1) / 2 signal processors, and the signal processor down-mixes the grouped audio channel signals into one audio channel signal, Two audio channel signals can be output.

In step 1140, the signal delay unit 230 may output an audio channel signal that can not be grouped in step 1110 by delaying the audio channel signal by the time the signal processor 400 performs steps 1120 and 1130 .

In step 1150, the M-channel encoding unit 112 encodes the (N-1) / 2 audio channel signals output in step 1130 and one audio channel signal output in step 1140, Can be generated. Since the sum of (N-1) / 2 audio channel signals output in step 1130 and one audio channel signal output in step 1140 is M (N-1) / 2 + Audio channel signals.

In step 1160, the M-channel encoding unit 112 may transmit the bitstream generated in step 1150 to the M-channel decoding unit 121 of the multichannel decoding apparatus 120. [ In addition, the signal processing unit 220 may transmit the additional information extracted in operation 1020 to the upmixing unit 720 of the multi-channel decoding apparatus 120.

12 is a flowchart illustrating a multi-channel audio signal decoding method according to an embodiment of the present invention.

In step 1210, the M-channel decoding unit 121 receives the bit stream from the multi-channel audio signal encoding apparatus 110 and decodes M audio channel signals in the received bit stream.

In step 1220, the N-channel decoding unit 122 can check whether N is an odd number. If N is an odd number, the N-channel decoding unit 122 may perform step 1250. [ If N is an even number, the N-channel decoding unit 122 may perform step 1230. [

In step 1230, the uncorrelated signal generator 710 may input M audio channel signals decoded in step 1210 to M signal generators, respectively, to generate M uncorrelated signals.

In step 1240, the upmixing unit 720 upmixes the M audio channel signals decoded in step 1210 and the M non-correlation signals 821 generated in step 1230 to generate N It is possible to output an audio channel signal.

In step 1250, the signal delay unit 730 receives one of the M audio channel signals decoded in step 1210, and outputs the delayed audio channel signal. At this time, the time for delaying the input audio channel signal by the signal delay unit 730 is determined based on the time when the steps 1230 to 1240 are performed or the time when the steps 1260 to 1270 are performed Can be determined.

In step 1260, the decorrelation signal generator 710 inputs (M-1) audio channel signals to (M-1) signal generators 800 to generate (M-1) Can be generated. At this time, the (M-1) audio channel signals may be audio channel signals excluding the audio channel signals input to the signal delay unit 730 in step 1250 among the M audio channel signals decoded in step 1210.

In step 1270, the upmixing unit 720 upmixes (M-1) audio channel signals and (M-1) non-correlation signals 821 generated in step 1260 to (M-1) * Two audio channel signals can be output.

At this time, the N-channel decoding unit 122 outputs both the (M-1) * 2 audio channel signals output in the step 1270 and the 1-channel audio channel signals output in the step 1250, The signal output from the demultiplexer 122 may be N audio channel signals of (M-1) * 2 + 1.

The present invention minimizes the distortion of the sound quality and generates the sound field effect close to the original sound by upmixing only one non-correlation signal for each audio channel signal. Also, according to the present invention, the number of channels to be downmixed is determined according to the number of channels of the input signal, so that only one non-correlation signal can be applied to each audio channel signal.

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

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

110: Multi-channel audio signal encoding apparatus
111: N-channel encoding unit
112: M-channel coding unit
120: Multichannel audio signal decoding device
121: M-channel decoding unit
122: N-channel decoding unit

Claims (20)

An N channel encoding unit for encoding N audio channel signals to generate M audio channel signals and additional information and outputting additional information; And
An M channel encoding unit for encoding the M audio channel signals and outputting a bitstream,
Channel audio signal encoding apparatus. The method according to claim 1,
Wherein when N is an even number, M is N / 2. 3. The method of claim 2,
Wherein the N-channel encoding unit comprises:
A signal grouping unit for grouping the N audio channel signals into two and outputting them; And
A signal processing unit for extracting and outputting the additional information from the two grouped audio channel signals, downmixing the grouped two audio channel signals into one signal and outputting the M audio channel signals,
Channel audio signal encoding apparatus. The method according to claim 1,
The additional information,
A channel level difference (CLD), an inter-channel correlation / coherence (ICC), and an inter-channel phase difference (IPD). The method according to claim 1,
Wherein when N is an odd number, M is (N-1) / 2 + 1. 6. The method of claim 5,
Wherein the N-channel encoding unit comprises:
A signal grouping unit for grouping the N audio channel signals into two and outputting them;
(N-1) / 2 channel audio channel signals by downmixing the two grouped audio channel signals into one signal, and outputs the audio channel signals A processor; And
A signal delay unit for delaying and outputting an audio channel signal of a non-grouped channel among the N audio channel signals,
Channel audio signal encoding apparatus. The method according to claim 6,
Wherein the signal delay unit comprises:
Wherein the signal processing unit extracts the additional information and delays an audio channel signal of a non-grouped channel based on a time period during which the two received signals are downmixed to one signal. The method according to claim 1,
Wherein when N is N '+ K and N' is an even number, M is N '/ 2 + K. The method according to claim 1,
Wherein the N-channel encoding unit comprises:
A signal grouping unit for grouping N 'audio channel signals into two and outputting them;
A signal processing unit for extracting and outputting the additional information from the two grouped audio channel signals, downmixing the grouped two audio channel signals into one signal, and outputting an N '/ 2 channel audio channel signal; And
A signal delay unit for delaying and outputting K audio channel signals,
Channel audio signal encoding apparatus. The method according to claim 1,
Wherein M is (N'-1) / 2 + 1 + K when N is N '+ K and N' is an odd number. The method according to claim 1,
Wherein the N-channel encoding unit comprises:
A signal grouping unit for grouping N 'audio channel signals into two and outputting them;
Extracts and outputs the additional information from the two grouped audio channel signals, downmixes the grouped two audio channel signals into one signal, and outputs (N'-1) / 2 channel audio channel signals A signal processor; And
An audio channel signal of a non-grouped channel among the N audio channel signals, and a signal delay unit for delaying and outputting K audio channel signals,
Channel audio signal encoding apparatus. An M-channel decoding unit decoding M audio channel signals in the received bitstream; And
An N-channel decoding unit for decoding N audio channel signals using the M audio channel signals and the additional information,
Channel audio signal. 13. The method of claim 12,
Wherein when N is an even number, N is M * 2. 14. The method of claim 13,
Wherein the N-channel decoding unit comprises:
An uncorrelated signal generating unit for generating M non-inertia signals using the M audio channel signals; And
An upmixing unit for upmixing the M audio channel signals and the M non-inertial signals with additional information and outputting N audio channel signals,
Channel audio signal. 13. The method of claim 12,
Wherein when N is an odd number, N is (M-1) * 2 + 1. 16. The method of claim 15,
Wherein the N-channel decoding unit comprises:
A signal delay unit for delaying and outputting one of the M audio channel signals;
An uncorrelated signal generating unit for generating (M-1) non-inferiority signals using (M-1) audio channel signals that are not delayed among the M audio channel signals; And
(M-1) * 2 audio channel signals by upmixing the (M-1) audio channel signals and the (M-1)
Channel audio signal. 17. The method of claim 16,
Wherein the M-
And grouping the decoded M audio channel signals into K audio channel signals and remaining audio channel signals when N is N '+ K. 18. The method of claim 17,
Wherein the N-channel decoding unit comprises:
A signal delay unit for delaying and outputting the K audio channel signals;
An uncorrelated signal generating unit for generating (MK) or (MK-1) non-correlation signals using the remaining audio channel signals; And
Mixer unit for upmixing the remaining audio channel signals with the (MK) or (MK-1) non-inertial signals as additional information and outputting N 'audio channel signals,
Channel audio signal. Encoding N audio channel signals to generate M audio channel signals and additional information;
Outputting the additional information; And
Encoding the M audio channel signals and outputting a bit stream;
Channel audio signal. Decoding M audio channel signals in the received bitstream; And
Decoding the N audio channel signals using the M audio channel signals and the additional information
Channel audio signal.

Images