WO2007069369A1

WO2007069369A1 - Signal processing device, and signal processing method

Info

Publication number: WO2007069369A1
Application number: PCT/JP2006/315932
Authority: WO
Inventors: Yasushi Sato
Original assignee: Kyushu Institute Of Technology
Priority date: 2005-12-14
Filing date: 2006-08-11
Publication date: 2007-06-21
Also published as: JP5103606B2; JP2007166315A

Abstract

It is intended to expand the bit number of information signals and to prevent the information at that time from degradation. For this intention, a signal from an input terminal (1) is fed through predetermined delay means (2) to a variable filter (3), and is then fed through an arbitrary low-pass filter (4) to operation means (5), and its difference from the output of the variable filter (3) is calculated and fed to coefficient correction means (6). The coefficient determined by the coefficient correction means (6) is fed to the variable filter (3). Moreover, the signal from the input terminal (1) is fed through predetermined delay means (7) and a less significant bit addition circuit (8) to operation means (9), and its difference from the output of the variable filter (3) is calculated. This difference is added through a weighting circuit (10) to the output of an addition circuit (11), and is extracted from an output terminal (12). Moreover, the difference from the operation means (9) is fed through a peak detection circuit (13) and a low-pass filter (14) to a limiter circuit (15), and the signal corresponding to the less significant value is deleted so that the weighting at the weighting circuit (10) is performed.

Description

Specification

Signal processing apparatus and signal processing method

Technical field

The present invention relates to a signal processing device and a signal processing method suitable for use when, for example, expanding the number of bits of a quantized acoustic signal. Specifically, the number of bits can be expanded well while blocking the harmonic components due to the expanded bits.

Background art

In a conventional signal processing apparatus and signal processing method, for example, as described in US Pat. No. 6055318 and US Pat. No. 6154547, processing such as noise reduction is performed using a feedback filter. It has been known. For quantized signals, as described in, for example, US Pat. No. 3,889,108, a device for converting the bandwidth using a feedback filter has been proposed. That is, a device for processing a quantized signal using a feedback filter is known. Disclosure of the invention

[0003] For example, for a compact disc (CD), it is generally recorded as an audio signal digitally with a sampling frequency of 44. lkHz and a quantization bit rate of 16 bits. Widely used. However, when the recorded 16-bit signal is directly converted to DZA (Digital to Analog), it appears that the subtlety of the playback sound quality is not enough nuance.

[0004] On the other hand, in the DZA conversion of acoustic signals, for example, 24-bit conversion means have become available at low cost, and it is conceivable to improve sound quality by using such conversion means. However, for example, if an extension is performed by simply inserting an 8-bit value of 0 below the 16-bit information, a large amount of noise is generated and the sound quality is significantly degraded. Moreover, if such noise is removed using a low-pass filter, it will result in a sound that feels full.

[0005] The present invention has been made in view of such problems, and an object of the present invention is to reduce an information signal quantized with a small number of bits, such as recording on a compact disc. Bi In addition to expanding the number of mobile stations, the deterioration of information such as the occurrence of noise during the expansion is prevented. As a result, it is possible to provide a signal processing apparatus and a signal processing method that can be directly applied to recording on a conventional compact disc or the like.

[0006] That is, in order to solve the above-described problems and achieve the object of the present invention, the present application provides a signal processing device for extending the number of bits of a quantized signal. Filter means that expands the number of bits of the quantized original signal and blocks the harmonic component due to the extended bits, and the output of the filter means is a signal in which any lower bits of the extension are added to the original signal The first calculating means for subtracting the force and the second calculating means for adding the output of the first calculating means to the output of the filter means.

[0007] In addition, in the signal processing device according to claim 2, the filter unit includes a variable filter including a delay unit and a correction unit that controls a coefficient thereof, and the correction unit is determined by a tendency of the original signal. The error ε between the output of the low-pass filter whose cutoff frequency is controlled and the output of the variable filter is calculated according to the external signal generated, and the coefficient of the variable filter is controlled using a correction algorithm that minimizes the error ε. It is characterized by that.

[0008] In the signal processing device according to claim 3, the moving correction average value of the peak is calculated with a time constant longer than the quantization sampling with respect to the output of the first arithmetic means, and the calculated movement is calculated. The amplitude of the output of the first calculation means is controlled by the corrected average value, and the second calculation means adds to the output of the filter means.

[0009] In the signal processing device according to claim 4, the original signal is a signal obtained by quantizing an acoustic signal.

[0010] In the signal processing device according to claim 5, the original signal is a signal obtained by quantizing the acoustic signal, and the tendency of the original signal includes the distinction between the conversation of the acoustic signal and the music, and the song or music. It is characterized by being Jeyangle.

[0011] Furthermore, in order to achieve the object of the present invention, in the invention described in claim 6, there is provided a signal processing method for extending the number of bits of a quantized signal, the method comprising: The number of bits is expanded and the harmonic components due to the expanded bits are cut off. The signal with the harmonic component cut off is also subtracted by subtracting the signal power obtained by adding any lower bits of the extension to the original signal. The number of bits is expanded by adding the signal to the signal whose harmonic components are cut off. It is characterized by obtaining a quantized signal.

Thus, according to the invention of the present application, for example, the number of bits of an information signal quantized with a small number of bits, such as recording on a compact disc, is expanded, and noise is generated during the expansion. Therefore, it is possible to provide a signal processing apparatus and a signal processing method that can be applied as they are to recording on a conventional compact disc or the like.

Brief Description of Drawings

FIG. 1 is a block diagram showing a configuration of an embodiment of an acoustic signal processing device to which a signal processing device and a signal processing method according to the present invention are applied.

FIG. 2 is a block diagram showing the configuration of the main part.

FIG. 3 is a diagram for explaining that.

FIG. 4 is a diagram for explaining that.

FIG. 5 is a diagram for explaining this.

FIG. 6 is a diagram for explaining that.

FIG. 7 is a diagram for explaining that.

FIG. 8 is a waveform diagram for explaining the effect.

BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an embodiment of an acoustic signal processing apparatus to which a signal processing apparatus and a signal processing method according to the present invention are applied. .

In FIG. 1, for example, a 16-bit PCM signal is supplied to the input terminal 1. A signal from the input terminal 1 is supplied to the variable filter 3 through a predetermined delay means 2. A signal from the input terminal 1 is supplied to the computing means 5 through the low-pass filter (LPF) 4, and a difference from the output of the variable filter 3 is calculated and supplied to the coefficient correcting means 6. Then, the coefficient obtained by the coefficient correcting means 6 is supplied to the variable filter 3.

[0016] Further, the signal from the input terminal 1 is supplied to the lower bit tracking circuit 8 through the predetermined delay means 7, and, for example, 8 bits of the value 0 are inserted into the lower order of the 16-bit input signal to obtain 24 bits. It is said that This 24-bit signal is supplied to the arithmetic means 9 and the output of the variable filter 3 is output. The difference from the force is calculated. Then, this difference value (24 bits) is added to the output of the variable filter 3 by the addition circuit 11 through the weighting circuit 10 and taken out to the output terminal 12.

In addition, the difference value from the calculation means 9 is supplied to the peak detection circuit 13, and the peak detected signal is supplied to the limiter circuit 15 through the low-pass filter (LPF) 14, and the signal corresponding to the lower 8 bits is deleted. Is done. The weighting circuit 10 performs weighting based on the output of the limiter circuit 15. As a result, the number of bits of the information signal is expanded and correction processing for information deterioration due to the expansion is performed.

Further, in this circuit, the variable filter 3 is an FIR type digital filter as shown in FIG. 2, for example, and is supplied to a plurality of unit delay means Z-1 in which input signals are connected in cascade. The outputs of each stage are added by the adding means (+) through the weighting circuit h. This added value is taken out as an output value y (n), and a difference from the target value d (n) is calculated by the calculation means 5 and the weighting circuit h is calculated by the coefficient correction means 6 so that this difference value is minimized. The coefficient of is obtained.

That is, if the signal supplied to the variable filter 3 is x (n), the output value y (n) is expressed as follows.

[Number 1]

M

k = 0

The difference value (error signal) ε (η) is

ε (n) = d (n) -y (n)

And this mean square error e

β = Ε {ε ² (η)}

The coefficient correction means 6 determines the coefficient so that is minimized. [0021] Therefore, the condition that the mean square error e is minimum is

e = E {d ² (n)}-2E {d (n) -y (n)} + E {y ² (n)}

As

E {d ² (n)} = Pd

Then, it becomes as follows.

[0022] That is,

[Equation 2] M

{d (n)-y (n)} = E {d (n) · ∑ h _k (n— k)} = ∑ h _k E {d (n) · x (n— k)} k = 0 k = 0, so

[Equation 3]

M M M

e = Pd-2∑ h _k P (k) + ∑∑ hkh _n y (| nk |) k = 0 k = 0 n = 0

It will be.

[0023] Meanwhile,

And

d (n) = Cos (2 π n / N) + s (n)

Where s (n) is a white signal,

s (n) = E {s ² (n)} = σ2

It is.

[0024] Here, for example, when a bit having a value of 0 is inserted in the lower order in a circuit as shown in B of FIG. 3 with respect to a sine wave signal having a frequency of 1 kHz as shown in FIG. As shown in Fig. 3C (a), the quantized signal has a large step at each quantization step as shown in Fig. 3 (b). Changes. Then, the frequency spectrum of such a signal is the same as the original sinusoidal signal as shown in D (a) of Fig. 3. Waves are generated.

[0025] On the other hand, by providing a low pass filter (LPF) as shown in Fig. 4A, from the signal in which harmonics are generated as shown in Fig. 4B (a), (b As shown in (), harmonic components can be removed. However, if you use it alone, you will end up with a muffled sound. Therefore, as shown in Fig. 5A, the low-pass filter (LPF) output is a signal obtained by deleting the bits below the original quantization with a limiter from the difference signal between the input signal and the output of the low-pass filter (LPF). Add to.

According to this, the waveform is smoothed as shown in (b) of FIG. 5 by the signal power low pass filter (LPF) quantized as shown in (a) of B of FIG. As shown in (c) of the figure, the high-frequency component deleted by the low-pass filter (LPF) is added, so that this kind of sound can be prevented. In other words, in this case, the low-pass filter (LPF) band has high sound quality of 24 bits, and high sound is added through as it is.

Furthermore, in the configurations of FIGS. 4 and 5, there is a problem in that the cutoff frequency is fixedly applied by force using a low-pass filter (LPF) as means for smoothing the waveform. Therefore, an adaptive filter (LMS) using delay means is used as shown in Fig. 6A. That is, such an adaptive filter is equivalent to the variable filter shown in FIGS. 1 and 2, and thus includes harmonic components as shown in FIG. 6B (a). As shown in (b) of the figure, harmonic components can be adaptively removed from the signal to smooth the waveform.

[0028] In the circuit A of Fig. 6, the quantization noise is high even in the case of force continuity signals that are extracted by an adaptive filter (LMS) focusing on the continuity signals. There are many. Therefore, as shown in Fig. 7, the low-pass filter (LPF) is used to remove high-frequency quantization noise and correct the adaptive filter (LMS) coefficient. Here, the cutoff frequency of the low-pass filter (LPF) is generally about 1 OkHz when the sampling frequency force is 4.1 kHz, that is, half the frequency of Nyquist.

[0029] Further, the cut-off frequency of the low-pass filter (LPF) is the original acoustic signal conversation and music. Depending on the type of music or the genre of music, changes such as narrowing the passband can be made. Thus, for example, better sound quality improvement can be performed according to the genre of music.

That is, in FIG. 1 described above, for example, the music information of a compact disc (CD) is 16 bits. Therefore, an adaptive filter is formed by the delay unit + variable filter unit, and the variable filter is controlled by the least square method so as to approach the signal source expanded to 24 bits by removing harmonics from the reference signal through LPF. As a result, for continuous (highly reproducible signals) due to the delay, the filtering effect is used to interpolate from 16 bits to 24 bits. However, since this sound is a reproducible signal, it becomes a muffled sound.

[0031] Then, the lower 8 bits of the source signal delayed by the variable filter processing are added (the lower 8 bits are 0), and the difference from the adaptive filter is calculated. This difference signal can be expanded to 24 bits by not adding it in the case of a signal with a magnitude less than the force quantization error, which is a high-frequency signal, and the conventional signal is added to the high-frequency information. Therefore, it can solve the problem.

[0032] In addition, the peak detection in FIG. 1 is a pop-up because a scene of sudden addition or supple force occurs if it is determined whether or not the force to add is based on the magnitude of the quantization error alone. There is a risk of noise. Therefore, the peak error is calculated as 1.0 at maximum, and the amount of error is multiplied by the amount of error, so the high frequency signal can be added smoothly, eliminating pop noise. be able to.

[0033] Further, the signal processing apparatus and signal processing method of the present invention!

By providing (or a fixed LPF), only the low frequency part of the target signal can be used as the target value of the adaptive filter. According to this, it is possible to extend the bit precision to 24 bits by LPF, and the adaptive filter does not follow high-frequency signals even with highly continuous signals.

[0034] In the case of a high-frequency signal, there is a cyclic signal due to a quantization error, so that it does not react to such a signal. In addition, peak detection of the error signal and use of the moving correction average (LPF) suppresses excessive response to sudden music signals. As a result, pop noise can be eliminated. In addition, limit the peak detection and error signal By controlling the amplitude, the rate of change can be made square.

Thus, according to the signal processing device and the signal processing method of the present invention, for example, the sound source of a compact disk (CD) is 16 bits, which is insufficient in terms of high sound quality. On the other hand, 24-bit DACs can be used at low cost, so there are many 24-bit sound sources. Therefore, the present invention has an effect of improving the sound quality by predicting and extending the lower 8 bits, and can easily cope with a conventional CD.

In FIG. 8, the effect of the present invention is described with reference to waveforms. Here, A in Fig. 8 shows the waveform immediately after the start of calculation, and here the quantization noise is generated. On the other hand, B in the figure shows the waveform after 100 ms from the start of the calculation, which shows that the quantization noise is reduced.

Accordingly, in the above-described embodiment, the number of bits of the information signal quantized with a small number of bits, for example, recording on a compact disc, is expanded, and information such as the occurrence of noise is generated during the expansion. As a result, it is possible to provide a signal processing apparatus and a signal processing method that can be directly applied to recording on a conventional compact disc or the like.

Thus, according to the signal processing device of the present invention, the signal processing device expands the number of bits of the quantized signal, and extends the number of bits of the quantized original signal and uses the expanded bits. Filter means for cutting off harmonic components, first calculation means for subtracting the signal power obtained by adding an arbitrary lower bit of the extension to the original signal, and filtering the output of the first calculation means By having the second calculation means for adding to the output of the means, the number of bits can be expanded well.

[0039] Further, according to the signal processing method of the present invention, the signal processing method extends the number of bits of the quantized signal, and extends the number of bits of the quantized original signal. The harmonic component due to the bit is cut off, the signal with the harmonic component cut off is also subtracted from the original signal, and the signal power with any lower bits added to the extension is subtracted, and the subtracted signal is subtracted from the harmonic component. By adding to the blocked signal to obtain a quantized signal with an extended number of bits, a good number of bits can be expanded.

[0040] The present invention is not limited to the above-described embodiment. Various modifications can be made without departing from the scope description. Explanation of quotation marks

DESCRIPTION OF SYMBOLS 1 ... Input terminal, 2 ... Delay means, 3 ... Variable filter, 4 ... Low pass filter (LPF), 5 ... Calculation means, 6 ... Coefficient correction means, 7 ... Delay means, 8 ... Lower bit addition circuit, 9 ... Arithmetic unit, 10 ... Weight circuit, 11 ... Adder circuit, 12 ... Output terminal, 13 ... Peak detection circuit, 14 ... Single pass filter (LPF), 15 ... Limiter circuit

Claims

The scope of the claims

[1] A signal processing device for extending the number of bits of a quantized signal,

Filter means for extending the number of bits of the quantized original signal and blocking harmonic components due to the expanded bits;

First arithmetic means for subtracting the output of the filter means from a signal obtained by adding an arbitrary lower bit of the extension to the original signal;

Second computing means for adding the output of the first computing means to the output of the filter means;

A signal processing apparatus comprising:

[2] In the signal processing device according to claim 1,

The filter means has a variable filter including a delay means and a correction unit for controlling the coefficient thereof,

The correction unit calculates an error ε between the output of the low-pass filter whose cutoff frequency is controlled according to the external signal determined by the tendency of the original signal and the output of the variable filter, and minimizes the error ε. The coefficient of the variable filter is controlled using a correction algorithm

A signal processing apparatus.

[3] The signal processing device according to claim 1,

A peak moving correction average value is calculated with a time constant longer than the quantization sampling with respect to the output of the first calculation means, and the output amplitude of the first calculation means is calculated based on the calculated movement correction average value. Is added to the output of the filter means by the second computing means.

A signal processing apparatus.

[4] In the signal processing device according to any one of claims 1 to 3,

The original signal is a signal obtained by quantizing an acoustic signal.

A signal processing apparatus.

[5] The signal processing device according to claim 2,

The original signal is a signal obtained by quantizing an acoustic signal, The tendency of the original signal is the conversation between the acoustic signal and the music, and the Z or music Giannole.

A signal processing apparatus.

A signal processing method for extending the number of bits of a quantized signal,

Extends the number of bits of the quantized original signal and cuts off harmonic components due to the expanded bits,

Subtract the signal power obtained by adding the arbitrary lower bits of the extension from the original signal to the signal from which the harmonic component is cut off,

The subtracted signal is added to the signal from which the harmonic component is cut off to obtain the quantized signal with the extended number of bits.

And a signal processing method.