JP3403925B2 - Signal transmission method via 1-bit digital signal, delta-sigma modulation circuit, and demodulation circuit - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a signal transmission method using delta sigma modulation, which is particularly preferably used for audio signal processing, a delta sigma modulation circuit, and a delta sigma modulation generated signal. The present invention relates to a bit digital signal demodulation circuit.

[0002]

2. Description of the Related Art Conventionally, as a method for transmitting a digital signal, a multi-bit encoding method for transmitting one word consisting of a plurality of bits as a break and a delta-sigma modulation are used to encode a 1-bit digital signal. A transmission method is known.

In the case of the multi-bit encoding method, the transmitting or recording side stores the data in 1 according to a predetermined format.
Encode into words. On the other hand, the receiving or reproducing side synchronizes the words to determine the delimiter of each word, and decodes each word to identify the data. Therefore, both sides need a signal processing circuit that performs signal processing according to the word format. As a result, once the word format has been determined and the sampling frequency, dynamic range, etc. have been standardized, it is difficult to change the standard. Furthermore, since the method requires word synchronization,
An error correction circuit is indispensable for correcting an error that occurs because it is easily affected by the transmission line and the like.

On the other hand, in the 1-bit digital encoding system, since the 1-bit digital signal is a flow of finely-divided data that does not require word synchronization, it is less susceptible to the influence of the transmission line and the like, and it is not susceptible to errors. It has the advantage of being strong.
Therefore, in this method, no error correction circuit is required in both the transmitting or recording device and the receiving or reproducing device. Further, when the 1-bit digital signal is an acoustic signal, the receiving or reproducing side can demodulate the 1-bit digital signal into an analog signal by a simple low-order low-pass filter, so that a complicated processing circuit for demodulation is unnecessary. Become. Therefore, in recent years, the 1-bit digital coding system, which has many advantages over the multi-bit coding system, has been attracting attention.

As shown in FIG. 8, in a typical conventional delta-sigma modulation circuit 100, an analog acoustic signal input from an input terminal 101 is integrated by cascaded integrators m101 to m107. . The integrator output of each stage is added by the adder 103, and then the quantizer 104
Entered in. The quantizer 104 derives an output of “1” at the output terminal 106 when the output of the adder 103 is 0 or more, and derives an output of “0” when the output of the adder 103 is less than 0. The output of the quantizer 104 is negatively fed back to the input side of the integrator m101 of the first stage via the digital / analog converter 105 and the feedback resistor r100.

On the other hand, in order to form a dip in the quantization noise distribution (noise floor) of the 1-bit digital signal output from the delta-sigma modulation circuit 100 and adjust the quantization noise distribution shape to a desired shape, the delta The integration circuit 102 of the sigma modulation circuit 100 includes three feedback circuits m111.
~ M113 are provided. The feedback circuit m111 outputs the output of the third-stage integrator m103 to the second-stage integrator m1.
Negative feedback to the input side of 02, feedback circuits m112 and m1
13 is the fifth and seventh stage integrators m105 and m10
7 is output to the fourth and sixth stage integrators m104 · m
Negative feedback is provided to the input side of 106.

These feedback circuits m111 to m113 form three partial negative feedback loops, and the quantization noise level of the 1-bit digital signal is centered on the frequency (zero point frequency) according to the gain of each partial negative feedback loop. Falls sharply. In the following, of the frequency characteristics of the quantization noise, the part where the level is lowered is referred to as a dip.
By these dips, the quantization noise in the high frequency band is suppressed, and the level of the quantization noise can be kept below a predetermined value up to the upper limit of the desired frequency band used, such as 20 kHz.

In the delta-sigma modulation circuit 100, after the acoustic signal is modulated into a 1-bit digital signal, the 1-bit digital signal is received by a not-shown receiving or reproducing apparatus, for example, by a low-order low-pass filter or the like. Demodulated to an analog acoustic signal.

[0009]

By the way, in the transmission of digital signals, for example, a main signal such as an acoustic signal,
For example, there is a case where it is required to simultaneously transmit both a sub signal such as a flag indicating channel information. However, since the 1-bit digital signal transmitted by the 1-bit digital encoding method is a data string that is continuous in the time domain method, it does not interfere with the advantage of being easy to demodulate, and the main signal and the sub signal It is difficult to transmit both of these.

In the conventional multi-bit encoding method, when both the main signal and the sub signal are transmitted, the main signal and the sub signal are temporally divided and transmitted. Here, in the multi-bit encoding method, word synchronization is required when returning an encoded data string to an analog signal. Therefore, the modulation circuit modulates the main signal and the sub signal in a data format such that the main signal and the sub signal are included in one word, and the demodulation circuit adjusts the main signal and the sub signal in accordance with the standardized data format. If the signals are separated, both the main signal and the sub signal can be temporally divided and transmitted.

Since word synchronization is essential in the multi-bit encoding system, signals other than the main signal are transmitted even if the sub-signal is not transmitted, and the demodulation circuit
It has a circuit for separating the main signal according to a predetermined data format. Therefore, when transmitting the sub-signal, simply change the data format of the circuit,
Since the auxiliary signal can be a circuit which separation the main signal, there is no necessity of adding a new circuit.

On the other hand, in the 1-bit digital encoding system, a continuous data string is transmitted in the time axis direction. Therefore, in order to divide and transmit the main signal and the sub-signal in the time axis direction, a circuit for separating the main signal and the sub-signal according to a standardized data format at the time of modulation is used as a demodulation circuit. Must be newly added, and the circuit configuration of the demodulation circuit becomes complicated. In addition, once the data format is standardized, it becomes difficult to change the data format. Therefore, dividing and transmitting in the time axis direction impairs the advantages of the 1-bit digital encoding method.

The present invention has been made in view of the above problems, and an object thereof is to be able to transmit both a main signal and a sub signal without impairing the advantages of the 1-bit digital encoding system. It is to provide a signal transmission method via a 1-bit digital signal, a delta-sigma modulation circuit, and a demodulation circuit.

[0014]

[Means for Solving the Problems] 1 according to the invention of claim 1
In order to solve the above-mentioned problems, a signal transmission method using a bit digital signal includes a step of modulating a main signal by delta-sigma modulation to a 1-bit digital signal, and the above-mentioned 1-bit via a transmission line or a recording medium. A method of transmitting a signal via a 1-bit digital signal including a step of transmitting a digital signal and a demodulation step of demodulating the transmitted 1-bit digital signal is characterized in that the following steps are provided.

That is, the modulation step includes a step of changing the shape of the quantization noise distribution of the main signal based on the sub signal, and the demodulation step includes the quantization of the transmitted 1-bit digital signal. It is characterized by including a step of extracting a side signal based on the shape of the noise distribution.

The shape of the quantization noise distribution of the 1-bit digital signal is changed, for example, by changing the coefficient of a multiplier provided between cascaded integrators or changing the loop gain of the partial negative feedback loop. Can be changed.

In the above configuration, the sub-signal is added as the shape of the quantization noise distribution of the 1-bit digital signal.
Therefore, the main signal can be demodulated as in the case where the sub signal is not added. As a result, both the main signal and the sub signal can be transmitted as a 1-channel 1-bit digital signal without disturbing the characteristic of delta-sigma modulation that demodulation is easy.

In addition, since the sub-signal is the quantization noise itself of the 1-bit digital signal, the main signal and the sub-signal exist in the same region in terms of time and frequency in the 1-bit digital signal. To do. Therefore, it is possible to prevent a third party from removing, changing, and falsifying the sub-signal.

Further, in order to solve the above-mentioned problems, the delta-sigma modulation circuit according to the invention of claim 2 is a sub-signal in a delta-sigma modulation circuit for delta-sigma modulating a main signal to generate a 1-bit digital signal. The sub-signal adding means for changing the shape of the quantization noise distribution of the 1-bit digital signal and the input signal to be the main signal are input to the first stage according to the above.
Multiple integrators connected in series with each other and each of the above products
The adder that adds the output of the divider and the output of the above adder
Quantizer that outputs a 1-bit digital signal by subdividing
And the output of the above integrator,
It has a partial negative feedback loop for negative feedback to the input side of
According to the sub-signal, the sub-signal adding means
Change the loop gain of the negative feedback loop.
It is characterized in that it is provided with a further means .

Furthermore, the delta-sigma modulation circuit according to the invention of claim 3 is the structure of the invention of claim 2, wherein, above
The partial negative feedback circuit that constitutes the partial negative feedback loop is
Including a differential amplifier provided between both integrators,
Signal adding means, in accordance with the sub-signal,
By changing the multiplier coefficient
It is characterized by changing .

In the structure according to the invention of claim 2 ,
The quantization noise level of the 1-bit digital signal generated by the delta-sigma modulation circuit sharply decreases around the frequency (zero point frequency) according to the loop gain.

When the sub-signal changes, the loop gain changing means has a function as in the configuration of the invention according to claim 3, for example.
The loop gain of the partial negative feedback loop is changed by, for example, changing the multiplier coefficient of the partial negative feedback circuit forming the partial negative feedback loop. As a result, the quantization noise level of the 1-bit digital signal decreases sharply around the zero-point frequency different from that before the change of the sub-signal, and the sub-signal adding means can largely change the shape of the quantization noise distribution. The zero frequency is
Since it is determined by the loop gain, if the loop gain is set to a predetermined value, the quantization noise distribution of the 1-bit digital signal changes to a predetermined shape. As a result, the delta-sigma modulation circuit can generate a 1-bit digital signal in which the sub-signal can be more easily discriminated on the demodulation side.

Therefore, the sub-signal adding means is the above-mentioned claim 2.
Of the partial negative feedback loop in the configuration according to the invention
By changing the gain, the invention according to claim 3 is achieved.
In such a configuration, it is provided between cascaded integrators.
By changing the multiplier coefficient of the differential amplifier,
Shape of quantization noise distribution of bit digital signal as sub-signal
Change accordingly.

Therefore, the delta-sigma modulation circuit is
Signal demodulation is easy, and the main and side signals are on a single channel
Can be transmitted by a third party, and a third party can remove, change, or
It is possible to generate a 1-bit digital signal that is difficult to tamper with.

In addition, in the delta-sigma modulation circuit according to the invention of claim 4, in the configuration of the invention of claim 2 or 3, the main signal is an acoustic signal and the sub-signal is
It is characterized by being a signal indicating at least one of channel information, presence / absence of pre-emphasis, a flag for copyright protection, and a mastering code.

In the above structure, each piece of information serving as a sub signal is information that is closely related to an acoustic signal serving as a main signal and has a small amount of information. Therefore, even when the sub-signal adding means cannot change the shape of the quantization noise distribution of the 1-bit digital signal into too many shapes, the delta-sigma modulation circuit can add the information as a sub-signal without any trouble. . In addition, the sub-signal is removed from the 1-bit digital signal generated by the delta-sigma modulation circuit having the above configuration,
Since it is difficult to change and tamper with, it is possible to reliably protect the above-mentioned information, which is closely related to the main signal and which causes great damage when removed.

On the other hand, the demodulation circuit according to the invention of claim 5 is
In order to solve the above problems, in a demodulation circuit that demodulates a 1-bit digital signal generated by subjecting a main signal to delta-sigma modulation, the quantization noise distribution of the 1-bit digital signal has a shape depending on the sub-signal. And a sub-signal discriminating means for discriminating the added sub-signal by comparing the pre-stored shape of the quantization noise distribution with the shape of the quantization noise distribution of the received 1-bit digital signal. It is characterized by

In the above structure, when the demodulation circuit receives the 1-bit digital signal, the sub-signal discriminating means determines the shape of the quantization noise distribution of the 1-bit digital signal and the shape of the previously stored quantization noise distribution. The comparison is performed, and the sub-signal is extracted based on the difference between the two. This allows the demodulation circuit to correctly extract the sub-signal from the received 1-bit digital signal. Furthermore, by storing the quantization noise distribution shape serving as a comparison reference in a storage medium such as an IC (Integrated Circuit) that cannot be read by a third party, the side signal can be removed, changed, and changed by the third party. Tampering can be securely prevented.

The demodulation circuit according to the invention of claim 6 is
In order to solve the above problems, in a demodulation circuit for respectively demodulating a 1-bit digital signal of a plurality of channels generated by delta-sigma modulating a main signal of a plurality of channels,
The shape of the quantization noise distribution of the 1-bit digital signal of at least one specific channel of the respective channels is determined according to the sub-signal, and the quantization noise distribution of the received 1-bit digital signal of the specific channel is received. Is compared with the shape of the quantization noise distribution of the 1-bit digital signal of another channel, and a sub-signal discriminating means for discriminating the added sub-signal is provided.

In the above structure, the demodulation circuit is, for example,
Receives 1-bit digital signals of multiple channels, such as left and right channels. Here, if the shape of the quantization noise distribution is changed in the 1-bit digital signal of the specific channel, the shape of the quantization noise distribution is different from the shape of the quantization noise distribution of other channels. The sub-signal discriminating means is a 1-bit digital signal of a specific channel and a 1-bit digital signal of another channel.
The shape of the quantization noise distribution is compared with that of the bit digital signal, and the side signal is extracted based on the difference. Further, unlike the configuration according to the fifth aspect of the invention, it is not necessary to store the shape of the quantization noise distribution in advance, so that the configuration of the demodulation circuit can be simplified. With this, with a relatively simple configuration,
It is possible to provide a demodulation circuit that can correctly extract a sub-signal from the received 1-bit digital signal.

Further, in the demodulation circuit according to the invention of claim 7, in the structure of the invention according to claim 5 or 6, the sub-signal discriminating means is one of the quantization noise distribution of the received 1-bit digital signal. The feature is that the sub-signal is discriminated based on the shape in a predetermined specific frequency band.

The method of limiting to a specific frequency band may be, for example, extracting only the component of the specific frequency band from the received 1-bit digital signal using a bandpass filter or the like, or Fourier transform. You may extract only the component of a specific frequency band using such calculation.

In this configuration, since the frequency band to be compared is limited, the comparison becomes easier as compared with the case where the shapes of the quantization noise distributions are compared over all the frequency bands. Particularly, as in the configuration of the invention described in claim 5,
The area required for storage can be reduced as compared to the case where the shape of quantization noise that serves as a comparison reference is stored in advance.

[0034]

BEST MODE FOR CARRYING OUT THE INVENTION [First Embodiment] The following will describe one embodiment of the present invention with reference to FIGS. 1 to 5.

That is, the acoustic signal transmitting apparatus according to this embodiment is an apparatus for transmitting an acoustic signal as a main signal,
As shown in FIG. 2, left and right acoustic signal sources 2R
A demodulation device 3 for delta-sigma modulating an analog or multi-bit acoustic signal output by L into a 1-bit digital signal and a 1-bit digital signal received via a transmission path 4R or 4L such as an optical fiber. The demodulator 5 is provided. The modulator 3 corresponds to the delta-sigma modulator described in the claims, and the demodulator 5 corresponds to the demodulator.

In the modulator 3, the acoustic signal from the right channel acoustic signal source 2R is delta-sigma modulated by the delta-sigma modulation circuit 31 and sent to the transmission line 4R as a 1-bit digital signal. On the other hand, in the demodulation device 5, the 1-bit digital signal received from the transmission path 4R is demodulated into an acoustic signal by the demodulation circuit 51R and a low-pass filter (hereinafter abbreviated as LPF) 52R. Similarly in the left channel, the acoustic signal from the acoustic signal source 2L is delta-sigma modulated into a 1-bit digital signal by the delta-sigma modulation circuit 32, and then the transmission line 4L.
Demodulation circuit 51L / LP of the demodulation device 5 transmitted through the
F52L demodulates to an acoustic signal. The acoustic signals of both channels demodulated by the demodulation device 5 are sent to the amplifier 6
It is sonicated by the left and right channel speakers 7R and 7L via the R and 6L. In the following, when referring to each member, when the left and right are not particularly distinguished, or when both are collectively referred to, the alphabetic character (R or L) added to the end of the reference numeral is omitted, and Reference as Source 2.

Further, in the acoustic signal transmission device 1 according to the present embodiment, in order to accurately discriminate and demodulate and output the acoustic signals of the left and right channels with respect to the switching of the transmission paths 4L and 4R, etc. The 1-bit digital signal output by the device 3 is added with channel information indicating which of the left and right channels the audio signal is, as a sub-signal added to the main signal.

Specifically, the modulator 3 includes either a delta-sigma modulation circuit 31 for delta-sigma modulating the right channel acoustic signal or a delta sigma modulation circuit 32 for delta sigma modulating the left channel acoustic signal. An additional information signal generating circuit (sub-signal adding means) 10 connected to the delta-sigma modulation circuit 31 of one channel (here, the right channel) is provided, and 1 in the right channel is provided in accordance with the channel information of the sub-signal. The shape of the quantization noise distribution of the bit digital signal can be changed.

In the following, for convenience of explanation, a delta-sigma modulation circuit having an integral order of 7th order and three partial negative feedback loops will be described as an example. However, the integral order and the number of partial negative feedback loops are used for various purposes. It can be freely set according to.

As shown in FIG. 1, the delta-sigma modulation circuit 31 integrates the analog acoustic signal input from the acoustic signal source 2 to the input terminal 21 in a high order.
And an adder 23 for adding integrated outputs of the respective orders, and an adder 2
A quantizer 24 that quantizes the output of 3 to output a 1-bit digital signal; and a digital / analog converter 25 that converts the output of the quantizer 24 into an analog value and feeds it back to the integration circuit 22. Is equipped with.

The quantizer 24 samples the output of the adder 23 at a predetermined sampling frequency FS, derives an output of "1" when the output is 0 or more, and outputs an output of "0" when the output is less than 0. Derive the output. As a result, a 1-bit digital signal having the sampling frequency FS is output from the output terminal 26.

In the 1-bit digital coding method for sampling a multi-bit digital signal at high speed, the quantizer 2
The sampling frequency FS of 4 is usually set to a predetermined multiple of fs, such as 32fs or 64fs, where fs is the sampling frequency of the multi-bit digital signal. Here, as in the case of a compact disc, f
When s = 44.1 kHz, FS is 1.42 MHz at 32 fs and 2.82 MH at 64 fs.
z.

On the other hand, the integrating circuit 22 is a feedback resistor provided between the 7th-order integrators m1 to m7 connected in cascade and the input side of the first-stage integrator m1 and the digital / analog converter 25. and r0. The feedback resistor r0 is connected to the inverting input terminal of the differential amplifier a1 described later.

The first-order integrator m1 is a differential amplifier a1.
And a capacitor c1 which is a time constant element provided between the input and output of the differential amplifier a1, and an input resistance r1 provided between the input of the integrator m1 and the inverting input terminal of the differential amplifier a1. Is equipped with. The non-inverting input terminal of the differential amplifier a1 is grounded. The output from the differential amplifier a1 is used as the output of the integrator m1 and is output from the integrator m2 of the next stage.
Is input to the adder 23.

The second and subsequent integrators m2 to m7 are also constructed in the same manner, and the reference numerals of the corresponding parts are the same alphabetic characters with a suffix corresponding to the order of each integrator m2 to m7. Shows. For example, in the third-order integrator m3, the output of the integrator m2 is input via the input resistor r3, and the output of the differential amplifier a3 is input to the next-stage integrator m4 and the adder 23.

Further, the integrating circuit 22 according to the present embodiment constitutes a partial negative feedback loop, and changes the loop gain of the partial negative feedback loop based on the instruction of the additional information signal generating circuit 10. Gain changing circuit (partial negative feedback circuit; loop gain changing means) 11a to 1
1c is provided.

Specifically, the loop gain changing circuit 1
1a is provided across the second-order integrator m2 and the third-order integrator m3, and the output of the integrator m3 can be negatively fed back to the input side of the integrator m2. Similarly, a loop gain changing circuit 11b is provided in association with the fourth-order integrator m4 and the fifth-order integrator m5,
A loop gain changing circuit 11c is provided in association with the sixth-order integrator m6 and the seventh-order integrator m7. By these loop gain changing circuits 11a to 11c,
Three partial negative feedback loops are formed in the integrating circuit 22. For example, in the partial negative feedback loop formed by the loop gain changing circuit 11a, the output of the integrator m2 is integrated and inverted by the integrator m3, further forwardly rotated by the loop gain changing circuit 11a, and then the integrator m2. Is negatively fed back to the non-inverting input terminal of the differential amplifier a2 provided in the above.

With these three partial negative feedback loops,
As shown in FIG. 3, three dips are formed in the frequency characteristic of the quantization noise level of the 1-bit digital signal. The center frequency (zero point frequency) f of the dip is determined by the loop gain Gp of each partial negative feedback loop, and f ≅FS × (Gp) ^1/2 / 2π (1 ). In the above formula (1), FS is the sampling frequency of the delta-sigma modulation circuit 31. In this way, the shape of the quantization noise distribution can be changed by lowering the quantization noise level of the 1-bit digital signal at each zero point frequency.

The gain Gp of the partial negative feedback loop is determined by the multiplier coefficient of the differential amplifier forming the partial negative feedback loop. For example, the gain Gp of the partial negative feedback loop formed by the loop gain changing circuit 11a is the difference between the multiplier coefficient of the differential amplifier a2, the multiplier coefficient of the differential amplifier a3, and the loop gain changing circuit 11a. It is determined by the product of the multiplier coefficient of the dynamic amplifier a11 (described later).

Here, the loop gain changing circuit 11a.
Is, for example, as shown in FIG.
One end of the output side is connected to the inverting input terminal of the differential amplifier a11, and the input resistance section ri whose resistance value can be changed according to the instruction of the additional information signal generating circuit 10 and the input / output of the differential amplifier a11. A feedback resistor rf11 provided between the output resistor ro and one end of which is connected to the output of the differential amplifier a11.
11 and 11. Further, one end of the input resistor portion ri on the input side is connected to the input of the loop gain changing circuit 11a, that is, the output of the integrator m3, and the other end of the output resistor ro11 is connected to the loop gain changing circuit 11a. Output, that is, the differential amplifier a2 provided in the integrator m2
It is connected to the inverting input terminal of. The differential amplifier a
The non-inverting input terminal 11 is grounded.

In the input resistance section ri, for example, a plurality of input resistances ri1, ri2 and ri3 connected in parallel to each other and individual contacts on the input side of each of the input resistances ri1 to ri3 are connected to change the loop gain. An input side switch si having a common contact connected to the input of the circuit 11a, that is, the output of the integrator m3, and the input resistors ri1 to ri.
3 to the output side of each of the individual contacts, the differential amplifier a
An output side switch so having a common contact connected to the inverting input terminal 11 is provided. Both switches si ・ s
o operates in conjunction with the control signal given from the additional information signal generating circuit 10. As a result, the resistance value of the input resistance unit ri is selected to be the resistance value according to the instruction from the additional information signal generation circuit 10.

In the above configuration, the additional information signal generation circuit 10
Need only control the switches si.so of the loop gain changing circuits 11a to 11c based on the channel information. Therefore, the additional information signal generation circuit 10
It is not necessary to generate a frequency-modulated signal, an amplitude-modulated signal, etc., and can be realized by a relatively simple circuit such as a logic circuit.

Here, the multiplier coefficient of the loop gain changing circuit 11a is the resistance value of the input resistance portion ri and the feedback resistance rf1.
And the resistance value of 1. Therefore, the loop gain changing circuit 11a can change the multiplier coefficient according to the instruction of the additional information signal generating circuit 10, and can change the loop gain Gp of the partial negative feedback loop formed by the loop gain changing circuit 11a. Since the configurations of the remaining loop gain changing circuits 11b and 11c are the same as the configurations of the loop gain changing circuit 11a, description thereof will be omitted.

The quantization noise distribution shape of the 1-bit digital signal that can be set by the delta sigma modulation circuit 31 is determined by the combination of the loop gains of the partial negative feedback loops. These quantization noise distribution shapes are It is selected so as to satisfy the following conditions. That is, each quantization noise distribution shape is selected such that the dynamic range of the acoustic signal that is the main signal can be secured and the demodulation side can distinguish each other.

For example, to list the conditions required for the current consumer digital audio equipment, the acoustic signal which is the main signal is in the frequency band of 10 kHz to 20 kHz.
It is required to maintain S / N of about 90 to 100 dB. Therefore, the gain Gp of each of the above partial negative feedback loops
As shown in FIG. 3, for example, in a region of 20 kHz or less, a desired dynamic range (for example, 90 d
The size is set so that (B) can be secured.

The acoustic signal transmission device 1 according to this embodiment is
In order to simplify the configuration of adding channel information, channel information is added only to the 1-bit digital signal of the right channel and not added to the 1-bit digital signal of the left channel. That is, the left channel delta-sigma modulation circuit 3
Reference numeral 2 directly modulates the analog acoustic signal from the acoustic signal source 2L by delta sigma. Specifically, as shown in FIG. 5, the delta sigma modulation circuit 32 is the loop gain changing circuit 1 of the delta sigma modulation circuit 31 shown in FIG.
Feedback circuits m11 to m13 are provided in place of 1a to 11c. Each of the feedback circuits m11 to m13 has a constant gain regardless of the instruction from the additional information signal generating circuit 10, except that the loop gain changing circuit 11 shown in FIG.
It has the same configuration as a. Specifically, for example, the feedback circuit m
11, the input of the feedback circuit m11 is
It is connected to the inverting input terminal of the differential amplifier a11 via the input resistor ri11. As a result, in the delta-sigma modulation circuit 32, the loop gain of each partial negative feedback loop becomes constant regardless of the channel information. Therefore, the delta-sigma modulation circuit 32 outputs a 1-bit digital signal having a constant quantization noise distribution shape. Since the remaining members are the same as those of the loop gain changing circuit 11a, the members having the same functions are designated by the same reference numerals and the description thereof will be omitted.

Thus, in the modulator 3 shown in FIG. 2, the delta-sigma modulation circuit 31 delta-sigma-modulates the acoustic signal output from the acoustic signal source 2R and changes the shape of the quantization noise distribution. , Channel information indicating the right channel is added. As a result, the 1-bit digital signal to which the channel information is added is output from the output terminal of the right channel. These 1-bit digital signals of the left and right channels are transmitted to the demodulation device 5 via the transmission lines 4R and 4L. In this embodiment,
Channel information is not added to the 1-bit digital signal output from the output terminal on the left channel side.

On the other hand, the demodulation device 5 includes demodulation circuits 51R and 51L and LPFs 52R and 52 for demodulating the left and right channel 1-bit digital signals into analog acoustic signals.
L and a channel switching circuit 53 for selecting whether to switch the left and right channels when outputting the demodulated acoustic signal to the amplifiers 6R and 6L for the left and right channels.

Each of the demodulation circuits 51 is realized by, for example, a low pass filter or the like. In this case, the cutoff frequency of the low-pass filter is set to the upper limit frequency Ft of the transmission band in which the 1-bit digital signal can be transmitted. As a result, the 1-bit digital signal is modulated into an analog signal. It should be noted that the low-pass filter 52 only needs to be able to remove noise components in the higher frequency band than the effective frequency band of the acoustic signal. Therefore, in particular, rather than a higher order filter,
The following filters are sufficient. In this case, for example, it can be realized by one resistor and one capacitor.

Further, the cutoff frequency of each low-pass filter 52 arranged after each demodulation circuit 51 is set to the upper limit frequency Fa of the band (audible band) for transmitting the acoustic signal in the above-mentioned transmission band. As a result, in each low-pass filter 52, the acoustic signal that is the main signal is extracted from the analog signal and input to the channel switching circuit 53.

Here, in the high-speed sampling 1-bit encoding method, if the sampling frequency is FS, FS /
It is known that 2 is the upper limit frequency Ft of the transmission band and FS / 6 is the upper limit frequency Fa of the frequency band that can be used as the acoustic band.

For example, if FS = 32fs and fs is 44.1 kHz as in the case of a compact disc, then Ft = 32 × 44.1 / 2 = 705.6 [kHz] (2) Fa = 32 × 44 .1 / 6 = 235.2 [kHz] (3)

However, when the circuit is actually implemented by hardware, it is difficult to sufficiently reduce the quantization noise in the frequency band up to the upper limit frequencies Ft and Fa. Therefore, the S / N condition required by the current consumer digital audio equipment, that is, 10 to 20 k
The upper limit frequencies Ft, F are set so that the S / N at Hz of about 90 to 100 dB can be realized relatively easily.
The realistic value of a is about 1/2 to 1/4 of those values. Specifically, for example, Fa is set to about 50 kHz and Ft is set to about 120 kHz. When the sampling frequency FS is increased to 64fs, the upper limit frequencies Fa and Ft are 100kH, respectively.
z, about 240 kHz.

Further, the channel switching circuit 53 is
For example, a relay or analog switch
Switches s1 and s2 for selecting one of two outputs and outputting one input are provided. Switch s
The common contact of No. 1 is connected to the LPF 52R, and the common contact of the switch s2 is connected to the LPF 52L. Further, one individual contact of the switch s1 and one individual contact of the switch s2 are commonly connected to the amplifier 6 of the right channel.
It is connected to the speaker 7R via R. Similarly, the remaining individual contacts of both switches s1 and s2 are commonly connected to the speaker 7L via the left channel amplifier 6L. The switches s1 and s2 are interlocked with each other and switched in response to an instruction from a channel discrimination circuit 61 described later. Thereby, the demodulation device 5 can select whether to switch the left and right channels when outputting the acoustic signals to the amplifiers 6R and 6L for both channels.

Further, the demodulation device 5 according to the present embodiment is connected to the outputs of the demodulation circuits 51R and 51L of each channel, and based on the channel information added to the 1-bit digital signal of the corresponding channel, Channel discriminating circuits (sub signal discriminating means) 61R and 61L for controlling the switching circuit 53 are provided. Each channel discrimination circuit 61 extracts the added channel information based on the shape of the quantization noise distribution of the 1-bit digital signal, and when the channel information is different from the normal one, switches the left and right channels to the channel switching circuit 53. Can be instructed to.

Each channel discrimination circuit 61 according to the present embodiment.
Stores in advance the shape of the quantization noise distribution serving as the discrimination criterion, and compares the shape of the received quantization noise distribution of the received 1-bit digital signal with the stored shape using, for example, Fourier transform. , The channel information added to the 1-bit digital signal is determined based on the difference between the two shapes.

For example, in the case where the sub-signal is not added, that is, the shape of the quantization noise distribution of the 1-bit digital signal when the left channel is shown is stored in each channel discrimination circuit 61 as a determination criterion, Each of the channel determination circuits 61 determines that the received 1-bit digital signal is the left channel when there is no difference between the two shapes above a certain level. On the other hand, when a difference is found,
Each channel determination circuit 61 determines that the channel information added to the 1-bit digital signal indicates the right channel.

In the above configuration, the operation of the entire acoustic signal transmission device 1 when the acoustic signal which is the main signal and the channel information which is the sub signal are simultaneously transmitted will be described below.

That is, in the modulator 3, the left-channel delta-sigma modulation circuit 32 has a predetermined quantization noise distribution, for example, as shown by the solid line in FIG. It outputs a 1-bit digital signal having a shape. On the other hand, when the 1-bit digital signal transmitted by the acoustic signal transmission device 1 is a stereo signal, the additional information signal generation circuit 10 changes the shape of the quantization noise distribution of the 1-bit digital signal so that the right channel is changed. To the delta-sigma modulation circuit 31. As a result, the delta-sigma modulation circuit 31 outputs a 1-bit digital signal having a quantization noise distribution shape different from that of the left channel, as indicated by the broken line in FIG.

On the other hand, when the correct channel 1-bit digital signal is input to the demodulator 5, the channel discrimination circuit 61R outputs the right-channel 1-bit digital signal,
That is, the additional information signal generation circuit 10 inputs a 1-bit digital signal whose quantization noise distribution shape has been changed, and the channel discrimination circuit 61L receives a left-channel 1-bit digital signal.

Therefore, the channel discrimination circuit 61R is
Since a difference is recognized between the shape of the quantization noise distribution of the received 1-bit digital signal and the shape stored in advance, it is determined that the 1-bit digital signal is the right channel, that is, the correct channel. Similarly, since the channel discrimination circuit 61L does not find a certain difference or more between the shape of the quantization noise distribution of the received 1-bit digital signal and the shape stored in advance, the 1-bit digital signal is The left channel, that is, the correct channel is determined. Therefore, the both-channel discrimination circuit 6
1R / 61L instructs the channel switching circuit 53 not to switch the left and right channels.

As a result, the right channel demodulation circuit 51R
The acoustic signal demodulated by the LPF 52R is output to the right channel amplifier 6R and is acoustically converted by the speaker 7R. Similarly, the left channel demodulation circuit 51L / LPF5
The acoustic signal demodulated by 2L is transmitted to the speaker 7L via the amplifier 6L for the left channel to be sonicated.

On the other hand, for example, the transmission lines 4R and 4L
When a 1-bit digital signal of a channel different from the normal channel is input to the demodulator 5 due to the replacement of, etc., the 1-bit digital signal of the left channel is input to the channel determination circuit 61R and is input to the channel determination circuit 61L. Is a right-channel 1-bit digital signal, that is, a 1-bit digital signal in which the shape of the quantization noise distribution is changed by the additional information signal generation circuit 10 is input.

In this state, the channel discrimination circuit 61R
Indicates that there is no difference above a certain level between the shape of the quantization noise distribution of the received 1-bit digital signal and the shape stored in advance, so that the 1-bit digital signal is the left channel, that is, the wrong channel. It is determined that
Similarly, the channel discriminating circuit 61L recognizes a difference of a certain amount or more between the shape of the quantization noise distribution of the received 1-bit digital signal and the shape stored in advance,
The 1-bit digital signal is the right channel, that is,
Determined to be the wrong channel. Therefore, the both channel discrimination circuits 61R and 61L instruct the channel switching circuit 53 to switch the left and right channels.

As a result, the right channel demodulation circuit 51R
The acoustic signal demodulated by the LPF 52R is output to the left channel amplifier 6L and is sonicated by the speaker 7L. Similarly, the left channel demodulation circuit 51L / LPF5
The acoustic signal demodulated by the 2L is transmitted to the speaker 7R via the right channel amplifier 6R and is acoustically converted. As a result, even if a 1-bit digital signal of an unusual channel is input to the demodulator 5 due to the switching of the transmission paths 4R and 4L, the demodulator 5 will not The channel of the 1-bit digital signal can be discriminated from the shape of the quantization noise distribution, and the acoustic signal of the correct channel can be output to the amplifiers 6R and 6L of each channel.

Here, the channel information is added to the 1-bit digital signal as the shape of the quantization noise distribution. Further, no matter which channel information is added, the quantization noise distribution of the 1-bit digital signal is selected so as to secure the dynamic range of the acoustic signal which is the main signal. Therefore, the demodulation circuit 51 / LPF 52 of each channel is
A bit digital signal can be demodulated into an acoustic signal.

As described above, in the acoustic signal transmission apparatus 1 according to the present embodiment, the unavoidable quantization noise distribution is used as the sub-signal in the analog / digital conversion, so that, for example, a simple low-pass filter or the like can be used. The main signal and the sub signal can be transmitted at the same time without impeding the advantage of the 1-bit digital encoding method that the circuit can demodulate the main signal.

Further, by selecting the loop gain so that the quantization noise distribution does not affect the main signal, even if it is reproduced by the demodulator which does not support the sub signal, the reproduction is performed without the influence of the sub signal. It can be carried out. Further, the main signal and the sub-signal, in the time direction and the frequency direction,
Since they are present at the same time, they are difficult to separate, and it is also difficult for a third party to remove, modify and tamper with side signals.

The channel discriminating circuits 61R and 61L according to the present embodiment store the shape of the quantization noise distribution when the sub-signal is not added as a criterion, but the present invention is not limited to this. . For example, it may be the shape of the quantization noise distribution of the 1-bit digital signal of the right channel to which the sub signal is added, or may be an intermediate value of the shape of the quantization noise distribution of the 1-bit digital signal of the left and right channels. . The same effect as in the present embodiment can be obtained as long as the shape of the received 1-bit digital signal makes it possible to determine which channel the 1-bit digital signal is from.

Further, the channel discrimination circuits 61R and 61R
L is, for example, using Fourier transform or the like to extract each frequency component of the quantization noise distribution, and frequency-determines the shape of the received quantization noise distribution of the 1-bit digital signal and the stored shape of the quantization noise distribution. Comparison is made for each component, but the present invention is not limited to this. For example, a predetermined calculation such as a weighted average may be performed on each frequency component of the quantization noise distribution of the 1-bit digital signal, and the calculation results may be compared. In this case, it suffices to store the result of the same calculation performed on the shape of the quantization noise that serves as the comparison reference. Further, instead of extracting frequency components by calculation such as Fourier transform, comparison may be performed using a filter or the like. In any case, if the channel information can be extracted by comparing the shape of the quantization noise distribution in the received 1-bit digital signal with the stored shape, various comparison methods can be applied.

Further, when it is difficult to compare the shapes of the quantization noise distributions of the 1-bit digital signal due to the level fluctuations of the main signal, for example, the quantization noise distributions are kept during the period when the level of the main signal is below a predetermined level. It is possible to improve the accuracy of extracting the channel information by detecting the period in which the shapes are easily compared and by demodulating device 5 comparing the shapes of the quantization noise distributions during the period.

Second Embodiment The channel discriminating circuit 61 according to the first embodiment stores the shape of the quantization noise distribution serving as a comparison reference in advance, and compares it with the shape to receive 1 The channel information added to the bit digital signal is determined.

On the other hand, in this embodiment, a case will be described in which the channel information is determined by comparing the quantization noise distributions in the received 1-bit digital signals of a plurality of channels. As shown in FIG. 6, the acoustic signal transmission device 1a according to the present embodiment is the first except that a channel discriminating circuit 61a is provided instead of the channel discriminating circuits 61R and 61L shown in FIG. It is similar to the embodiment. Therefore, members having the same functions as those in the first embodiment are designated by the same reference numerals and the description thereof will be omitted.

The channel discrimination circuit 61a according to the present embodiment.
Are connected to the outputs of both demodulation circuits 51R and 51L, and can control the channel switching circuit 53 based on the comparison result of both outputs. Specifically, the channel discrimination circuit 6
1a compares both outputs, and determines from the left channel or the right channel from the difference in the quantization noise distribution of both outputs. If the result of determination is that the left and right channels have been switched, the channel switching circuit 53 is instructed to switch and output the left and right channels.

As a result, the demodulation device 5 according to the present embodiment.
The sub-signal a can be extracted based on the shape of the quantization noise distribution of the 1-bit digital signal. As a result, the acoustic signal transmission device 1a can simultaneously transmit the main signal and the sub signal, as in the acoustic signal transmission device 1 according to the first embodiment.

Further, the channel discriminating circuit 61a according to the present embodiment extracts the sub-signal by comparing the shapes of the quantization noise distributions among the 1-bit digital signals of a plurality of channels. Therefore, unlike each channel discrimination circuit 61 according to the first embodiment, the sub signal can be extracted without storing the shape serving as the comparison reference. As a result, the storage area can be reduced and the circuit configuration can be simplified as compared with the channel discrimination circuit 61.

The modulation apparatus 3 according to the present embodiment is based on the channel information, and in only one channel,
Although the shape of the quantization noise distribution of the 1-bit digital signal is changed, the present invention is not limited to this, and the shape of the quantization noise distribution of both channels may be changed. In this case, by changing the quantization noise distribution shape of one channel so that the difference from the other shape becomes large, the quantization noise distribution shape of one channel becomes larger than that when the quantization noise distribution shape of one channel is fixed. The difference in shape can be expanded. In any case,
On the demodulator 5a side, if channel information can be extracted by comparing the quantization noise distribution shapes of both channels, substantially the same effect as this embodiment can be obtained.

[Third Embodiment] In the first and second embodiments described above, the channel discrimination circuits 61R, 61L, and 61a are the demodulation circuits 51R and 51L.
Since the shape of the quantization noise distribution is compared over the entire frequency band output by, there is a possibility that the comparison procedure will be complicated, and a high-speed channel discrimination circuit is required for real-time comparison. May be.

On the other hand, in the acoustic signal transmission apparatus according to this embodiment, a case will be described in which each channel discrimination circuit compares quantization noise distributions in a predetermined limited section. Note that the configuration can be applied to both the first and second embodiments, but in the following, with reference to FIG. 7, the case of application to the second embodiment will be described as an example.

That is, in the acoustic signal transmission device 1b according to the present embodiment, the channel discriminating circuit 61b provided in place of the channel discriminating circuit 61a and each demodulating circuit 51R.5.
A band pass filter (hereinafter,
62R and 62L are abbreviated as BPF, respectively. The center frequency of the BPF62R / 62L is
For example, the quantization noise distribution shape shown in FIG. 3 is set to a frequency that makes it easy to distinguish the shape of the quantization noise distribution indicating each channel information. Further, the channel discrimination circuit 61b compares the outputs of the BPFs 62R and 62L and extracts the channel information.

In this configuration, the band to be compared by the channel discriminating circuit 61b has a predetermined BPF 62R.
It is limited to the pass band of 62L. Therefore, the comparison procedure can be simplified as compared with the case where comparison is performed over the entire frequency band output by the demodulation circuits 51R and 51L. For example, if the band to be compared is sufficiently narrow, the channel information can be extracted only by comparing the levels of the quantization noise distribution in the band in the left and right channels.

When applied to the first embodiment, each channel discriminating circuit 61R / 61L has a corresponding BPF6.
Channel information is extracted by comparing the output of 2R / 62L with a predetermined value stored in advance.

In the above description, the BPF62R / 6
A predetermined range is extracted from the quantization noise distribution by 2L, but the present invention is not limited to this, and is used when comparing the quantization noise distributions, for example, by limiting the frequency band to be calculated in Fourier transform. The frequency band to be used may be limited.

As described above, the acoustic signal transmitters 1 • 1a • 1b according to the first to third embodiments have a modulation step of delta-sigma modulating a main signal to a 1-bit digital signal, and a transmission step. 1 via the channel or recording medium
The step of transmitting a bit digital signal and the above-mentioned 1 transmitted
A device for transmitting a main signal and a sub-signal by using a signal transmission method via a 1-bit digital signal having a demodulation step of demodulating a bit digital signal, and further based on the sub-signal when modulating. In addition, the shape of the quantization noise distribution of the main signal is changed, and at the time of the demodulation, the sub-signal is extracted based on the transmitted shape of the quantization noise distribution of the 1-bit digital signal. .

In the above configuration, the sub-signal is added as the shape of the quantization noise distribution of the 1-bit digital signal.
Therefore, the main signal can be demodulated as in the case where the sub signal is not added. As a result, both the main signal and the sub signal can be transmitted as a 1-channel 1-bit digital signal without disturbing the characteristic of delta-sigma modulation that demodulation is easy.

In addition, since the sub-signal is the quantization noise itself of the 1-bit digital signal, the main signal and the sub-signal exist in the same region in terms of time and frequency in the 1-bit digital signal. To do. Therefore, it is possible to prevent a third party from removing, changing, and falsifying the sub-signal.

In the first to third embodiments, the shape of the quantization noise distribution of the 1-bit digital signal is changed by changing the loop gain of the partial negative feedback loop in the delta sigma modulation circuit 31. However, it is not limited to this. For example, each integrator m
The shape of the quantization noise distribution may be changed by changing the multiplier coefficients of the differential amplifiers a1 to a7 (see FIG. 1) provided in 1 to m7. In any case, if the shape of the quantization noise distribution of the 1-bit digital signal can be selected based on the sub-signal, substantially the same effect as this embodiment can be obtained.

However, in the delta-sigma modulation circuit having the partial negative feedback loop, the frequency at which the dip is formed (zero point frequency) is changed by the loop gain of the partial negative feedback loop. As described above, the quantization noise level of the 1-bit digital signal sharply decreases in the vicinity of the zero point frequency. Therefore, when the loop gain of the partial negative feedback loop is changed to change the zero point frequency, the loop gain is changed. The shape of the quantization noise distribution changes significantly compared to the case where it is not changed.

Therefore, as shown in the first to third embodiments, each of the loop gain changing circuits 11a to 11c of the delta-sigma modulation circuit 31 responds to the instruction of the additional information signal generating circuit 10 in the partial negative feedback loop. By changing the loop gain of, the modulator 3 can generate a 1-bit digital signal in which the shape of the quantization noise distribution can be easily discriminated on the demodulation side.

In the configuration shown in FIG. 4, in order to simplify the circuit, the gain of each loop gain changing circuit 11 is changed by changing only the resistance value of the input resistance portion ri. Although explained, it is not limited to this.
If the configuration is such that the gain of each loop gain changing circuit 11 can be changed according to the instruction of the additional information signal generating circuit 10,
The same effect as this embodiment can be obtained. Furthermore, for example,
As shown in FIG. 1, the loop gain of the partial negative feedback loop may be changed by changing the multiplier coefficients of the differential amplifiers a1 to a7 provided in the integrators m1 to m7 of the respective orders. As long as the loop gain of the partial negative feedback loop can be changed according to the instruction from the additional information signal generation circuit 10, the same effect as this embodiment can be obtained.

In the first to third embodiments, the case where the sub signal added to the main signal is channel information has been described, but the present invention is not limited to this. For example,
Presence or absence of pre-emphasis, flag for copyright protection,
Alternatively, it may be a mastering code or the like. Since this information has a small amount of information, it can be added as a sub-signal without any trouble even if the shape of the quantization noise distribution cannot be set too much in order to sufficiently secure the dynamic range of the main signal. . In addition, since these pieces of information are closely related to the main signal, they are greatly damaged when they are removed, changed, or tampered with by a third party. However, when the 1-bit digital signal transmission method according to each of the above-described embodiments is used, a third party can remove the sub-signal,
Since it is difficult to change and tamper, this information can be surely protected and it is extremely effective. In any case, the same effect can be obtained as long as it is a sub-signal that is closely related to the main signal and has a small amount of information.

In the first to third embodiments, the case where the 1-bit digital signals of the left and right channels are transmitted has been described as an example, but the present invention is not limited to this. For example, it can be applied when transmitting two or more multi-channel 1-bit digital signals. Furthermore, the first
In the case where the sub-signal discriminating means (channel discriminating circuit) stores the shape of the quantization noise distribution to be compared as in the third embodiment and the third embodiment, a single-channel 1-bit digital signal is transmitted. It can also be applied in cases. In the case of a single channel, since the channel information has no meaning, other information is transmitted as a sub signal.

In addition, in the first to third embodiments, the case where the channel information indicates the right channel, that is, the case where the sub-signal is binary has been described as an example, but the present invention is not limited to this. Not a thing. If the quantization noise distributions of different shapes are associated with the values of the sub-signals, the sub-signals having a plurality of values can be added. For example, when each of the loop gain changing circuits 11a to 11c shown in FIG. 1 can select two loop gains, a combination of the loop gains can select 2 ³ quantization noise distribution shapes. It is only necessary to select a quantization noise distribution shape that can secure the dynamic range of the main signal and that can be distinguished from each other on the demodulation side.

On the other hand, the demodulators (5, 5b) shown in the first and third embodiments are the above-mentioned 1-bit digital signals in the demodulator circuit for demodulating the 1-bit digital signal generated by delta-sigma modulating the main signal. The shape of the quantization noise distribution of the signal is determined according to the sub-signal, and the shape of the quantization noise distribution stored in advance is compared with the shape of the quantization noise distribution of the received 1-bit digital signal. In addition, a sub signal discriminating means (channel discriminating circuits 61R, 61L, 61b) for discriminating the added sub signal is provided.

In the above structure, when the demodulation circuit receives the 1-bit digital signal, the sub-signal discriminating means determines the shape of the quantization noise distribution of the 1-bit digital signal and the shape of the previously stored quantization noise distribution. The comparison is performed, and the sub-signal is extracted based on the difference between the two. This allows the demodulation circuit to correctly extract the sub-signal from the received 1-bit digital signal. Furthermore, by storing the quantization noise distribution shape serving as a comparison reference in a storage medium such as an IC (Integrated Circuit) that cannot be read by a third party, the side signal can be removed, changed, and changed by the third party. Tampering can be securely prevented.

Also, the demodulation device 5a according to the second embodiment.
Is a demodulation circuit for respectively demodulating multi-channel 1-bit digital signals generated by delta-sigma-modulating multi-channel main signals.
The shape of the quantization noise distribution of the 1-bit digital signal of at least one specific channel is determined according to the sub-signal, and the shape of the quantization noise distribution of the received 1-bit digital signal of the specific channel and other Comparing the shape of the quantization noise distribution of the 1-bit digital signal of the channel,
It is characterized in that a sub signal discriminating means (channel discriminating circuit 61a) for discriminating the added sub signal is provided.

In the above structure, the demodulation circuit is, for example,
Receives 1-bit digital signals of multiple channels, such as left and right channels. Here, if the shape of the quantization noise distribution is changed in the 1-bit digital signal of the specific channel, the shape of the quantization noise distribution is different from the shape of the quantization noise distribution of other channels. The sub-signal discriminating means is a 1-bit digital signal of a specific channel and a 1-bit digital signal of another channel.
The shape of the quantization noise distribution is compared with that of the bit digital signal, and the side signal is extracted based on the difference.

Further, in this configuration, it is not necessary to previously store the shape of the quantization noise distribution as shown in the channel discriminating circuits 61R and 61L according to the first embodiment, so that the demodulating circuit can be simplified in configuration. . Accordingly, it is possible to provide a demodulation circuit that can correctly extract the sub-signal from the received 1-bit digital signal with a relatively simple configuration.

Also, the demodulator 5b according to the third embodiment.
In the sub-signal discrimination means (channel discrimination circuit 61
b) is characterized in that the sub-signal is discriminated based on the shape in the predetermined specific frequency band in the quantization noise distribution of the received 1-bit digital signal. Note that the method of limiting to a specific frequency band may be, for example, extracting only a component of a specific frequency band from a received 1-bit digital signal by using a bandpass filter or the like, or calculating a Fourier transform or the like. You may extract only the component of a specific frequency band using.

In this configuration, since the frequency band to be compared is limited, the comparison becomes easier compared to the case where the shapes of the quantization noise distributions are compared over all the frequency bands. In particular, it is possible to reduce the area required for storage, as compared with the case where the shape of the quantization noise serving as the comparison reference is stored in advance like the channel discrimination circuits 61R and 61L according to the first embodiment.

When the sub-signal is multi-valued, if the frequency band to be compared is limited as shown in the third embodiment, it may be difficult to distinguish the shape of each quantization noise distribution. In this case, for example, by providing a plurality of BPFs and setting a plurality of frequency bands to be extracted, a multi-valued sub-signal can be extracted without any trouble.

In the first to third embodiments, the case where the 1-bit digital signal is transmitted via the transmission lines 4R and 4L has been described as an example, but the present invention is not limited to this. The present invention can also be applied, for example, when the modulator 3 records a 1-bit digital signal on a recording medium and the demodulator 5 demodulates the 1-bit digital signal from the recording medium. The modulator 3 has 1 based on the main signal and the sub signal.
As long as the demodulator 5 acquires the main signal and the sub signal from the 1-bit digital signal while generating the bit digital signal, the same effects as those of the above-described embodiments can be obtained.

[0113]

According to the first aspect of the present invention, there is provided a method of transmitting a signal via a 1-bit digital signal by performing delta sigma modulation.
The modulation step of generating the bit digital signal includes a step of changing the shape of the quantization noise distribution of the main signal based on the side signal, and the demodulation step includes the transmitted quantization noise of the 1-bit digital signal. This is a configuration including a step of extracting a sub-signal based on the shape of the distribution.

In the above configuration, the sub-signal is added as the shape of the quantization noise distribution of the 1-bit digital signal.
Therefore, there is an effect that both the main signal and the sub signal can be transmitted without impairing the characteristics of the 1-bit digital encoding system.

In addition, since the sub signal is the quantization noise itself of the 1-bit digital signal, the main signal and the sub signal exist in the same region both in terms of time and frequency. Therefore, it is possible to prevent the removal, alteration and tampering of the sub signal by a third party.

In the delta-sigma modulation circuit according to the invention of claim 2, the main signal is delta-sigma modulated as described above,
In a delta-sigma modulation circuit that generates a 1-bit digital signal, a sub-signal adding unit that changes the shape of the quantization noise distribution of the 1-bit digital signal according to the sub-signal ,
The input signals, which are signals, are input to the first stage and are connected in cascade.
Add the output of each of the above integrators with the continuous integrators
Adder and the output of the adder are quantized to 1 bit
Quantizer that outputs a digital signal and the output of the integrator
Is negatively fed back to the input side of the preceding integrator from the integrator.
A partial negative feedback loop, and the auxiliary signal adding means described above.
Is the loop gain of the above partial negative feedback loop, depending on the side signal.
This is a configuration including loop gain changing means for changing the IN .

As described above, the delta-sigma modulation circuit according to the invention of claim 3 is the partial negative feedback circuit which constitutes the partial negative feedback loop in the configuration of the invention of claim 2.
Includes a differential amplifier provided between the integrators,
The sub-signal adding means is responsive to the sub-signal for the differential signal.
The above loop by changing the multiplier coefficient of the amplifier
This is a configuration for changing the gain .

In the configurations according to the inventions of claims 2 and 3, since the loop gain is changed according to the sub-signal, 1
The quantization noise level of the bit digital signal sharply decreases around a zero frequency different from that before the change of the sub-signal, and the sub-signal adding means can largely change the shape of the quantization noise distribution. As a result, the delta-sigma modulation circuit has an effect that it can generate a 1-bit digital signal in which the sub-signal can be more easily discriminated on the demodulation side.

Therefore, the delta-sigma modulation circuit is
Signal demodulation is easy, and the main and side signals are on a single channel
Can be transmitted by a third party, and a third party can remove, change, or
Can generate a 1-bit digital signal that is difficult to tamper with
Produces the effect.

As described above, in the delta-sigma modulation circuit according to the invention of claim 4, in the configuration of the invention of claim 2 or 3, the main signal is an acoustic signal and the sub-signal is channel information. , A signal indicating at least one of the presence / absence of pre-emphasis, a flag for copyright protection, and a mastering code.

Therefore, there is an effect that each of the above-mentioned information, which is closely related to the main signal and which is greatly damaged when removed, can be transmitted while being surely protected.

As described above, the demodulation circuit according to the invention of claim 5 compares the shape of the quantization noise distribution stored in advance with the shape of the quantization noise distribution of the received 1-bit digital signal, This is a configuration including a sub-signal discriminating means for discriminating the added sub-signal.

In the above configuration, the shape of the quantization noise distribution stored in advance is compared with the shape of the quantization noise distribution of the received 1-bit digital signal to extract the sub-signal. As a result, the demodulation circuit has an effect of correctly extracting the sub-signal from the received 1-bit digital signal.

In addition, by storing the quantization noise distribution shape serving as a comparison reference in a storage medium that cannot be read by a third party, it is possible to reliably prevent removal, alteration and tampering of the side signal by the third party. This effect is also played.

As described above, in the demodulation circuit according to the invention of claim 6, the shape of the quantization noise distribution of the received 1-bit digital signal of the specific channel and the quantization noise distribution of the 1-bit digital signal of another channel are received. The sub-signal discriminating means for discriminating the added sub-signal by comparing with the shape of

In the above configuration, it is not necessary to store the shape of the quantization noise distribution in advance, unlike the configuration of the invention described in claim 5. Therefore, it is possible to provide a demodulation circuit that can correctly extract the sub-signal from the received 1-bit digital signal with a relatively simple configuration.

As described above, the demodulation circuit according to the invention of claim 7 has the configuration of the invention of claim 5 or 6,
The sub-signal discriminating means is configured to discriminate the sub-signal based on the shape of the received quantization noise distribution of the 1-bit digital signal in a predetermined specific frequency band.

In this configuration, since the frequency band to be compared is limited, the effect of facilitating the comparison can be obtained as compared with the case of comparing the shapes of the quantization noise distributions over all the frequency bands. Play.

[Brief description of drawings]

FIG. 1 shows an embodiment of the present invention and is a block diagram showing a main configuration of a delta-sigma modulation circuit.

FIG. 2 is a block diagram showing a case of extracting a sub-signal by comparing it with predetermined quantization noise, and is a block diagram showing a main part configuration of an acoustic signal transmission apparatus including the delta-sigma modulation circuit.

FIG. 3 is a graph showing a distribution state of quantization noise in a 1-bit digital signal output from the delta sigma modulation circuit.

FIG. 4 is a block diagram showing a configuration example of a loop gain changing circuit in the delta sigma modulation circuit.

FIG. 5 is a block diagram showing a main part configuration of a delta-sigma modulation circuit for a channel whose quantization noise distribution shape is not changed according to channel information.

FIG. 6 shows another embodiment of the present invention, and is a block diagram showing a main configuration of an acoustic signal transmission device in the case of comparing sub-signals by comparing quantization noise of both channels.

FIG. 7 shows still another embodiment of the present invention, which is a main part of an acoustic signal transmission device in the case of extracting a side signal by comparing a specific part of the characteristics of the quantization noise of both channels. It is a block diagram which shows a structure.

FIG. 8 is a block diagram showing a conventional example and showing a main part configuration of a delta-sigma modulation circuit.

[Explanation of symbols]

3 modulator (delta sigma modulator) 5.5a 5b demodulator (demodulator) 10 additional information signal generator (sub signal adder) 11a loop gain change circuit (partial negative feedback circuit; loop gain changer) 11b loop Gain changing circuit (partial negative feedback circuit; loop gain changing means) 11c Loop gain changing circuit (partial negative feedback circuit; loop gain changing means) 23 Adder 24 Quantizer 61R / 61L / 61a / 61b Channel discrimination circuit (sub signal) Discriminating means) m1 to m7 integrator

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H03M 3/02 H03M 1/08

Claims (7)

(57) [Claims] 1. A step of modulating a main signal by delta sigma modulation to a 1-bit digital signal, a step of transmitting the 1-bit digital signal via a transmission line or a recording medium, and the transmitted 1-bit digital signal. A method of transmitting a signal via a 1-bit digital signal, which comprises a demodulation step of demodulating the signal, wherein the modulating step changes the shape of the quantization noise distribution of the generated 1-bit digital signal according to the sub-signal. And the step of demodulating includes a step of extracting a side signal based on the shape of the quantization noise distribution of the transmitted 1-bit digital signal. Signal transmission method. 2. A delta-sigma modulation circuit for delta-sigma-modulating a main signal to generate a 1-bit digital signal, wherein a sub-signal is added to change the shape of the quantization noise distribution of the 1-bit digital signal according to the sub-signal. Means and the input signal that is the main signal are input to the first stage and are cascaded with each other.
A plurality of connected integrators, an adder for adding the outputs of the integrators, and a 1-bit digital signal by quantizing the output of the adder.
And the output of the above integrator to the input of the integrator in the previous stage.
And a partial negative feedback loop for negatively feeding back the force side, the sub-signal addition means, in response to the sub signal, said portion negative retrace
Loop gain change hand to change the loop gain of the return loop
A delta-sigma modulation circuit characterized by having stages . 3. A partial negative feedback constituting the partial negative feedback loop.
The return circuit includes a differential amplifier provided between both integrators.
Wherein, the sub-signal addition means, in response to the sub signal, the differential
The above loop by changing the multiplier coefficient of the amplifier
The delta-sigma modulation circuit according to claim 2, wherein the gain is changed . 4. The main signal is an acoustic signal, and the sub-signal is a signal indicating at least one of channel information, presence / absence of pre-emphasis, a flag for copyright protection, and a mastering code. The delta-sigma modulation circuit according to claim 2 or 3. 5. A demodulation circuit for demodulating a 1-bit digital signal generated by subjecting a main signal to delta-sigma modulation, wherein the quantization noise distribution of the 1-bit digital signal has a shape determined according to a sub-signal. And the shape of the quantization noise distribution stored in advance and the received 1
A demodulation circuit characterized by comprising sub-signal discriminating means for discriminating an added sub-signal by comparing with a shape of a quantization noise distribution of a bit digital signal. 6. A demodulation circuit for respectively demodulating multi-channel 1-bit digital signals generated by delta-sigma modulating multi-channel main signals, wherein at least one specific channel 1-bit digital signal of each of the channels is demodulated. The shape of the quantization noise distribution of the signal is determined according to the sub-signal, and the shape of the received quantization noise distribution of the 1-bit digital signal of the specific channel and the quantization of the 1-bit digital signal of the other channel A demodulation circuit comprising: a sub-signal discriminating means for discriminating an added sub-signal by comparing the shape of a noise distribution. 7. The sub signal discriminating means receives the 1
7. The demodulation circuit according to claim 5, wherein the sub-signal is discriminated based on the shape in a predetermined specific frequency band of the quantization noise distribution of the bit digital signal.