JPH06326674A - Composite sound signal generation circuit - Google Patents

Abstract

PURPOSE:To improve the performance of a circuit and to provide simple circuit constitution with less processing amount by constituting a composite sound signal generation circuit by a digital signal processing. CONSTITUTION:The operation sampling frequency of the circuit is set at a sub carrier wave frequency or the power of two and balanced modulation and sub carrier wave signal generation, etc., are realized by the simple processings of a high-pass filter 74, polarity inversion and signal changeover, etc. Further, by using a digital interpolation device, the sampling frequency of input signals is converted and the signals of the optional sampling frequency are made inputtable. Since the circuit scale of the composite sound generation circuit by the digital signal processing can be substantially reduced, the device can be miniaturized, power consumption can be lowered and circuit operations can be stabilized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a composite audio signal generation circuit, and more particularly to a composite audio signal generation circuit for synthesizing stereo audio signals used in FM stereo broadcasting.

[0002]

2. Description of the Related Art In order to transmit a stereo audio signal,
It is necessary to multiplex the left and right two-channel audio signals by some method, and carry them on one carrier. Furthermore, since some receivers do not receive and reproduce stereo signals, it is desired that they be compatible with conventional monaural signals. A device for performing such signal processing is a composite voice signal generation circuit. There are various types of this method, but carrier suppression AM-FM is used in Japan.
The method is adopted. This method, as shown in FIG.
The L + R signal and the L-R signal are created from the (left) signal and the R (right) signal, and the L + R signal is used as the main signal, and 38 kHz
Sub-carrier signal of the carrier suppression AM modulation by the L-R signal and a pilot signal of 19 kHz (3 at the receiver side)
Used to reproduce the 8 kHz subcarrier signal. ) Is a multiplex. Finally, FM modulation is performed on the transmission carrier with this composite audio signal. Thus, stereo FM broadcasting compatible with conventional FM broadcasting waves can be realized.

FIG. 2 shows a conventional example of a composite voice signal generation circuit. In the figure, 21 and 22 are low-pass filters, 23 and 24 are adders / subtractors, 25 is a balanced modulator, 26 is a signal adder, and 27.
Is a subcarrier generator, and 28 is a 1/2 frequency divider. The two systems of audio input L and R signals are band-limited by the low-pass filters 21 and 22. The low-pass filters 21 and 22 also have a pre-emphasis function, and in order to prevent the high-frequency signal component of noise from being emphasized due to FM modulation and preventing the S / N from deteriorating in the high band, the high-frequency component of the audio signal Emphasize. The audio signal that has passed through the low-pass filter is converted into an L + R signal and an LR signal by a matrix circuit including adders / subtractors 23 and 24. The L-R signal is input to the balanced modulator 25 for carrier suppression AM modulation, and the 38 kHz subcarrier signal obtained from the subcarrier generator 27 is balanced-modulated to obtain a subsignal. The frequency of the subcarrier signal is halved by the 1/2 frequency divider 28 to generate a pilot signal of 19 kHz. Signal adder 2 for L + R main signal, LR sub signal, and pilot signal
6, and a composite voice signal is generated.

Known examples of the composite audio signal generating device include Japanese Patent Application Laid-Open No. 1-291536 and Japanese Patent Application Laid-Open No. 2-189038.

[0005]

When the above-mentioned composite audio signal generating circuit is specifically realized, if it is configured by an analog circuit, the precision of the circuit elements is limited, so that a highly accurate matrix circuit, balanced modulator, It was difficult to obtain the pre-emphasis circuit. In addition, it is difficult to avoid the influence of physical fluctuation factors such as temperature, voltage, and aging fluctuation, and there is a drawback that initial adjustment and regular maintenance become complicated. Although these problems can be avoided by introducing digital signal processing technology, digitizing balanced modulation as it is requires a subcarrier generator and a filter with a sharp cutoff characteristic, which makes it difficult to increase the circuit scale. There was Recently, digital audio signals have been digitized to use digital audio signals as input signals to a composite audio signal generation circuit. However, since the sampling frequency of these signals is not a simple integer ratio, some kind of It is necessary to transform the sampling frequency by means.

It is an object of the present invention to provide a circuit configuration which can avoid these problems in a composite voice signal generation circuit using a digital signal processing technique.

[0007]

In order to achieve the above object, the operation sampling frequency of the composite audio signal generation circuit is selected to be a power of 2 of the subcarrier frequency 38 kHz. A digital interpolation circuit using a time-varying coefficient filter is used to convert the sampling frequency of the digital input voice signal.

First, in order to simplify the balanced modulator and the subcarrier signal generator, the relationship of harmonics by sampling a digital signal is used.

The principle of the present invention will be described with reference to FIG. In the figure, 31 is a low pass filter, 32 is a high pass filter, and 33 is a polarity inverter. The operation sampling frequency of the digital filters 31 and 32 is 2 fs, and the polarity inverter 3
3 operates to invert the polarity of the signal into positive and negative for each sample.

The operation when a signal of sampling frequency fs is input to the circuit of FIG. 3 will be described with reference to the operation spectrum diagram of FIG. As shown in FIG. 4 (a), the signal of the sampling frequency fs has a spectral structure in which the spectral frequency components from 0 (DC) to fs / 2 are folded at fs / 2 (this is called Nyquist frequency). have. Further, as shown in FIG. 3, the signal of the sampling frequency fs is converted into the operation sampling frequency 2f.
When input to the digital signal processing device of s, when the sampling frequency is 2fs, the signal spectrum has components 0 to fs repeatedly appearing from fs to 2fs. Therefore, the signal of the sampling frequency fs has the operation sampling frequency of 2f.
s low-pass filter 31 and high-pass filter 32
(Each frequency characteristic is the characteristic shown by the broken line in (b) and (c) of FIG. 4), each filter output is shown in FIG.
As shown by the solid lines in (b) and (c), the output of the low-pass filter is the signal obtained by re-sampling the input signal at 2fs, but the high-pass filter is balanced-modulated with the carrier signal of fs. A signal with the same spectrum as that obtained is obtained.

That is, fs is set to the same value as the subcarrier frequency of the composite voice signal, and the sampling frequency of the input signal is f.
s, if the operation sampling frequency of the signal processing device is doubled, it can be seen that (carrier suppression) balanced modulation can be performed by performing high-pass filtering and extracting harmonic components.

Further, the output signal of the low pass filter 31 is input to the polarity inverter 33. The polarity inverter 33 inverts the polarity of the signal having the sampling frequency of 2fs in the period of fs.
That is, +1, -1, +1, -1, ... Are multiplied. This operation

[0013]

## EQU1 ## cos (2πnfs T) = cos (2πnfs / 2fs) =
It is equivalent to multiplying cos (πn) = + 1, −1, +1, −1, ... In Equation 1, T = 1 / 2fs is the motion sampling period, and n is a positive integer 0, 1, 2, .... That is, the polarity reversal causes the low-pass filter output to have a frequency fs
It means that the cosine wave of is multiplied and the balanced modulation is possible.

Thus, by taking the sampling frequency of the input equal to the subcarrier frequency (or its power of 2), balanced modulation is possible with a very simple signal processing operation. When the input audio signal is an analog signal, the sampling frequency of the AD converter may be selected as fs. However, in recent years, digitization of audio signals has been actively carried out, and the sampling frequencies thereof are 48 kHz, 44.1 kHz, and 32 kHz.
Hz, etc., 38 kHz subcarrier frequency of composite audio signal
And not a simple integer ratio. Therefore, a device capable of converting the sampling frequency with an arbitrary conversion ratio is required. Therefore, a digital interpolation device using a time-varying coefficient FIR (non-cyclic Finite Impulse Response) filter is used.

An interpolator using a time-varying coefficient filter is described in detail in, for example, Japanese Patent Application No. 3-102474.

According to the sampling theorem, as shown in FIG. 5, from the data string f (nT ₁ ) sampled at the period T ₁ ,
The original time function f (t) uses Sinc (t) = sint / t,

[0017]

[Number 2] f (t) = Σf (nT ) Sinc {π (t-nT 1) / T 1} = Σf (nT 1) can be represented as Sc (n, τ). Here, τ = t / T ₁ is a fraction when the output time t is measured in the T ₁ cycle. Equation ₁ indicates that the coupling coefficient Sc (n, τ) is a function of t when a data value at time t is predicted by linear combination of discrete data f (nT ₁ ). The time-varying coefficient Sc (n, τ) is t = nT ₁ and t = m
It is a function that has the property that it becomes 0 at T ₁ (m ≠ n, m, n is an integer), and it is Sinc (t) in Equation 2 or Lagr used in numerical analysis.
Various functions such as ange's interpolation polynomial can be used.

Mathematical Expression 2 is approximated by a finite number of data N, the interpolated value f (t) is FIR having a time varying coefficient Sc (n, τ).
It is shown that it can be obtained as the output of the filter. From this, it is understood that the interpolation (or sampling frequency conversion) can be realized as hardware by the time-varying coefficient filter. Parameters n, τ that determine the time-varying coefficient Sc (n, τ)
Is expressed by Equation 3 by the data output time t given by the sampling period T ₂ of the output data series.

[0019]

[Equation 3] t = nT ₁ + τ = mT ₂ FIG. 6 shows the overall hardware configuration of the interpolator. In the figure, 61 ₁ , 61 ₂ , ..., 61 _N are delay elements, and 62 ₀ , 6
2 ₁ , 62 ₂ , ~, 62 _N-1 , 62 _N are coefficient multipliers, 6
3 ₁ , 63 ₂ , ~, 63 _N-1 , 63 _N is an adder / subtractor, 64 is R
OM, 65 is a clock device, 66 is a counter, and 67 is a latch. The time measuring device 65 for obtaining τ that determines the interpolation time t in the equation 3 inputs a clock pulse sufficiently faster than T ₁ to the counter 66, resets it in the T ₁ cycle, and reads the count value to the latch 67 in the T ₂ cycle. It can be realized.
If the time-varying coefficient Sc (n, τ) is written in the ROM 64 in advance, and the τ obtained is read out and given as the coefficient of the FIR filter, an interpolating device using the time-varying coefficient filter is realized.

[0020]

As described above, in the composite voice signal generation circuit of the present invention, the sampling frequency of the input signal and the operation sampling frequency of the device are balanced by taking the subcarrier frequency of the composite voice signal or its power of two. A complicated circuit such as a modulator or a subcarrier signal generator can be replaced with a simple circuit, which greatly simplifies the configuration. A filter is required for the balanced modulation, but this can be used together with the pre-emphasis filter for the FM signal, and the amount of components does not increase particularly.

Further, since the operation sampling frequency of the device is a power of 2 of the subcarrier signal frequency, the frequency of the pilot signal has the same relationship, and in this case, the configuration of the pilot signal generator is simplified. For example, sampling frequency 7
To generate a pilot signal (19 kHz) at 6 kHz,

[0022]

## EQU00004 ## cos (2.pi.n19 / 76) = cos (n.pi./2) = 1,
Since it is 0, -1, 0, 1, ..., It can be configured by a very simple circuit of the polarity reversing device and the changeover switch.

When a digital audio signal is used as the input signal, it is necessary to convert the sampling frequency. In the conventional sampling frequency conversion, a digital filter having an operation sampling frequency of the least common multiple of the input / output sampling frequency had to be configured, but by using the digital sampling frequency converter as in the present invention, It can be configured without increasing the operation sampling frequency of the device.

[0024]

FIG. 7 is a block diagram of a first embodiment according to the present invention. In the figure, 71 and 72 are adders and subtractors, and 73 and 7
6 is a low pass filter, 74 is a high pass filter, 75
Is a signal adder. FIG. 8 is an operation spectrum diagram of FIG. The operation of the embodiment of FIG. 7 will be described with reference to FIG.

Two-system voice input L of sampling frequency fs,
The R signal is converted into an L + R signal and an L-R signal by a matrix circuit including adders / subtractors 71 and 72. The L + R signal is interpolated by the low-pass filter 31 having the operation sampling frequency 2fs to be a signal having the sampling frequency 2fs as shown in FIG. On the other hand, the L-R signal is passed through the high pass filter 32.
, The sampling frequency is converted to 2fs, and at the same time, the spectrum component around fs is extracted and carrier suppression balanced modulation processing is performed. The low pass filter 31 and the high pass filter 32 can have a pre-emphasis function at the same time as the sampling frequency conversion. The low-pass filter output and the high-pass filter output are combined by the signal adder 75, and a pilot signal of frequency fs / 2 is added,
By removing the harmonic component of the L + R signal around 2fs with the low-pass filter 76, a composite voice signal can be obtained. Figure 7
As can be seen from the above embodiment, according to this embodiment, without using a complicated circuit such as a subcarrier signal generator and a balanced modulator,
A composite audio signal generation circuit can be configured.

A second embodiment of the present invention is shown in FIG. 9 and its operation spectrum diagram is shown in FIG. In FIG. 9, 9
1, 92 and 95 are low-pass filters, 93 and 94 are adders and subtractors, 96 is a polarity inverter, and 97 is a signal adder. The operation of the embodiment shown in FIG. 9 will be described with reference to FIG. The two systems of L and R signals of sampling frequency fs or 2fs are input to low pass filters 21 and 22 and pre-emphasized. When the sampling frequency of the input signal is fs, the low pass filters 21 and 22 also have a sampling frequency conversion function. The outputs of the low-pass filters 21 and 22 are L by a matrix circuit including adders / subtractors 93 and 94.
It is converted into + R signal and LR signal. The L + R signal is passed through the low-pass filter 95 with the operation sampling frequency of 4fs for 2fs.
Remove the surrounding harmonic components. The polarity of the L-R signal is inverted by the polarity inverter 96 every fs to perform balanced modulation. The modulated LR signal, L + R signal, and pilot signal of frequency fs / 2 are combined by the signal adder 97,
A composite audio signal is obtained. In the embodiment of FIG. 9, the frequency characteristic of the low pass filter 95 is the low pass filter 7 of FIG.
There is an advantage that it can be looser than 6.

FIG. 11 shows a third embodiment according to the present invention. In the figure, 110 is a digital interpolator, 111 and 112 are low-pass filters, 113 and 114 are adders and subtractors, and 115.
Is a polarity inverter, 116 is a changeover switch, and 117 is a signal adder. The input signal is a digital stereo audio signal, and signals of two systems of L and R having a sampling frequency fs' are multiplexed. The sampling frequency of the audio signal is converted into 4fs by the digital interpolator 110.

Unlike the configuration example of FIG. 6, the digital interpolator 110 is multiplexed twice, so that the delay element array 61 of FIG.
There are two signals, ₁ to 61 _N , and every two signals are connected to the coefficient multiplier. The L and R signals output from the digital interpolator 110 are input to the low pass filters 111 and 112 having a motion sampling frequency of 4fs and are subjected to pre-emphasis frequency characteristics. Further, adder / subtractor 113, 1
L + R signal, L- by a matrix circuit consisting of 14
The R-R signal is converted into the R signal and the L-R signal is passed through the polarity inverter 115 and the changeover switch 116. The polarity inverter 115 switches the polarity of the signal in the fs cycle, and the changeover switch 116 switches between the polarity inverter output and the ground (logic 0) every 2fs. Since the sampling frequency of the L-R signal x is 4fs, the L-R signal is converted into a sequence of x, 0, -x, 0, x, ... By this operation. This sequence is obtained by multiplying the original L-R signal sequence x by 1,0, -1,0,1, ... Shown in Equation 4, and has a 1/4 (ie, fs) of the operation sampling frequency 4fs. It is equivalent to multiplying a cosine wave signal. In this way, changeover switch 1
A modulated L-R signal is obtained at the output of 16. Signal adder 1
At 17, the L + R signal, the modulated L-R signal, and the pilot signal are combined to obtain a composite voice signal.

The application of the digital interpolator is not limited to the sampling frequency of the input signal of 4 fs as in the embodiment of FIG. 11, and it can be either fs or 2fs as in the embodiments of FIGS. It can be carried out. In any case,
Since the sampling frequency between signals that is not a simple integer ratio can be converted arbitrarily, even if the operation sampling frequency of the composite audio signal generation circuit according to the present invention is set to a power of 2 of the subcarrier frequency, any sampling can be performed. It is possible to input a signal having a digitized frequency.

The embodiments of the present invention applied to an apparatus for generating a composite audio signal by digital processing have been described above. In an actual configuration, it can be realized by a general-purpose digital signal processing technique and does not require a special function. Therefore, it can be realized by hardware by a logic circuit or software by a general-purpose digital signal processor.

[0031]

According to the present invention, a balanced modulator and a sub-carrier signal generator having a complicated structure by setting the relationship between the sampling frequency of the signal and the sampling frequency of the sub-carrier signal to be a power of two. ， The pilot signal generator can be realized with a simple circuit, and the scale of the device can be greatly reduced. Also, when the digital input signal sampling frequency is different from the subcarrier sampling frequency, the sampling frequency conversion can be easily performed using a digital interpolator. Is not a constraint. Further, since it is digital processing, it operates stably against various physical fluctuation factors, labor saving in initial adjustment, periodic maintenance, etc., and change in circuit constants can be easily dealt with.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a composite audio signal used in FM broadcasting.

FIG. 2 is a conventional composite voice signal generation circuit diagram.

FIG. 3 is an explanatory diagram of the principle of balanced modulation according to the present invention.

FIG. 4 is an operation spectrum diagram illustrating the operation of FIG.

FIG. 5 is an explanatory diagram of the principle of digital interpolation used in the present invention.

FIG. 6 is a block diagram of a digital interpolator.

FIG. 7 is a block diagram of the first embodiment of the present invention.

FIG. 8 is an operation spectrum diagram of the embodiment of FIG.

FIG. 9 is a circuit diagram of a second embodiment of the present invention.

FIG. 10 is an operation spectrum diagram of the embodiment of FIG.

FIG. 11 is a circuit diagram of a third embodiment of the present invention.

[Explanation of symbols]

71, 72 ... Adder / subtractor, 73, 76 ... Low-pass filter, 74 ... High-pass filter, 75 ... Signal adder.

Claims (6)

[Claims] 1. In a composite audio signal generation circuit for multiplexing two systems of audio signals, a left audio signal and a right audio signal, and converting into a composite audio signal, the operation sampling frequency of the circuit is a sub-signal of the composite audio signal generation circuit. A composite audio signal generation circuit, which has the same frequency as a frequency fs of a carrier signal or a power of 2. 2. The composite audio signal according to claim 1, wherein an input signal having a sampling frequency equal to the subcarrier frequency is processed by a high-pass filter operating at an operating sampling frequency twice the subcarrier frequency. Generation circuit. 3. The polarity of an audio signal sampled at a sampling frequency twice the subcarrier frequency as defined in claim 1.
A composite audio signal generation circuit that inverts at a cycle of / fs. 4. The polarity of an audio signal sampled at a sampling frequency which is four times as high as the subcarrier frequency, according to claim 1.
A composite audio signal generation circuit which inverts at a cycle of / fs and further inserts "0" data at a cycle of 1/2 fs. 5. The digital sampling frequency conversion circuit according to claim 1, 2, 3 or 4, wherein the sampling frequency of the input signal is converted to the same frequency as the operation sampling frequency of the composite audio signal generation circuit. Complex audio signal generation circuit. 6. The digital sampling frequency conversion circuit according to claim 5, wherein a timekeeping device that is periodically initialized by a sampling pulse signal of a first sampling frequency,
An input signal sampled at the first sampling period is measured by measuring the time of a sampling pulse having a second sampling frequency and using a time-varying coefficient filter having a filter coefficient determined by the second sampling time. A composite audio signal generation circuit using a sampling frequency conversion circuit for converting a sequence into an output signal sequence resampled at a second sampling period.