close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JP2011211725

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JP2011211725
An object of the present invention is to generate an analog signal of high quality even when there
is a variation in elements constituting a digital-to-analog converter for converting a digital signal
into an analog signal, and has a high resolution and a circuit scale To realize a small digital-toanalog converter. A first data converter for reducing the number of bits of an input signal, a
second data converter for converting the format of the first output signal, and a history of the
output of the second data converter A data converter is provided, having a third data converter
for converting into a corresponding code. [Selected figure] Figure 9
Digital to analog converter
[0001]
The present invention relates to a digital-to-analog converter for converting a digital signal into
an analog signal and its application.
[0002]
US Pat. No. 5,862,237 as a prior art digital-to-analog converter for converting digital signals to
analog signals, and a digital-to-analog converter for converting voice signals to multiple digital
signals and reproducing multiple voice signals using multiple speaker drivers. And US Pat. No.
5,909,496 have been proposed.
[0003]
11-04-2019
1
In FIG. 1 of US Pat. No. 5,862, 237, a digital serial audio signal is converted into a plurality of
digital signals by a serial-to-parallel converter and a decoder circuit.
Here, it is a feature of the conventional example to convert a plurality of digital signals so as to
be weighted by the amplitude of the audio signal.
Thereby, when driving a plurality of speakers, the amount of current of the current sources of
the plurality of driving devices is controlled according to the weighting to drive the plurality of
speaker devices, thereby making the sound according to the amplitude of the audio signal We
propose a system to play.
[0004]
In FIG. 4 of US Pat. No. 5,909,496, the digital serial audio signal is once converted into a plurality
of digital signals by a serial-to-parallel converter and a decoder circuit as in US Pat. No.
5,862,237. Here, the plurality of digital signals are converted to be weighted by the amplitude of
the audio signal, and the direction of the current of the drive circuit for driving the plurality of
speakers is one specific bit of the plurality of digital signals (in the known example) Control using
MSB) is a feature of the conventional example. Thereby, when driving the plurality of speakers,
the amount of current of the current sources of the plurality of driving devices is controlled
according to the weighting to drive the plurality of speaker devices to reproduce the sound
according to the amplitude of the audio signal. At the same time, it is possible to configure the
drive circuit with a simpler circuit.
[0005]
In these conventional examples, since digital signals subjected to serial-to-parallel conversion are
used as they are as signals for driving a plurality of speakers, firstly, manufacturing variations
among current sources of weighted drive circuits are nonlinear noises. Second, there is a problem
that quantization noise generated when reproducing a digital signal is superimposed as a noise
component in an audio frequency band, etc., so that high-quality audio signals can be
reproduced. It has the drawback of being difficult.
[0006]
In order to avoid the first problem, it is necessary to have means for suppressing manufacturing
variations among the plurality of drive devices.
11-04-2019
2
[0007]
In FIG. 33 of US Pat. No. 5,872,532, a technique comprising a selection circuit and an integrator
for controlling the selection circuit is proposed as a means for suppressing variations between
current sources driving a plurality of speaker drive devices.
In this proposal, a plurality of loudspeakers are controlled by inputting a signal for driving a
plurality of loudspeakers to the selection device and controlling the circuit for integrating the
presence or absence of the plurality of loudspeaker drive circuits one or more times. The
frequency of use of each of the drive devices is integrated, and the selection circuit is controlled
so as to keep the integration result constant.
This makes it possible to reduce noise due to manufacturing variations among the drive devices.
A technique for suppressing variations among a plurality of drive devices is called a mismatch
shaping method.
[0008]
A method is proposed in FIG. 1 of US Pat. No. 5,592,559, in which the input digital serial audio
signal is once subjected to digital modulation using a ?? modulator to drive a voice coil to
reproduce voice. This conventional example is a proposal for driving a speaker in the positive
and negative directions of two voice coils using a digitally modulated ternary signal, but it is
possible to drive two or more voice coils and to drive a plurality of driving devices. There is no
mention of technology to reduce the variation in
[0009]
USP 7,058,463 Fig. In No. 3, it has been proposed to discharge the input digital serial voice
signal to a frequency higher than the audio frequency by applying digital modulation using a
?? modulator and oversampling. A technique that spouts quantization noise out of the
frequency of interest in this way is called a noise shaping method. In this conventional example,
quantization noise generated when reproducing a digital signal is moved to a high frequency
11-04-2019
3
band outside the audible frequency using a noise shaping method. This avoids the problem of the
second problem of quantization noise being superimposed as a noise component in the audio
frequency band.
[0010]
Further, in the conventional example, in order to avoid the problem of noise caused by
manufacturing variations among a plurality of driving devices, which is the first problem, the
control is performed by the DEM (Dynamic Element Matching) method using a pseudo random
signal. It is proposed to introduce a mismatch shaping method using a selection circuit.
[0011]
However, since the speaker drive circuit is driven as it is without attenuating the quantization
noise emitted to a frequency higher than the audio frequency by applying the digital modulation
using the ? ? ? modulator and oversampling, the high frequency band There is a problem that
the quantization noise moved to is emitted from the speaker.
[0012]
Also, simply switching the selection circuit by the DEM method using a random signal has the
disadvantage that white noise caused by this random signal is superimposed on the reproduced
audio signal.
In order to avoid the problem of noise caused by manufacturing variations among a plurality of
drive devices, it is necessary to operate the switching operation of the selection circuit by the
DEM method at high speed as the number of speaker drive circuits increases.
The details of the operation of the DEM method are described in Section 8.3.3 of the reference
"Delta-Sigma Data Converters" IEEE Press 1997 ISBN 0-7803-1045-4 and Figure 8.5. In the
mismatch shaping method using the DEM method, the selection circuit needs to operate at high
speed is a serious drawback in the implementation of this conventional example. Incidentally, this
drawback is already pointed out as a problem in US Pat. No. 5,872,532 and is known.
[0013]
11-04-2019
4
As in the above conventional example, by using the noise shaping method by the digital
modulation using the ? ?? ? modulation circuit and the oversampling, the quantization noise
generated by reproducing the digital signal is emitted to the frequency band higher than the
audio frequency. Things are generally well known techniques. Reference "Over sampling DeltaSigma Data Converters" IEEE Press 1991 ISBN 0-87942-285-8 pp. Equation (22) of 7 shows the
relationship between the oversampling ratio and the strength of the noise shaped noise with
respect to the order of the modulator. Generally, by noise shaping, the effective intensity of
quantization noise decreases by 3 (2L + 1) dB every time the oversampling ratio is doubled,
where L is the order of the ?? modulator. Therefore, to reduce quantization noise, the
oversampling ratio must be increased or the order of the ?? modulator must be increased. On
the other hand, when the oversampling ratio is increased, it is necessary to operate the ??
modulator at high speed. Further, if the order of the ?? modulator is increased, the operation of
the ?? modulator becomes unstable.
[0014]
As described above, in the noise shaping method by digital modulation using the ?? modulation
circuit and oversampling, quantization noise generated by reproducing a digital signal is
discharged to a frequency band higher than an audio frequency. Therefore, it is necessary to
attenuate the noise-shaped unnecessary quantization noise generated in the ?? modulation
circuit and the component outside the audio frequency band with the continuous time LPF
(Continuous-Time Low Pass Filter).
[0015]
FIG. 1A shows an example of a general system using a ?? modulation circuit. The noise-shaped
unnecessary quantization noise and out-of-band components generated in the ?? modulator
(100) are attenuated by the continuous time LPF (101). Since oversampling is performed, the
LPF may be a low-order one, but when the passband is narrow, the time constant becomes large,
and the area occupied by the LPF can not be ignored when built in a semiconductor integrated
device.
[0016]
11-04-2019
5
As shown in FIG. 1 (b), there is a method of reducing the characteristic request of the LPF, which
is disposed downstream of the modulator, as a multi-bit ?? modulator (110) as shown in FIG. 1
(b). In this case, since the quantization noise can be reduced by 6 dB by increasing the number of
bits of the ?? modulator by 1 bit, the cutoff frequency characteristic of the LPF can be relaxed.
However, increasing the number of bits of the modulator increases the circuit size of the internal
modulator.
[0017]
As another method of relaxing the characteristic request of the LPF, a method of inserting the
switched capacitor filter (121) shown in FIG. 1C between the ?? modulator and the LPF is also
proposed. In this case, in addition to the need for an OP amplifier to realize a switched capacitor
filter, a large capacitor may be needed to lower the cutoff frequency, which increases chip area
and power consumption. There is a drawback.
[0018]
As another method of reducing the characteristic requirement of the LPF, a method of inserting
the analog FIR filter (131) shown in FIG. 1 (d) between the ?? modulator and the LPF has been
proposed. In this method, an analog FIR filter is configured by adding each tap of the FIR filter in
an analog manner to be an output. In this case, the amount of attenuation for out-of-band noise
can be increased by increasing the number of taps. The method using an analog FIR filter also
has the effect of reducing the degradation of the SNR due to clock jitter, and is an effective
method when using a clock signal with low accuracy or when using multiple clocks on the same
chip.
[0019]
However, when the ?? modulator has multiple bits, the delay elements constituting the analog
FIR filter are required for the number of cells of the segment type modulator constituting the bits
of the ?? modulator О the number of taps. There is a disadvantage that the circuit scale
increases rapidly.
[0020]
The operation will be described in more detail with respect to a method of postfixing an analog
FIR filter to a system using a general noise shaping method using a ?? modulation circuit,
11-04-2019
6
particularly when using a cascade ?? modulator.
[0021]
First, FIG. 2 shows a general configuration of the cascade ?? modulator (200).
The input digital signal (210) is quantized by the first stage ?? modulator (201), and the first
stage quantization noise (211) is further quantized by the second stage ?? modulator (202). Be
done.
The output Y2 of the second stage is converted by the digital signal processing block (220), and
then the output of the first stage is added with Y1 (230) and output.
[0022]
The output of the first stage Y1 and the output of the second stage Y2, the noise transfer
function of the first and second stages NTF1 (z), NTF2 (z), the quantization noise of the first and
second stages Assuming that the gain from the first stage to the second stage is A1 and H3 =
NTF1 (z) / A1, the total output Y is Y = Y1 + Y2H3 = Y1 + Y2NTF1 / A1 = X + NTF1Q1 + (?
A1Q1 + NTF2Q2) NTF1 / A1 = X + NTF1Q1-NTF1Q1 + NTF1NTF2Q2 / A1 = X + NTF1NTF2Q2 /
A1 (Equation 1) The first stage quantization noise can be canceled out.
[0023]
A general configuration (300) in which an analog FIR filter (301) is added to this cascaded delta
sigma modulator is shown in FIG.
[0024]
This configuration can also be converted to a configuration (400) in which an analog FIR filter is
placed after each stage of the cascade ?? modulator as shown in FIG.
The operation of the second stage in the configuration in which the analog FIR filter is disposed
at each stage of the cascade ?? modulator as shown in FIG. 4 will be described in detail below.
11-04-2019
7
[0025]
The signal from Y2 is multiplied by H3 (z) in the digital signal processing block (220) and then
multiplied by the transfer function HFIR (z) of the FIR filter (300).
[0026]
Now, consider the case where the first-stage ? ? ? modulator and the FIR filter are moving
average filters.
The transfer function of the FIR filter is defined as H3 (z) = NTF1 = (1-z <-1>) HFIR (z) = 1 + z <1> + z <-2>... + Z <-(n-1)> .. ... (Equation 2) H3HFIR = (1-z <-1>) (1 + z <-1> + z <-2>... + Z <-(n-1)>)
= 1-z <-n And can be configured by a 2-tap post filter regardless of the number of taps of the FIR
filter.
In other words, when an analog FIR filter is added to the cascade ?? modulator, the number of
taps of the second stage after put filter is always 2 taps by using the configuration shown in FIG.
Even if the number is increased, the number of taps of the post filter does not increase and it is
suitable for miniaturization.
[0027]
Similarly, consider a configuration in which the first stage is a second-order ?? modulator and
the FIR filter is a moving average filter. Since H3 = NTF1 = (1-z <-1>) <2>, H3HFIR = (1-z <-1>)
<2> (1 + z <-1> + z <-2>... + Z <- n-1)>) = 1?z <?1> ?z <?n> + z <? (n + 1)> (Equation 4), and
the number of taps in the second stage post-filter There are four taps regardless of the tap length
of the FIR filter.
[0028]
In other words, when an analog FIR filter is added to a cascade ?? modulator, the second stage
can be obtained even if the number of taps of the FIR filter is increased regardless of the order of
the ?? modulator by adopting the configuration of FIG. It is possible to suppress an increase in
the number of taps of the post filter, and it is understood that it is suitable for miniaturization.
11-04-2019
8
[0029]
Note that YFIR in the case where an analog FIR filter is added to a cascade ?? modulator is
YFIR = (1 + z <?1> + z <?2>. (Equation 5)
U.S. Pat. No. 5,862,237 U.S. Pat. No. 5,909,496 U.S. Pat. No. 5,872,532 U.S. Pat. No. 5,592,559
U.S. Pat. No. 7,058,463 U.S. Pat.
[0030]
As described above, FIG. 5 shows a general block diagram in the case where an analog FIR filter
is placed after the modulator of each stage of the cascaded ?? modulator. Here, for
convenience of explanation, the number of taps of the FIR filter is n, the noise transfer function
of the first stage of the cascade ?? modulator is NTF1 = (1-z <-1>) <2>, and the first stage is 1
bit The second stage consists of an n-bit internal modulator.
[0031]
Here, the digital input signal (510) is input to the first stage internal modulator (201) of the
cascade ?? modulator, and the second stage internal modulator (202) is the first stage internal
modulator 201) is cascade-connected, and an output signal (520) from the first internal
modulator (201) is input to the analog FIR filter (301). The output signal (530) from the second
stage internal modulator (202) is converted and output from the binary code to the thermometer
code by the formatter circuit (501). The signal (531) converted to the thermometer code is input
to the post filter circuit (502). An output signal (521) from the analog FIR filter (301) and an
output signal (532) from the post filter circuit (502) are added in an analog form by a summing
block (540) and output.
[0032]
In a cascade ?? modulator using an analog FIR filter, consider the influence when tap
coefficients forming the analog FIR filter have an error.
11-04-2019
9
[0033]
If the first stage internal modulator is configured with one bit, the mismatch causes a tap
coefficient error to affect the frequency characteristics of the analog FIR filter.
However, since the linearity from the digital input to the analog output is not affected, the
distortion characteristics and the SNR are not degraded.
[0034]
On the other hand, when the first stage internal modulator has three or more levels, the
mismatch of the analog FIR filter unit directly affects the output as in a general ?? modulator,
and the distortion and the SNR characteristic are degraded, so one stage In order to increase the
number of levels of the internal modulator in the eye, a separate mismatch shaper is required.
[0035]
Mismatches of the elements that make up the second stage post-filter also affect the output, but
the SNR is degraded because the second stage input signal is the first stage quantization noise,
but the signal component is included. If not, the distortion characteristics do not deteriorate.
[0036]
Here, let us calculate the influence of the tap coefficients of the analog FIR filter and the post
filter on the output YFIR.
[0037]
Here, when the first and second internal modulators both have two levels and NTF1 = NTF2 = (1z <-1>) <2>, the characteristic of the analog FIR filter is H1FTR, post-filter Let H 2 FTR be the
characteristic of the first stage and let b 0, b 1,..., B n be the first stage tap coefficients a 0, a 1,. z)
= H1FTRY1 + H2FTRY2 = H1FTR {X + NTF1Q1} + H2FTR {-Q1 + NTF2Q2 / A1} = (a0 + a1z <-1> +
a2z <-2>... + an1z <-(n-1)>) {X + (1-z <-1>) <2> Q1}-(b0-b1z <-1> -bn-1z <-n> + bnz <-(n + 1)>) {Q1
+ (1-z <-1>) <2> Q2 / A1} ... (Equation 6)
[0038]
11-04-2019
10
When the influence of the tap coefficient at direct current is determined, YFIR (z) | z = 1 = (a0 +
a1 + a2... + An1) X? (b0?b0?bn?1 + bn) Q1 (Equation 7)
It can be seen that the quantization noise of the first stage appears in proportion to the tap
coefficients of the elements constituting the second stage postfilter.
Assuming that the tap coefficients of the second stage post filter are b0 = 1 + ?b0, b1 = 1 +
?b1, bn?1 = 1 + ?bn?1, bn = 1 + ?bn for simplicity, YFIR (z) | z = 1 = ( a0 + a1 + a2 ... + an1)
X-(? b0-? b1-? bn-1 +? bn) Q1 (equation 8).
Therefore, at the output, the first stage quantization noise Q1 appears at the output in proportion
to the product sum of the errors ?bi of the taps.
[0039]
As described above, although it has become possible to reduce out-of-band noise by using a
cascaded ?? modulator using an analog FIR filter, noise due to a mismatch of elements
constituting the post filter is in-band. There was a problem of increasing the noise.
[0040]
The present invention comprises a first circuit receiving a first input signal, a second circuit
receiving a second input signal, a third circuit receiving an output signal from the second circuit,
and the third circuit. A fourth circuit receiving an output signal from the first circuit, and an
adder circuit combining the output signal of the first circuit and the output signal of the fourth
circuit and outputting the combined signal; Is formed by combining a digital analog conversion
circuit and an analog FIR filter, and the transfer coefficient of one of the second circuit and the
third circuit is (1-z <-1>), and the second The other transfer coefficient of the circuit and the third
circuit is (1-z <-n>), and the transfer coefficient of the fourth circuit is HFIR (z) = 1 + z <-1> + z <2>. In the case of <-(n-1)>, one of the second circuit and the third circuit having the transfer
coefficient (1-z <-1>) is an analog It consists of the road, and the other one of the second circuit
and the third circuit having a transfer coefficient (1-z <-n>) is composed of a digital circuit.
[0041]
According to the present invention, even when there is a variation in elements constituting a
digital-to-analog converter that converts digital signals into analog signals, high-quality analog
11-04-2019
11
signals can be generated, and they have high resolution. A small digital-to-analog conversion
device can be realized.
[0042]
In the cascaded ?? modulator according to the present invention, the postfilter placed after the
second-stage modulator in the case where the analog FIR filter is placed is characterized as
follows.
[0043]
Assuming that the order of the internal modulator of the cascade ?? modulator is 1 and H3 =
NTF1 = (1-z <-1>), then H3HFIR = (1-z <-1>) (1 + z <-1> + z <?2>... + Z <? (n?1)>) = (1?z
<?n>) ииииииииии (Equation 9) The order of the internal modulator is equated to H3 = NTF1 = If (1-z <1>) <2> = (1-z <-1>) <> (1-z <-1>), then H3HFIR = (1-z <-1>) (1 + z) <?1> + z <?2>... + Z <?
(n?1)>) (1-z <?1>) = (1-z <?n>) и (1?z <?1>) ... (Equation 10)
[0044]
In both cases of Eq. 9 and Eq. 10, since H3HFIR includes (1-z <-n>), this (1-z <-n>) term is
separated from the post-filter and pre-digitalized. The first feature is to do.
[0045]
On the other hand, it is a second feature that terms other than (1-z <-n>) are once converted into
a thermometer code by a formatter and then subjected to calculation processing by a post filter.
[0046]
Example 1 FIG. 6 shows a first example in which an analog FIR filter and a post filter are
provided behind the cascaded ?? modulator of the digital-to-analog converter according to the
present invention.
In this embodiment, it is assumed that the first stage of the cascaded ?? modulator is a 1-bit
internal modulator and the second stage is an n-bit internal modulator.
[0047]
11-04-2019
12
Here, the digital input signal (510) is input to the first stage internal modulator (201) of the
cascade ?? modulator, and the second stage internal modulator (202) is the first stage internal
modulator 201) is cascade-connected, and an output signal (520) from the first internal
modulator (201) is input to the analog FIR filter (301).
The output signal (530) from the second stage internal modulator (202) performs calculation of
(1-z <-n>) in the digital signal processing block (601).
The output (631) from the digital calculation block (601) is converted and output from the
binary code to the thermometer code by the formatter circuit (602).
The signal (632) converted to the thermometer code is input to the post filter circuit (603).
An output signal (521) from the analog FIR filter (301) and an output signal (633) from the post
filter circuit (603) are added in an analog form by a summing block (540) and output.
[0048]
FIG. 7a shows a first embodiment of the digital-to-analog converter of the present invention.
In the analog FIR filter of this embodiment, a delay element (701) composed of DFF performing
one clock delay, a drive buffer (702) connected to the output, and one end are connected to the
drive buffer, and one end is an analog The resistance element (703) connected to the output
terminal is used as a single-stage unit so that the voltages are weighted and added in an accurate
manner, and the units are connected in a plurality of stages.
[0049]
As shown in Equation 10, when the order of the internal modulator is quadratic, the transfer
function of the second stage is (1?z <?n>) и (1?z <?1>).
11-04-2019
13
(1Since -z <-n>) is processed digitally, the post filter needs to calculate (1-z <-1>) in an analog
manner.
FIG. 7b shows an embodiment of the 1-bit unit of the post filter, which is placed after the cascade
? ? ? ? ? ? modulator of the present invention.
[0050]
Here, the input signal Y2-m (632) indicates a signal of one bit of the digital signal converted to
the thermometer code by the formatter. The post filter unit to which the input signal Y2-m (632)
is input is a switch controlled by a delay element (711) configured by DFF that executes one
clock delay and a signal 00 obtained by dividing one clock Resistor element (713) connected to
the input via (715a) and one end connected to the drive buffer, and one end connected to the
output terminal such that the voltage is weighted and added in an analog manner. And a drive
inverter (714) connected to the output via a switch (715a) controlled by a signal 00 similarly
dividing one clock, and one end is connected to the drive inverter and one end is an analog
voltage A resistive element (715) connected to the output terminal so as to perform weighted
addition, an inverter (714) connected to the input and output of a delay element (711) composed
of DFF that performs one clock delay, and the inverter Signal divided by 1 clock ? And a switch
(715b) that is controlled by.
[0051]
Here, the connection between the input / output of the delay element (711), the drive buffer
(712) and the drive inverter (714) is connected by switching between the input / output with the
switches (715a) and (715b). The switches (715a) and (715b) are controlled by the signals .PHI.0
and .PHI.1 obtained by dividing one clock, so that they constitute a swapping circuit in which the
connection relationship of the resistance elements is swapped for each clock. By this swapping
circuit, the transfer function Y2 (z) with respect to the input digital signal sequence y21 (n), y21
(n + 1), y21 (n + 2),..., Y21 (n + k) (k: integer) is Y2, o (z) = b0 (1-Z <-1>), Y2, e (z) = b1 (1-Z <-1>)
However, Y2, o (z) represents k = odd, Y2 and e (z) indicate k = even. Further, it is assumed that a
variation error of b0 and b1 resistance elements (713) and (715). ????
11-04-2019
14
[0052]
Therefore, 1-Z <-1> is multiplied to variation errors b0 and b1. Therefore, if calculated as z = 1,
then Y2 (z) | z = 1 = 0. It shows that the influence of the mismatch disappears, and the first-order
mismatch shaping is applied to the variation.
[0053]
The simulation result of the output spectrum of the entire ?? modulator when the elements
constituting the post filter have a 1% mismatch is shown in FIG.
In the conventional method (conventional) which does not use a changeover switch, it can be
confirmed that the noise in the low band is significantly increased. On the other hand, when the
proposed method is used (Proposed), it can be seen that the low band noise is reduced by 6 dB
when the frequency is 1?2.
[0054]
As described above, it is understood that, when the present method is used, high SNR can be
realized even when element values such as resistors constituting the digital-to-analog conversion
device have variations, and a high-resolution digital-to-analog conversion device can be
configured. In LSI, element value variation is generally about 0.1%. Even in such a case, it is
possible to configure a high accuracy and high resolution digital-to-analog converter by using
this method.
[0055]
Second Embodiment FIG. 9 shows a second embodiment of the digital-to-analog converter
according to the present invention. In this embodiment, it is assumed that the first stage of the
cascade ?? modulator is a 1-bit internal modulator, and the second stage is an n-bit internal
modulator. As shown in Formula 9, when the order of the internal modulator is one, the transfer
function of the second stage is (1?z <?n>) и 1. (1Since -z <-n> is processed digitally, it is
necessary to calculate 1 in an analog manner. Here, the input signal Y2-m (632) converted into
the thermometer code by the formatter is once input to the selection circuit (910), and is
connected corresponding to each bit of the output from the selection device. A buffer (901) and
11-04-2019
15
one end are connected to a drive buffer, and a resistive element (902) connected to an output
terminal is formed by a unit group so that one end adds a voltage in an analog manner. In order
to remove the variation of the drive buffer (901) and the resistance element (902) by the
mismatch shaping method, the selection circuit (910) uses the output (921) of the selection
circuit (910) as a delay element and an adder. The frequency of use of the unit constituted by the
drive buffer (901) and the resistance element (902) is calculated by the constituted integration
circuit (911) and integration circuit (912), and selected in ascending order of usage frequency. It
is characterized by operating as follows.
[0056]
According to this embodiment, when performing mismatch shaping, the selection circuit is
switched instead of simply using the DEM method using a random signal, so a random signal that
causes a problem when using the DEM method is caused It is not necessary to introduce white
noise and to introduce a circuit for switching the selection circuit.
[0057]
Third Embodiment FIG. 10 shows a third embodiment of the digital-to-analog converter
according to the present invention.
In this embodiment, it is assumed that the first stage of the cascade ?? modulator is a 1-bit
internal modulator, and the second stage is an n-bit internal modulator. As shown in Equation 10,
when the order of the internal modulator is quadratic, the transfer function of the second stage is
(1?z <?n>) и (1?z <?1>). (1Since -z <-n> is processed digitally, it is necessary to calculate (1-z
<-1>) in an analog manner. Here, the input signal Y2-m (632) converted to the thermometer code
by the formatter is once input to the selection circuit (1010), and corresponds to each bit of the
output (1020) from the selection device. It is input to the post filter unit (603).
[0058]
The post filter unit (603) is connected to the input via a delay element (711) composed of a DFF
that performs one clock delay and a switch (715a) controlled by a signal 00 obtained by dividing
one clock. Also, a signal obtained by dividing one clock in the same manner as a resistance
element (713) connected to the drive buffer (712) and one end to the drive buffer and one end
connected to the output terminal so as to perform weighted addition of voltages in an analog
manner. The drive inverter (714) and one end connected to the output via the switch (715a)
controlled by ? 0 are connected to the drive inverter, and one end is connected to the output
11-04-2019
16
terminal so that the voltage is weighted and added in an analog manner. An inverter (714)
connected to the input and output of the delay element (711) configured of the resistive element
(715) and the DFF performing one clock delay, and the output of the inverter divided by one
clock Controlled by the signal ?? 1 That is composed of a switch (715b).
[0059]
Here, the connection between the input / output of the delay element (711), the drive buffer
(712) and the drive inverter (714) is connected by switching between the input / output with the
switches (715a) and (715b).
Since the switches (715a) and (715b) are controlled by the signals 00 and 11 obtained by
dividing one clock, a swapping circuit in which the resistance element is swapped for each clock
is configured.
[0060]
The selection circuit (1010) comprises an output (1021) of the selection circuit (1010) with a
delay element and an adder in order to further eliminate the variation between the units (603) of
the post filter by the mismatch shaping method. The integration circuit (1011) is characterized in
that it operates to calculate the frequency of use of the unit (603) of the post filter and select the
frequency of use in ascending order. As in the second embodiment, it is possible to increase the
order of mismatch shaping by repeatedly using the integrating circuit (1011) for controlling the
selection circuit (1010) used in the mismatch shaping method.
[0061]
According to this embodiment, high-order mismatch shaping can be easily realized by
superimposing mismatch shaping by the mismatch shaper and mismatch shaping by the
swapping circuit. The high-order mismatch shaping function, which conventionally had a large
hardware size, can be realized with a small amount of hardware. For example, second-order
mismatch shaping can be realized by using DWA (Data Weighted Averaging) for the mismatch
shaper.
11-04-2019
17
[0062]
In the first to third embodiments, an example is shown in which the first stage of the cascaded
?? modulator is a 1-bit internal modulator, and the second stage is an n-bit internal modulator.
The effect of the present embodiment can be similarly realized using the configuration of any
internal modulator composed of cascaded ?? modulators.
[0063]
Fourth Embodiment A fourth embodiment of a digital-to-analog converter according to the
present invention is shown in FIG. 11a.
In this embodiment, it is assumed that the ?? modulator has an output of n bits. In this
embodiment, the signal Y2-m obtained by converting the n-bit output of the ?? modulator
(1101) into a thermometer code by the formatter (1102) is mismatch shaped by the post filter
(1103), and the output is An analog addition is performed via the drive buffer circuit (1104) and
the resistance element (1105).
[0064]
FIG. 11 b shows an embodiment of the post filter (1103). In order to remove the variations of the
drive buffer circuit (1104) and the resistance element (1105) by the mismatch shaping method,
the selection circuit (1110) is configured of the output of the selection circuit (1110) by a delay
element and an adder. The frequency of use of the output signal is calculated by the integration
circuit (1111) and the integration circuit (1112), and the operation is performed so as to select in
order of the frequency of use. Here, the integration circuit performs the operation with the input
signal as an m-bit vector signal.
[0065]
In this embodiment, in order to remove the digital signal modulated by the ?? modulator by the
mismatch shaping method by the post filter using the integrating circuit when driving the
plurality of speakers by the plurality of driving circuits, There is no need to superimpose white
noise caused by a random signal or introduce a circuit that switches the selection circuit at high
speed, which is a problem when using a mismatch shaping method based on a certain DEM
method.
11-04-2019
18
[0066]
In this embodiment, an example in which a plurality of resistance elements are driven to add
voice in an analog manner is shown, but the present invention can be applied to all methods in
which a plurality of driving devices add in an analog manner.
[0067]
In this embodiment, the integration circuit (1110) for controlling the selection circuit (1110)
used in the mismatch shaping method is repeatedly used twice, but the mismatch shaping effect
is repeatedly used the integration circuit (1110) one or more times. It can be obtained by
[0068]
FIG. 12 shows another embodiment of the post filter (1103).
In order to eliminate variations between the speaker drive devices by the mismatch shaping
method, the selection circuit (1110) comprises an integration circuit (1111) comprising an
output of the selection circuit (1110) and a delay element and an adder and an integration
circuit. The frequency of use of the output signal is calculated by (1112), and it operates so as to
select in ascending order of frequency of use. Furthermore, the control circuit (1201) provided
between the input / output of the integration circuit and the selection circuit The output signal
selected by the selection circuit is limited according to the magnitude of the amplitude.
The control circuit (1201) operates such that the limited output drive circuit is selected when the
amplitude of the input digital signal is small, and controls such that all the output drive circuits
are selected when the amplitude is large. Do.
As a result, since only one output drive circuit is selected when the signal of small amplitude is
used, it is possible to suppress the influence of variations between the drivers at the time of small
amplitude, and at the time of small amplitude. Since sound is emitted only from a specific drive
device, localization of the sound image is improved. Further, by dynamically controlling the
number of drive of the drive device according to the amplitude of the input signal, it becomes
possible to optimize the power consumption consumed by the drive device. As described in the
first to fourth embodiments, the feature of the present invention is that noise shaping is
11-04-2019
19
performed on the digital input signal by the ?? modulator, then division is performed by the
formatter and mismatch shaping is performed by the post filter. The driving circuit is driven to
perform addition in an analog manner.
[0069]
As a result, even if the power for driving the plurality of driving devices is small, it is possible to
obtain a large output by adding in an analog manner.
[0070]
As described in the first to fourth embodiments, the digital-to-analog converter for converting
digital signals into analog signals converts the digital audio signals into a plurality of digital
signals and analogizes the outputs of a plurality of driving devices It is possible to apply to all the
devices which add.
[0071]
Fifth Embodiment FIG. 13A shows a fifth embodiment in which the digital-to-analog conversion
devices shown in the first to fourth embodiments are configured to be added by a current.
In this embodiment, the drive buffer and the resistance element which are the components of the
previous embodiments are respectively digitalized with a current source (1300), a switch circuit
(1302) provided between the current source and the output, and the switch. A configuration in
which a buffer circuit (1301) controlled by signals is replaced is shown.
[0072]
Sixth Embodiment FIG. 13 b shows a sixth embodiment in which the digital-to-analog conversion
devices shown in the first to fourth embodiments are added by sound pressure in FIG. 13 a. Is
shown.
In this embodiment, the drive buffer and the resistance element, which are components of the
previous embodiments, are respectively controlled by a speaker device (1310) and a voice coil
(1312) for driving the speaker device and the voice coil by digital signals. The configuration is
11-04-2019
20
shown in which the drive circuit (1311) is replaced.
[0073]
Seventh Embodiment Furthermore, FIG. 13 c shows the seventh embodiment in the case where
the digital-analog conversion devices shown in the first to fourth embodiments are added by light
in FIG. 13 a. It shows. In this embodiment, the drive buffer and the resistance element which are
the components of the previous embodiments are respectively replaced with a light emitting
element (1320) and a drive circuit (1321) for controlling the light emitting element by digital
signals. There is. In the present embodiment, as the light emitting element, any device capable of
emitting light by electric power such as a lamp or an LED can be used.
[0074]
Eighth Embodiment FIG. 14A shows an eighth embodiment in which the digital-to-analog
conversion devices shown in the first to fourth embodiments in FIG. 13A are configured by
addition using piezoelectric elements (piezo elements). An example is shown. In this embodiment,
the drive buffer and the resistive element which are the components of the previous
embodiments are respectively replaced with a piezoelectric element (1400) and a buffer circuit
(1401) for controlling the piezoelectric element with digital signals. It shows. Since the
piezoelectric element can convert an electrical signal into a physical displacement force, by
arranging a plurality of piezoelectric elements on a plane as in Example 14b (1410), the physical
displacement amount is converted to a sound wave in space. The present invention can be used
for applications such as combining, driving by adding a common vibration plate, and laminating
(1420) and adding a plurality of piezoelectric elements as in Example 14c.
[0075]
Since each piezoelectric element is driven by a 1-bit signal, it is possible to improve the power
efficiency and to reduce the influence of non-linearity of the piezoelectric element.
[0076]
In the present embodiment, the piezoelectric element is shown as means for converting an
electrical signal into a physical displacement force. However, any element capable of converting
an electrical signal into a physical displacement force can be used.
11-04-2019
21
[0077]
When a plurality of piezoelectric elements are stacked (1420) and added as in Example 14c
above, the strength of the physical displacement generated by each piezoelectric element can be
measured using another piezoelectric element It is.
That is, since variations in the strength of physical displacement generated by a plurality of
piezoelectric elements can be measured, the physical force generated by the plurality of
piezoelectric elements can be obtained by adjusting the driving force of the piezoelectric
elements according to the measured variations. It is also possible to improve the accuracy of
combined physical displacement by adding various displacements.
[0078]
Ninth Embodiment FIG. 15A shows a ninth embodiment in which the digital-analog conversion
devices shown in the first to fourth embodiments are configured to add magnetic fields
generated by coils.
In this embodiment, the drive buffer and the resistance element, which are components of the
previous embodiments, are respectively replaced with a buffer circuit (1501) for controlling the
coil (1500) and the front coil with digital signals. Since coils can convert electric signals into
magnetic field force, a plurality of coils are stacked and arranged (1510) as in Example 15b to
add magnetic fields or as in Example 15c, It can also be used in applications where the magnetic
field is added by simultaneously winding 1520 coils.
[0079]
As described above, since the magnetic fields can be added, the present invention can also be
applied to a digital-to-analog converter that reproduces an audio signal using a speaker drive
device using a plurality of voice coils. .
[0080]
11-04-2019
22
As in the above embodiments, it is also possible to measure the strength of the magnetic field
generated by each coil using another coil.
That is, since variations in the strength of the magnetic field generated by the plurality of coils
can be measured, the driving force of the coils is adjusted according to the measured variations
to combine the magnetic fields of the plurality of coils. It is also possible to improve the accuracy.
[0081]
Tenth Embodiment FIG. 16A shows a tenth embodiment in which the digital-to-analog converter
shown in the first to fourth embodiments is applied to a speaker driving device using a plurality
of voice coils. In this embodiment, a configuration is shown in which the resistive element which
is a component of the previous embodiments is replaced with a voice coil (1600). Since the voice
coil can convert an electrical signal into sound pressure by means of a cone (1601) or a dome,
the sound pressure is added by overlapping and arranging a plurality of coils (1610) as in the
embodiment 16a. Becomes possible. Further, in the method using the voice coil, the portion
emitting the audio signal is one cone (1601) or dome, so that the localization of the sound image
is improved.
[0082]
Further, as shown in FIG. 16b, the present invention can also be used in applications where
sound pressure is added by bundling and winding a plurality of voice coils (1620). By bundling
and winding a plurality of voice coils, respective voice coil characteristics can be made uniform.
Thereby, the error of the characteristic between voice coils is reduced, and it becomes possible to
reproduce a high-quality sound signal.
[0083]
As in the above embodiments, it is also possible to measure the strength of the magnetic field
generated by each voice coil using another voice coil. That is, since the variation in the strength
of the magnetic field generated by the plurality of voice coils can be measured, the driving force
of the voice coil is adjusted according to the measured variation to combine the magnetic fields
11-04-2019
23
of the plurality of voice coils. The accuracy of the voice signal is improved, and the voice signal
can be reproduced with high sound quality.
[0084]
In the above fifth to tenth embodiments, a plurality of driving devices are added in an analog
manner and output using the formatter and the post filter with the n-bit output from the
cascaded ?? modulator. Since the n-bit signal is converted into a thermometer code of m = 2
<n> by the formatter, 2 <n> post-filters and drive circuits are required. Here, by setting m = 2 <n>
= 16 or less, it is possible to suppress an increase in the circuit scale of the mismatch shaping
circuit and the swapping circuit. Similarly, by setting m = 2 <n> = 16 or less, as shown in the
embodiment shown in FIG. 14c, variation in the difference in the characteristics due to the
difference in the stacking order of the elements when the piezoelectric elements are stacked. Can
be reduced. Further, as in the embodiment shown in FIG. 15b and FIG. 16a, it is possible to
suppress the variation in the difference in the characteristics caused by the difference in the
stacking order of the coils when the coils are stacked. Furthermore, even in the embodiment of
bundling the coils as shown in FIG. 15c and 16b, it becomes possible to suppress the difference
in the characteristics of each coil.
[0085]
Eleventh Embodiment An eleventh embodiment according to the present invention is shown in
FIG. 17 using a modulator having another configuration than the cascaded ?? modulator of the
digital-to-analog converter shown in the first to fourth embodiments. Show. In order to transmit
the input signal to the output (1711) of the first stage of the cascaded delta-sigma modulator
(1700), a configuration using coefficients B0 (1720) and B1 (1730), respectively, is also possible.
When such a connection is made, part of the input signal is output also from the output of the
second and subsequent ?? modulators, so the number of stages of the cascade ?? modulator
may be increased, and a plurality of speaker devices may be used. In applications where sound
pressure is added in space by using it, it becomes possible to improve output sound pressure as
the number of speaker devices increases.
[0086]
(Twelfth Embodiment) The sound pressure is added in space by using a plurality of driving
11-04-2019
24
devices as described in the sixth embodiment, the eighth embodiment and the tenth embodiment
in the digital-to-analog converter shown in the first to fourth embodiments In such an
application, FIG. 18 shows a twelfth embodiment in which a block (1802) for digitally delaying a
signal for driving a driving device (1801) is inserted. It is possible to change the directivity of the
acoustic signal radiated in space by controlling the phase shift of the signal to each drive by thus
digitally delaying from the ?? modulator and the formatter. Become.
[0087]
For example, assuming that the distance between the speakers is d, the wavelength of the signal
is ?s, and the declination is ? when the front of the speaker is 0 radian, the phase of SP2 is
delayed by (2?d sin ?) / ?s with respect to SP3. By setting the phase of SP1 to (4?d sin ?) /
?s, it becomes possible to give the directivity characteristic on the SP1 side by only ?.
[0088]
In order to control the phase of a plurality of speakers in this manner, conventionally, an analog
phase shifter having a complicated structure is required, but since the input / output signal is a
digital signal, a digital delay (such as DFF) It is possible to easily control the exact phase shift
using
[0089]
(Example 13) Sound pressure is added in space using a plurality of driving devices as described
in Example 6, Example 8, and Example 10 for the digital-to-analog conversion devices shown in
the first to fourth examples. In such an application, FIG. 19a shows a thirteenth embodiment for
feeding back ambient noise as an input to a digital to analog converter.
The feedback control circuit (1900) is necessary to generate a signal whose phase is 180 degrees
out of phase with noise that cancels out ambient noise based on ambient noise information from
the microphone (1901) to which ambient sound is input. Calculate sound pressure and phase.
According to the present invention, it is possible to directly control the speaker with a digital
circuit, so it is possible to configure a precise noise reduction device. Further, as shown in FIG.
19b, since it is possible to generally detect the direction of the noise source by using a plurality
of microphones, the phase to each speaker driving device is controlled using the technique of the
twelfth embodiment. This allows the noise reduction speaker to have directivity. That is, not only
11-04-2019
25
the front direction of the noise reduction speaker but also noise in other directions can be
reduced.
[0090]
When noise reduction is performed in a car, a plurality of external noise sources may vary and
noise sources may vary, but a plurality of noise reduction speakers can be easily arranged by
using this embodiment. In addition, noise in directions other than the front can be reduced by
using a plurality of speakers, so that the interior of the vehicle can be muffled efficiently.
Furthermore, using a piezoelectric speaker makes it possible to realize a thin noise suppressor, so
noise can be silenced without reducing the space in the vehicle.
[0091]
(Example 14) The sound pressure is added in space using a plurality of driving devices as in
Example 6, Example 8, and Example 10 for the digital-to-analog converter shown in the first to
fourth examples. In such an application, FIG. 20 shows a fourteenth embodiment in which the
drive device (2000) is configured of a switching amplifier. As the switching amplifier, an analog
class D amplifier, a digital class D amplifier, an analog ?? modulator, a digital ?? modulator,
or the like can be used. Since the input digital signal is converted to a switching signal (binary
signal or ternary signal) by the switching amplifier, it is possible to improve the efficiency and
the output power.
[0092]
(Fifteenth Embodiment) The sound pressure is added in space using a plurality of driving devices
as described in the sixth embodiment, the eighth embodiment, and the tenth embodiment in the
digital-to-analog converter shown in the first to fourth embodiments. A fifteenth example of a
loudspeaker (2100) in such an application and a method of arranging electrical elements capable
of converting an electrical signal into a physical displacement force is shown in FIG.
[0093]
FIG. 21a shows an embodiment in the case of a grid arrangement.
11-04-2019
26
With such an arrangement, when covering a rectangular, square, etc. casing, subunits can be
arranged efficiently, and the horizontal and vertical directions are similar, and equivalent phase
characteristics can be realized. It can. In addition, when a rectangular or square speaker is used,
it is possible to arrange the rectangular surface most closely, and to maximize the radiation
sound pressure per unit area. Those arranged in this way are visually beautiful.
[0094]
FIG. 21b shows an embodiment in which the arrangement position is shifted by half for each row.
By arranging in a staggered manner in this manner, the surface density can be improved as
compared to the lattice arrangement. In particular, when a large number of speakers are
arranged, it is possible to increase the sound pressure per area. Furthermore, if hexagonal shapes
are used for the shape of the speaker, they can be arranged in a staggered arrangement and
without gaps. In this case, since it becomes possible to arrange without gaps, a high sound
pressure level can be realized. When the mismatch shaping technology is used, the distance
between the speakers is short, so that the mismatch shaping effect can be effectively realized.
[0095]
FIG. 21b shows an embodiment in which the loudspeakers are arranged concentrically. Thus,
since the distance of the speakers arranged on each concentric circle is equal from the central
axis of the whole speaker, the phase characteristics from the same concentric circle to the central
axis become equal, and the addition of the acoustic signals in the front is ideal. To be done. For
this reason, it becomes possible to improve acoustic characteristics.
[0096]
(Sixteenth Embodiment) The sound pressure is added in space using a plurality of driving devices
as described in the sixth embodiment, the eighth embodiment, and the tenth embodiment in the
digital-to-analog converter shown in the first to fourth embodiments. In such an application, FIG.
22 shows a sixteenth embodiment in which the digital-to-analog converter is configured in
stereo. Here, channel (2201) is a digital-to-analog converter for R signals, and channel (2202) is
a digital-to-analog converter for L signals. As described above, by providing the digital-to-analog
converter according to the present invention as a plurality of channels in parallel as well as
11-04-2019
27
reproducing the stereo sound of the digital signal, it becomes possible to reproduce an arbitrary
sound field generated by digital processing.
[0097]
(Seventeenth Embodiment) As shown in the sixteenth embodiment, sound pressure is added in
space using a plurality of digital-to-analog converters using a plurality of driving devices as in the
sixth, eighth, and tenth embodiments. A seventeenth embodiment relating to a speaker (2100) in
such an application and a method of arranging electric elements capable of converting an electric
signal to a physical displacement force is shown in FIG.
[0098]
FIG. 23a shows how to arrange speakers driving stereo L and R signals.
By arranging L and R symmetrically as described above, it is possible to enhance the stereo
effect. In the figure, L represents the left channel and R represents the right channel. FIG. 23b
shows a method of arranging speakers driving C signal in addition to stereo L and R signals. C in
the figure indicates a center channel. In the present invention, the assignment of channels to a
plurality of speakers can be easily changed dynamically, but it is possible to dynamically change
the assignment of channels by the music source to be reproduced and the sound field effect to be
realized. It is possible to perform stereo effects and sound field effects more effectively. FIG. 23c
shows a speaker arrangement method in the case of dynamically controlling assignment of a
plurality of speakers to channels. It shows how to arrange speakers driving C signal in addition to
stereo L and R signals. In the figure, L / C shows a speaker capable of driving both L signal and C
signal, and in the figure, R / C shows a speaker capable of driving both R signal and C signal.
[0099]
(Example 18) The sound pressure is added in space using a plurality of driving devices as in
Example 6, Example 8, and Example 10 with the digital-to-analog converter shown in the first to
fourth examples. In such an application, FIG. 24 shows an eighteenth embodiment in which a
digital filter processed digital signal is made into a stereo configuration of digital analog
conversion devices of a plurality of channels. Here, a plurality of digital signals obtained by
dividing the frequency band by the digital filter signal processing block (2401) are represented
by a plurality of channels (2402) and (2204) as digital-to-analog converters. For example, the
11-04-2019
28
digital-to-analog converter according to the present invention is provided in parallel as a plurality
of channels in order to divide the digital signal for high band and the digital signal for low band
in the digital filter signal processing block and reproduce each signal. Thus, it is possible to
reproduce with a speaker device that is optimal for the frequency band generated by digital
processing.
[0100]
(Example 19) The sound pressure is added in space using a plurality of driving devices as
described in Example 6, Example 8, and Example 10 for the digital-to-analog converter shown in
the first to fourth examples. In such an application, as shown in FIG. 25, the signal for driving the
driving device is once transmitted to the transmission line by the digital signal transmitter (2501)
and then received by the digital signal receiving device (2502) to drive the speaker by the driving
device. Shows an embodiment of the invention.
[0101]
As described above, by transmitting digital signals from the ?? modulator and the formatter by
the digital signal transceiver, it is possible to transmit a signal for driving the distributed
speakers as a digital transmission signal.
Since the digital signal is oversampled by the ?? modulator, even if there is an error in the
transmission line, the influence can be reduced. As the transmission line, it is possible to use any
transmission line that transmits digitally, such as a digital wired transmission line, a wireless
transmission line, and an optical transmission line.
[0102]
Also, when applied to a noise reduction device, it is necessary to distribute a plurality of noise
reduction speakers, but by using this embodiment, it is possible to transmit drive information to
easily separated sub speakers using a digital transmission line. Is possible.
[0103]
(Twenty-Sixth Embodiment) The sound pressure is added in space using a plurality of driving
devices as described in the sixth embodiment, the eighth embodiment, and the tenth embodiment
in the digital-to-analog converter shown in the first to fourth embodiments. In such an
application, FIG. 26 shows a twentieth embodiment in which an ultra-low frequency signal is
11-04-2019
29
superimposed on a signal for driving a drive device.
[0104]
Generally, the audio frequency is 20 to 20 KHz, and the sound of 20 Hz or less, which is the
lower limit frequency, is called an ultra-low frequency.
Although the sound in this band can not usually be recognized by human hearing unless it has a
considerable sound pressure, studies are being conducted regarding it as being related to health
and mental stress.
[0105]
In order to generate ultra-low frequency using conventional analog speakers, it is necessary to
drive the speakers with very slow signals, and there are many power consumption problems with
analog speakers with poor power efficiency.
If the configuration of the digital speaker according to the present invention is used to generate
an ultra-low frequency, it is possible to drive the electroacoustic transducer with a 1-bit signal,
thereby reducing the power efficiency and the influence of nonlinearity of the electroacoustic
transducer. It is possible to efficiently generate an ultra low frequency signal.
[0106]
In general, since the very low frequency signal is not included in the signal source (broadcast
signal or recording medium), when the very low frequency signal is generated, it needs to be
generated by the very low frequency generator (2600). The ultra low frequency generator should
use a digital circuit (2600) to generate an arbitrary frequency pattern, for example a pseudo
random signal of the digital circuit to generate an ultra low frequency signal with 1 / f
fluctuation It can be easily generated by using. Since the generated very low frequency signal can
be easily added digitally to the digital voice signal, it becomes possible to easily superimpose the
very low frequency signal.
11-04-2019
30
[0107]
(Twenty-first Embodiment) Physical displacements are synthesized using a plurality of
piezoelectric elements as in the eighth embodiment, using the digital-to-analog conversion
devices shown in the first to fourth examples. In such an application, FIG. 27 shows a twenty-first
embodiment in which a plurality of piezoelectric elements are used to drive a reflecting mirror.
27b, a device for driving a plurality of driving devices (2701) using a plurality of piezoelectric
elements (2702) to drive the stacked piezoelectric elements (2700) as shown in FIG. 27a. (2711)
A plurality of pieces are arranged on the upper substrate (2712) so that the support part (2714)
is a fixed shaft. The base of the upper places a reflector (2713). As shown in FIG. 27c, by driving
the piezoelectric element (2700), it is possible to change the reflection angle of the reflecting
mirror (2713) by deforming the upper base (2712) centering on the support part (2714). A
device in which such a piezoelectric element and a reflecting mirror are combined is applicable to
a compact projector device, but since the reflection angle can be controlled by a digital signal by
driving with a digital analog conversion device and a plurality of piezoelectric elements, the
device is compact Suitable for projector devices.
[0108]
(Example 22) A physical displacement is synthesized using a plurality of piezoelectric elements
as in the eighth example using the digital-to-analog conversion devices shown in the first to
fourth examples. In such an application, FIG. 28 shows another twenty-second embodiment in
which a plurality of piezoelectric elements are used to drive a reflecting mirror. As shown in FIG.
28a, a plurality of piezoelectric elements (2800) arranged in parallel on the substrate are used.
As shown in FIG. 28b, this piezoelectric element (2800) is placed around the reflecting mirror
(2701) whose center is fixed and driven along the X-Y axis to drive the reflecting mirror (2801)
with the center of the support as the base point. Can change the reflection angle of A device in
which such a piezoelectric element and a reflecting mirror are combined is applicable to a
compact projector device, but the reflection angle of a digital signal can be obtained by driving
with a digital analog conversion device and a plurality of piezoelectric elements arranged in
parallel. It is suitable for thin and small projector devices because it can control
[0109]
(Twenty-Third Embodiment) FIG. 29 shows a twenty-third embodiment where a bandpass ??
11-04-2019
31
modulator is used as the cascade ? ? ? modulator of the digital-to-analog converter shown in
the first to fourth embodiments. ing. In general, a bandpass ? ? ? modulator can be realized by
performing Z ? ?Z <2> conversion. This conversion converts the integrator into a resonator. In
this embodiment, the output of the internal ? ? ? modulator of the second stage is connected
to a two-clock delay that realizes Z <?2>, and a changeover switch is connected to the input and
output of the delay. This changeover switch switches the two inputs as shown in the figure in
accordance with a signal of 1/2 frequency of the clock. With such a configuration, even when
there is a mismatch in the elements constituting the DAC 21 and the DAC 22, it is possible to
reduce noise at a frequency of 1?4 of the clock frequency.
[0110]
As shown here, by performing frequency conversion, it is possible to realize arbitrary noise
shaping characteristics including band pass characteristics.
[0111]
Twenty-Fourth Embodiment FIG. 30A shows a twenty-fourth embodiment of the present
invention.
In this embodiment, it is assumed that the ?? modulator has an output of n bits. The n-bit
output of the ? ? ? ? ? ? modulator (2401) is converted into m sets of p-bit codes by the
formatter (2402), mismatch shaping and frequency selection are performed by the post filter
(2403), The output is converted to an analog signal by an internal digital-to-analog converter
(2404), and is added in an analog manner by an adder (2405). With this configuration, it is
possible to obtain a high-precision analog signal even using a multi-level internal digital-toanalog converter.
[0112]
FIG. 30 b shows an embodiment of the post filter. In order to reduce the influence of the
mismatch inside the internal digital-to-analog converter, the selection circuit (2410) selects the
output of the selection circuit (2410) according to the value of the output signal of the filter
circuit (2411). It is characterized by operating. Here, in the filter, the filter operation is
performed for each output level of the internal digital-to-analog converter. For example, using a
filter in which integrators or integrators are connected in multiple stages, selection is performed
11-04-2019
32
in ascending order of filter output, and selection is performed so that an output corresponding to
the input signal can be obtained by the selection. Even if the output from the formatter is due to
a plurality of signals representing a plurality of levels, it is possible to reduce noise in the low
frequency region due to mismatch.
[0113]
FIG. 30c shows a more specific embodiment of the internal digital-to-analog converter (2404)
and the adder (2405). In this embodiment, an analog current corresponding to each 1-bit signal
is output by an inverter (2421) and a resistor (2422), and a plurality of these currents are
connected to add output currents. In this embodiment, the values represented by the input
signals of the plurality of internal digital-to-analog converters need not be identical but may have
different weights. In this case, the value of the resistor (2422) may be set according to the weight
represented by each input digital signal. Also, this weight is not limited to the power of two. By
selecting in the selection circuit (2410) such that the selection result is equal to the input signal
of the selection circuit (2410), it becomes possible to perform conversion accurately even when
the weights are different.
[0114]
It is an example of a digital-to-analog converter using a ? ? modulation circuit. It is an example
of a cascaded ?? modulator. This is an example of a structure in which an analog FIR filter is
added to a cascade ?? modulator. It is another example of the structure which attached an
analog FIR filter to a cascade type delta-sigma modulator. It is an example of the block diagram
of the structure which attached an analog FIR filter to a cascade type delta-sigma modulator. It is
an example of a digital analog conversion device using a cascade ?? modulator of the present
digital analog conversion device according to the first embodiment of the present invention. It is
a circuit block diagram of a 1st Example. It is a simulation result of the effect of the digital analog
converter using the cascade type delta-sigma modulator of the present digital analog converter
according to the first embodiment of the present invention. It is a block diagram of a 2nd
Example. It is a block diagram of a 3rd Example. It is a block diagram of a 4th Example. It is a
circuit block diagram of a 4th Example. It is a block diagram of a 5th Example. It is a block
diagram of a 6th Example. It is a block diagram of a 7th Example. It is a block diagram of an 8th
Example. It is a block diagram of a 9th Example. It is a block diagram of a 10th Example. It is a
block diagram of 11th Example. It is a block diagram of a 12th Example. It is a block diagram of
a 13th Example. It is a block diagram of a 14th Example. It is a block diagram of a 15th Example.
It is a block diagram of the sixteenth embodiment. It is a block diagram of a 17th Example. It is a
11-04-2019
33
block diagram of the eighteenth embodiment. It is a block diagram of the nineteenth
embodiment. It is a block diagram of the twentieth embodiment. It is a block diagram of the
twenty-first embodiment. It is a block diagram of the twenty-second embodiment. It is a block
diagram of the twenty-third embodiment. It is a block diagram of the twenty-fourth embodiment.
11-04-2019
34
cted to serial-to-parallel conversion are
used as they are as signals for driving a plurality of speakers, firstly, manufacturing variations
among current sources of weighted drive circuits are nonlinear noises. Second, there is a problem
that quantization noise generated when reproducing a digital signal is superimposed as a noise
component in an audio frequency band, etc., so that high-quality audio signals can be
reproduced. It has the drawback of being difficult.
[0006]
In order to avoid the first problem, it is necessary to have means for suppressing manufacturing
variations among the plurality of drive devices.
11-04-2019
2
[0007]
In FIG. 33 of US Pat. No. 5,872,532, a technique comprising a selection circuit and an integrator
for controlling the selection circuit is proposed as a means for suppressing variations between
current sources driving a plurality of speaker drive devices.
In this proposal, a plurality of loudspeakers are controlled by inputting a signal for driving a
plurality of loudspeakers to the selection device and controlling the circuit for integrating the
presence or absence of the plurality of loudspeaker drive circuits one or more times. The
frequency of use of each of the drive devices is integrated, and the selection circuit is controlled
so as to keep the integration result constant.
This makes it possible to reduce noise due to manufacturing variations among the drive devices.
A technique for suppressing variations among a plurality of drive devices is called a mismatch
shaping method.
[0008]
A method is proposed in FIG. 1 of US Pat. No. 5,592,559, in which the input digital serial audio
signal is once subjected to digital modulation using a ?? modulator to drive a voice coil to
reproduce voice. This conventional example is a proposal for driving a speaker in the positive
and negative directions of two voice coils using a digitally modulated ternary signal, but it is
possible to drive two or more voice coils and to drive a plurality of driving devices. There is no
mention of technology to reduce the variation in
[0009]
USP 7,058,463 Fig. In No. 3, it has been proposed to discharge the input digital serial voice
signal to a frequency higher than the audio frequency by applying digital modulation using a
?? modulator and oversampling. A technique that spouts quantization noise out of the
frequency of interest in this way is called a noise shaping method. In this conventional example,
quantization noise generated when reproducing a digital signal is moved to a high frequency
11-04-2019
3
band outside the audible frequency using a noise shaping method. This avoids the problem of the
second problem of quantization noise being superimposed as a noise component in the audio
frequency band.
[0010]
Further, in the conventional example, in order to avoid the problem of noise caused by
manufacturing variations among a plurality of driving devices, which is the first problem, the
control is performed by the DEM (Dynamic Element Matching) method using a pseudo random
signal. It is proposed to introduce a mismatch shaping method using a selection circuit.
[0011]
However, since the speaker drive circuit is driven as it is without attenuating the quantization
noise emitted to a frequency higher than the audio frequency by applying the digital modulation
using the ? ? ? modulator and oversampling, the high frequency band There is a problem that
the quantization noise moved to is emitted from the speaker.
[0012]
Also, simply switching the selection circuit by the DEM method using a random signal has the
disadvantage that white noise caused by this random signal is superimposed on the reproduced
audio signal.
In order to avoid the problem of noise caused by manufacturing variations among a plurality of
drive devices, it is necessary to operate the switching operation of the selection circuit by the
DEM method at high speed as the number of speaker drive circuits increases.
The details of the operation of the DEM method are described in Section 8.3.3 of the reference
"Delta-Sigma Data Converters" IEEE Press 1997 ISBN 0-7803-1045-4 and Figure 8.5. In the
mismatch shaping method using the DEM method, the selection circuit needs to operate at high
speed is a serious drawback in the implementation of this conventional example. Incidentally, this
drawback is already pointed out as a problem in US Pat. No. 5,872,532 and is known.
[0013]
11-04-2019
4
As in the above conventional example, by using the noise shaping method by the digital
modulation using the ? ?? ? modulation circuit and the oversampling, the quantization noise
generated by reproducing the digital signal is emitted to the frequency band higher than the
audio frequency. Things are generally well known techniques. Reference "Over sampling DeltaSigma Data Converters" IEEE Press 1991 ISBN 0-87942-285-8 pp. Equation (22) of 7 shows the
relationship between the oversampling ratio and the strength of the noise shaped noise with
respect to the order of the modulator. Generally, by noise shaping, the effective intensity of
quantization noise decreases by 3 (2L + 1) dB every time the oversampling ratio is doubled,
where L is the order of the ?? modulator. Therefore, to reduce quantization noise, the
oversampling ratio must be increased or the order of the ?? modulator must be increased. On
the other hand, when the oversampling ratio is increased, it is necessary to operate the ??
modulator at high speed. Further, if the order of the ?? modulator is increased, the operation of
the ?? modulator becomes unstable.
[0014]
As described above, in the noise shaping method by digital modulation using the ?? modulation
circuit and oversampling, quantization noise generated by reproducing a digital signal is
discharged to a frequency band higher than an audio frequency. Therefore, it is necessary to
attenuate the noise-shaped unnecessary quantization noise generated in the ?? modulation
circuit and the component outside the audio frequency band with the continuous time LPF
(Continuous-Time Low Pass Filter).
[0015]
FIG. 1A shows an example of a general system using a ?? modulation circuit. The noise-shaped
unnecessary quantization noise and out-of-band components generated in the ?? modulator
(100) are attenuated by the continuous time LPF (101). Since oversampling is performed, the
LPF may be a low-order one, but when the passband is narrow, the time constant becomes large,
and the area occupied by the LPF can not be ignored when built in a semiconductor integrated
device.
[0016]
11-04-2019
5
As shown in FIG. 1 (b), there is a method of reducing the characteristic request of the LPF, which
is disposed downstream of the modulator, as a multi-bit ?? modulator (110) as shown in FIG. 1
(b). In this case, since the quantization noise can be reduced by 6 dB by increasing the number of
bits of the ?? modulator by 1 bit, the cutoff frequency characteristic of the LPF can be relaxed.
However, increasing the number of bits of the modulator increases the circuit size of the internal
modulator.
[0017]
As another method of relaxing the characteristic request of the LPF, a method of inserting the
switched capacitor filter (121) shown in FIG. 1C between the ?? modulator and the LPF is also
proposed. In this case, in addition to the need for an OP amplifier to realize a switched capacitor
filter, a large capacitor may be needed to lower the cutoff frequency, which increases chip area
and power consumption. There is a drawback.
[0018]
As another method of reducing the characteristic requirement of the LPF, a method of inserting
the analog FIR filter (131) shown in FIG. 1 (d) between the ?? modulator and the LPF has been
proposed. In this method, an analog FIR filter is configured by adding each tap of the FIR filter in
an analog manner to be an output. In this case, the amount of attenuation for out-of-band noise
can be increased by increasing the number of taps. The method using an analog FIR filter also
has the effect of reducing the degradation of the SNR due to clock jitter, and is an effective
method when using a clock signal with low accuracy or when using multiple clocks on the same
chip.
[0019]
However, when the ?? modulator has multiple bits, the delay elements constituting the analog
FIR filter are required for the number of cells of the segment type modulator constituting the bits
of the ?? modulator О the number of taps. There is a disadvantage that the circuit scale
increases rapidly.
[0020]
The operation will be described in more detail with respect to a method of postfixing an analog
FIR filter to a system using a general noise shaping method using a ?? modulation circuit,
11-04-2019
6
particularly when using a cascade ?? modulator.
[0021]
First, FIG. 2 shows a general configuration of the cascade ?? modulator (200).
The input digital signal (210) is quantized by the first stage ?? modulator (201), and the first
stage quantization noise (211) is further quantized by the second stage ?? modulator (202). Be
done.
The output Y2 of the second stage is converted by the digital signal processing block (220), and
then the output of the first stage is added with Y1 (230) and output.
[0022]
The output of the first stage Y1 and the output of the second stage Y2, the noise transfer
function of the first and second stages NTF1 (z), NTF2 (z), the quantization noise of the first and
second stages Assuming that the gain from the first stage to the second stage is A1 and H3 =
NTF1 (z) / A1, the total output Y is Y = Y1 + Y2H3 = Y1 + Y2NTF1 / A1 = X + NTF1Q1 + (?
A1Q1 + NTF2Q2) NTF1 / A1 = X + NTF1Q1-NTF1Q1 + NTF1NTF2Q2 / A1 = X + NTF1NTF2Q2 /
A1 (Equation 1) The first stage quantization noise can be canceled out.
[0023]
A general configuration (300) in which an analog FIR filter (301) is added to this cascaded delta
sigma modulator is shown in FIG.
[0024]
This configuration can also be converted to a configuration (400) in which an analog FIR filter is
placed after each stage of the cascade ?? modulator as shown in FIG.
The operation of the second stage in the configuration in which the analog FIR filter is disposed
at each stage of the cascade ?? modulator as shown in FIG. 4 will be described in detail below.
11-04-2019
7
[0025]
The signal from Y2 is multiplied by H3 (z) in the digital signal processing block (220) and then
multiplied by the transfer function HFIR (z) of the FIR filter (300).
[0026]
Now, consider the case where the first-stage ? ? ? modulator and the FIR filter are moving
average filters.
The transfer function of the FIR filter is defined as H3 (z) = NTF1 = (1-z <-1>) HFIR (z) = 1 + z <1> + z <-2>... + Z <-(n-1)> .. ... (Equation 2) H3HFIR = (1-z <-1>) (1 + z <-1> + z <-2>... + Z <-(n-1)>)
= 1-z <-n And can be configured by a 2-tap post filter regardless of the number of taps of the FIR
filter.
In other words, when an analog FIR filter is added to the cascade ?? modulator, the number of
taps of the second stage after put filter is always 2 taps by using the configuration shown in FIG.
Even if the number is increased, the number of taps of the post filter does not increase and it is
suitable for miniaturization.
[0027]
Similarly, consider a configuration in which the first stage is a second-order ?? modulator and
the FIR filter is a moving average filter. Since H3 = NTF1 = (1-z <-1>) <2>, H3HFIR = (1-z <-1>)
<2> (1 + z <-1> + z <-2>... + Z <- n-1)>) = 1?z <?1> ?z <?n> + z <? (n + 1)> (Equation 4), and
the number of taps in the second stage post-filter There are four taps regardless of the tap length
of the FIR filter.
[0028]
In other words, when an analog FIR filter is added to a cascade ?? modulator, the second stage
can be obtained even if the number of taps of the FIR filter is increased regardless of the order of
the ?? modulator by adopting the configuration of FIG. It is possible to suppress an increase in
the number of taps of the post filter, and it is understood that it is suitable for miniaturization.
11-04-2019
8
[0029]
Note that YFIR in the case where an analog FIR filter is added to a cascade ?? modulator is
YFIR = (1 + z <?1> + z <?2>. (Equation 5)
U.S. Pat. No. 5,862,237 U.S. Pat. No. 5,909,496 U.S. Pat. No. 5,872,532 U.S. Pat. No. 5,592,559
U.S. Pat. No. 7,058,463 U.S. Pat.
[0030]
As described above, FIG. 5 shows a general block diagram in the case where an analog FIR filter
is placed after the modulator of each stage of the cascaded ?? modulator. Here, for
convenience of explanation, the number of taps of the FIR filter is n, the noise transfer function
of the first stage of the cascade ?? modulator is NTF1 = (1-z <-1>) <2>, and the first stage is 1
bit The second stage consists of an n-bit internal modulator.
[0031]
Here, the digital input signal (510) is input to the first stage internal modulator (201) of the
cascade ?? modulator, and the second stage internal modulator (202) is the first stage internal
modulator 201) is cascade-connected, and an output signal (520) from the first internal
modulator (201) is input to the analog FIR filter (301). The output signal (530) from the second
stage internal modulator (202) is converted and output from the binary code to the thermometer
code by the formatter circuit (501). The signal (531) converted to the thermometer code is input
to the post filter circuit (502). An output signal (521) from the analog FIR filter (301) and an
output signal (532) from the post filter circuit (502) are added in an analog form by a summing
block (540) and output.
[0032]
In a cascade ?? modulator using an analog FIR filter, consider the influence when tap
coefficients forming the analog FIR filter have an error.
11-04-2019
9
[0033]
If the first stage internal modulator is configured with one bit, the mismatch causes a tap
coefficient error to affect the frequency characteristics of the analog FIR filter.
However, since the linearity from the digital input to the analog output is not affected, the
distortion characteristics and the SNR are not degraded.
[0034]
On the other hand, when the first stage internal modulator has three or more levels, the
mismatch of the analog FIR filter unit directly affects the output as in a general ?? modulator,
and the distortion and the SNR characteristic are degraded, so one stage In order to increase the
number of levels of the internal modulator in the eye, a separate mismatch shaper is required.
[0035]
Mismatches of the elements that make up the second stage post-filter also affect the output, but
the SNR is degraded because the second stage input signal is the first stage quantization noise,
but the signal component is included. If not, the distortion characteristics do not deteriorate.
[0036]
Here, let us calculate the influence of the tap coefficients of the analog FIR filter and the post
filter on the output YFIR.
[0037]
Here, when the first and second internal modulators both have two levels and NTF1 = NTF2 = (1z <-1>) <2>, the characteristic of the analog FIR filter is H1FTR, post-filter Let H 2 FTR be the
characteristic of the first stage and let b 0, b 1,..., B n be the first stage tap coefficients a 0, a 1,. z)
= H1FTRY1 + H2FTRY2 = H1FTR {X + NTF1Q1} + H2FTR {-Q1 + NTF2Q2 / A1} = (a0 + a1z <-1> +
a2z <-2>... + an1z <-(n-1)>) {X + (1-z <-1>) <2> Q1}-(b0-b1z <-1> -bn-1z <-n> + bnz <-(n + 1)>) {Q1
+ (1-z <-1>) <2> Q2 / A1} ... (Equation 6)
[0038]
11-04-2019
10
When the influence of the tap coefficient at direct current is determined, YFIR (z) | z = 1 = (a0 +
a1 + a2... + An1) X? (b0?b0?bn?1 + bn) Q1 (Equation 7)
It can be seen that the quantization noise of the first stage appears in proportion to the tap
coefficients of the elements constituting the second stage postfilter.
Assuming that the tap coefficients of the second stage post filter are b0 = 1 + ?b0, b1 = 1 +
?b1, bn?1 = 1 + ?bn?1, bn = 1 + ?bn for simplicity, YFIR (z) | z = 1 = ( a0 + a1 + a2 ... + an1)
X-(? b0-? b1-? bn-1 +? bn) Q1 (equation 8).
Therefore, at the output, the first stage quantization noise Q1 appears at the output in proportion
to the product sum of the errors ?bi of the taps.
[0039]
As described above, although it has become possible to reduce out-of-band noise by using a
cascaded ?? modulator using an analog FIR filter, noise due to a mismatch of elements
constituting the post filter is in-band. There was a problem of increasing the noise.
[0040]
The present invention comprises a first circuit receiving a first input signal, a second circuit
receiving a second input signal, a third circuit receiving an output signal from the second circuit,
and the third circuit. A fourth circuit receiving an output signal from the first circuit, and an
adder circuit combining the output signal of the first circuit and the output signal of the fourth
circuit and outputting the combined signal; Is formed by combining a digital analog conversion
circuit and an analog FIR filter, and the transfer coefficient of one of the second circuit and the
third circuit is (1-z <-1>), and the second The other transfer coefficient of the circuit and the third
circuit is (1-z <-n>), and the transfer coefficient of the fourth circuit is HFIR (z) = 1 + z <-1> + z <2>. In the case of <-(n-1)>, one of the second circuit and the third circuit having the transfer
coefficient (1-z <-1>) is an analog It consists of the road, and the other one of the second circuit
and the third circuit having a transfer coefficient (1-z <-n>) is composed of a digital circuit.
[0041]
According to the present invention, even when there is a variation in elements constituting a
digital-to-analog converter that converts digital signals into analog signals, high-quality analog
11-04-2019
11
signals can be generated, and they have high resolution. A small digital-to-analog conversion
device can be realized.
[0042]
In the cascaded ?? modulator according to the present invention, the postfilter placed after the
second-stage modulator in the case where the analog FIR filter is placed is characterized as
follows.
[0043]
Assuming that the order of the internal modulator of the cascade ?? modulator is 1 and H3 =
NTF1 = (1-z <-1>), then H3HFIR = (1-z <-1>) (1 + z <-1> + z <?2>... + Z <? (n?1)>) = (1?z
<?n>) ииииииииии (Equation 9) The order of the internal modulator is equated to H3 = NTF1 = If (1-z <1>) <2> = (1-z <-1>) <> (1-z <-1>), then H3HFIR = (1-z <-1>) (1 + z) <?1> + z <?2>... + Z <?
(n?1)>) (1-z <?1>) = (1-z <?n>) и (1?z <?1>) ... (Equation 10)
[0044]
In both cases of Eq. 9 and Eq. 10, since H3HFIR includes (1-z <-n>), this (1-z <-n>) term is
separated from the post-filter and pre-digitalized. The first feature is to do.
[0045]
On the other hand, it is a second feature that terms other than (1-z <-n>) are once converted into
a thermometer code by a formatter and then subjected to calculation processing by a post filter.
[0046]
Example 1 FIG. 6 shows a first example in which an analog FIR filter and a post filter are
provided behind the cascaded ?? modulator of the digital-to-analog converter according to the
present invention.
In this embodiment, it is assumed that the first stage of the cascaded ?? modulator is a 1-bit
internal modulator and the second stage is an n-bit internal modulator.
[0047]
11-04-2019
12
Here, the digital input signal (510) is input to the first stage internal modulator (201) of the
cascade ?? modulator, and the second stage internal modulator (202) is the first stage internal
modulator 201) is cascade-connected, and an output signal (520) from the first internal
modulator (201) is input to the analog FIR filter (301).
The output signal (530) from the second stage internal modulator (202) performs calculation of
(1-z <-n>) in the digital signal processing block (601).
The output (631) from the digital calculation block (601) is converted and output from the
binary code to the thermometer code by the formatter circuit (602).
The signal (632) converted to the thermometer code is input to the post filter circuit (603).
An output signal (521) from the analog FIR filter (301) and an output signal (633) from the post
filter circuit (603) are added in an analog form by a summing block (540) and output.
[0048]
FIG. 7a shows a first embodiment of the digital-to-analog converter of the present invention.
In the analog FIR filter of this embodiment, a delay element (701) composed of DFF performing
one clock delay, a drive buffer (702) connected to the output, and one end are connected to the
drive buffer, and one end is an analog The resistance element (703) connected to the output
terminal is used as a single-stage unit so that the voltages are weighted and added in an accurate
manner, and the units are connected in a plurality of stages.
[0049]
As shown in Equation 10, when the order of the internal modulator is quadratic, the transfer
function of the second stage is (1?z <?n>) и (1?z <?1>).
11-04-2019
13
(1Since -z <-n>) is processed digitally, the post filter needs to calculate (1-z <-1>) in an analog
manner.
FIG. 7b shows an embodiment of the 1-bit unit of the post filter, which is placed after the cascade
? ? ? ? ? ? modulator of the present invention.
[0050]
Here, the input signal Y2-m (632) indicates a signal of one bit of the digital signal converted to
the thermometer code by the formatter. The post filter unit to which the input signal Y2-m (632)
is input is a switch controlled by a delay element (711) configured by DFF that executes one
clock delay and a signal 00 obtained by dividing one clock Resistor element (713) connected to
the input via (715a) and one end connected to the drive buffer, and one end connected to the
output terminal such that the voltage is weighted and added in an analog manner. And a drive
inverter (714) connected to the output via a switch (715a) controlled by a signal 00 similarly
dividing one clock, and one end is connected to the drive inverter and one end is an analog
voltage A resistive element (715) connected to the output terminal so as to perform weighted
addition, an inverter (714) connected to the input and output of a delay element (711) composed
of DFF that performs one clock delay, and the inverter Signal divided by 1 clock ? And a switch
(715b) that is controlled by.
[0051]
Here, the connection between the input / output of the delay element (711), the drive buffer
(712) and the drive inverter (714) is connected by switching between the input / output with the
switches (715a) and (715b). The switches (715a) and (715b) are controlled by the signals .PHI.0
and .PHI.1 obtained by dividing one clock, so that they constitute a swapping circuit in which the
connection relationship of the resistance elements is swapped for each clock. By this swapping
circuit, the transfer function Y2 (z) with respect to the input digital signal sequence y21 (n), y21
(n + 1), y21 (n + 2),..., Y21 (n + k) (k: integer) is Y2, o (z) = b0 (1-Z <-1>), Y2, e (z) = b1 (1-Z <-1>)
However, Y2, o (z) represents k = odd, Y2 and e (z) indicate k = even. Further, it is assumed that a
variation error of b0 and b1 resistance elements (713) and (715). ????
11-04-2019
14
[0052]
Therefore, 1-Z <-1> is multiplied to variation errors b0 and b1. Therefore, if calculated as z = 1,
then Y2 (z) | z = 1 = 0. It shows that the influence of the mismatch disappears, and the first-order
mismatch shaping is applied to the variation.
[0053]
The simulation result of the output spectrum of the entire ?? modulator when the elements
constituting the post filter have a 1% mismatch is shown in FIG.
In the conventional method (conventional) which does not use a changeover switch, it can be
confirmed that the noise in the low band is significantly increased. On the other hand, when the
proposed method is used (Proposed), it can be seen that the low band noise is reduced by 6 dB
when the frequency is 1?2.
[0054]
As described above, it is understood that, when the present method is used, high SNR can be
realized even when element values such as resistors constituting the digital-to-analog conversion
device have variations, and a high-resolution digital-to-analog conversion device can be
configured. In LSI, element value variation is generally about 0.1%. Even in such a case, it is
possible to configure a high accuracy and high resolution digital-to-analog converter by using
this method.
[0055]
Second Embodiment FIG. 9 shows a second embodiment of the digital-to-analog converter
according to the present invention. In this embodiment, it is assumed that the first stage of the
cascade ?? modulator is a 1-bit internal modulator, and the second stage is an n-bit internal
modulator. As shown in Formula 9, when the order of the internal modulator is one, the transfer
function of the second stage is (1?z <?n>) и 1. (1Since -z <-n> is processed digitally, it is
necessary to calculate 1 in an analog manner. Here, the input signal Y2-m (632) converted into
the thermometer code by the formatter is once input to the selection circuit (910), and is
connected corresponding to each bit of the output from the selection device. A buffer (901) and
11-04-2019
15
one end are connected to a drive buffer, and a resistive element (902) connected to an output
terminal is formed by a unit group so that one end adds a voltage in an analog manner. In order
to remove the variation of the drive buffer (901) and the resistance element (902) by the
mismatch shaping method, the selection circuit (910) uses the output (921) of the selection
circuit (910) as a delay element and an adder. The frequency of use of the unit constituted by the
drive buffer (901) and the resistance element (902) is calculated by the constituted integration
circuit (911) and integration circuit (912), and selected in ascending order of usage frequency. It
is characterized by operating as follows.
[0056]
According to this embodiment, when performing mismatch shaping, the selection circuit is
switched instead of simply using the DEM method using a random signal, so a random signal that
causes a problem when using the DEM method is caused It is not necessary to introduce white
noise and to introduce a circuit for switching the selection circuit.
[0057]
Third Embodiment FIG. 10 shows a third embodiment of the digital-to-analog converter
according to the present invention.
In this embodiment, it is assumed that the first stage of the cascade ?? modulator is a 1-bit
internal modulator, and the second stage is an n-bit internal modulator. As shown in Equation 10,
when the order of the internal modulator is quadratic, the transfer function of the second stage is
(1?z <?n>) и (1?z <?1>). (1Since -z <-n> is processed digitally, it is necessary to calculate (1-z
<-1>) in an analog manner. Here, the input signal Y2-m (632) converted to the thermometer code
by the formatter is once input to the selection circuit (1010), and corresponds to each bit of the
output (1020) from the selection device. It is input to the post filter unit (603).
[0058]
The post filter unit (603) is connected to the input via a delay element (711) composed of a DFF
that performs one clock delay and a switch (715a) controlled by a signal 00 obtained by dividing
one clock. Also, a signal obtained by dividing one clock in the same manner as a resistance
element (713) connected to the drive buffer (712) and one end to the drive buffer and one end
connected to the output terminal so as to perform weighted addition of voltages in an analog
manner. The drive inverter (714) and one end connected to the output via the switch (715a)
controlled by ? 0 are connected to the drive inverter, and one end is connected to the output
11-04-2019
16
terminal so that the voltage is weighted and added in an analog manner. An inverter (714)
connected to the input and output of the delay element (711) configured of the resistive element
(715) and the DFF performing one clock delay, and the output of the inverter divided by one
clock Controlled by the signal ?? 1 That is composed of a switch (715b).
[0059]
Here, the connection between the input / output of the delay element (711), the drive buffer
(712) and the drive inverter (714) is connected by switching between the input / output with the
switches (715a) and (715b).
Since the switches (715a) and (715b) are controlled by the signals 00 and 11 obtained by
dividing one clock, a swapping circuit in which the resistance element is swapped for each clock
is configured.
[0060]
The selection circuit (1010) comprises an output (1021) of the selection circuit (1010) with a
delay element and an adder in order to further eliminate the variation between the units (603) of
the post filter by the mismatch shaping method. The integration circuit (1011) is characterized in
that it operates to calculate the frequency of use of the unit (603) of the post filter and select the
frequency of use in ascending order. As in the second embodiment, it is possible to increase the
order of mismatch shaping by repeatedly using the integrating circuit (1011) for controlling the
selection circuit (1010) used in the mismatch shaping method.
[0061]
According to this embodiment, high-order mismatch shaping can be easily realized by
superimposing mismatch shaping by the mismatch shaper and mismatch shaping by the
swapping circuit. The high-order mismatch shaping function, which conventionally had a large
hardware size, can be realized with a small amount of hardware. For example, second-order
mismatch shaping can be realized by using DWA (Data Weighted Averaging) for the mismatch
shaper.
11-04-2019
17
[0062]
In the first to third embodiments, an example is shown in which the first stage of the cascaded
?? modulator is a 1-bit internal modulator, and the second stage is an n-bit internal modulator.
The effect of the present embodiment can be similarly realized using the configuration of any
internal modulator composed of cascaded ?? modulators.
[0063]
Fourth Embodiment A fourth embodiment of a digital-to-analog converter according to the
present invention is shown in FIG. 11a.
In this embodiment, it is assumed that the ?? modulator has an output of n bits. In this
embodiment, the signal Y2-m obtained by converting the n-bit output of the ?? modulator
(1101) into a thermometer code by the formatter (1102) is mismatch shaped by the post filter
(1103), and the output is An analog addition is performed via the drive buffer circuit (1104) and
the resistance element (1105).
[0064]
FIG. 11 b shows an embodiment of the post filter (1103). In order to remove the variations of the
drive buffer circuit (1104) and the resistance element (1105) by the mismatch shaping method,
the selection circuit (1110) is configured of the output of the selection circuit (1110) by a delay
element and an adder. The frequency of use of the output signal is calculated by the integration
circuit (1111) and the integration circuit (1112), and the operation is performed so as to select in
order of the frequency of use. Here, the integration circuit performs the operation with the input
signal as an m-bit vector signal.
[0065]
In this embodiment, in order to remove the digital signal modulated by the ?? modulator by the
mismatch shaping method by the post filter using the integrating circuit when driving the
plurality of speakers by the plurality of driving circuits, There is no need to superimpose white
noise caused by a random signal or introduce a circuit that switches the selection circuit at high
speed, which is a problem when using a mismatch shaping method based on a certain DEM
method.
11-04-2019
18
[0066]
In this embodiment, an example in which a plurality of resistance elements are driven to add
voice in an analog manner is shown, but the present invention can be applied to all methods in
which a plurality of driving devices add in an analog manner.
[0067]
In this embodiment, the integration circuit (1110) for controlling the selection circuit (1110)
used in the mismatch shaping method is repeatedly used twice, but the mismatch shaping effect
is repeatedly used the integration circuit (1110) one or more times. It can be obtained by
[0068]
FIG. 12 shows another embodiment of the post filter (1103).
In order to eliminate variations between the speaker drive devices by the mismatch shaping
method, the selection circuit (1110) comprises an integration circuit (1111) comprising an
output of the selection circuit (1110) and a delay element and an adder and an integration
circuit. The frequency of use of the output signal is calculated by (1112), and it operates so as to
select in ascending order of frequency of use. Furthermore, the control circuit (1201) provided
between the input / output of the integration circuit and the selection circuit The output signal
selected by the selection circuit is limited according to the magnitude of the amplitude.
The control circuit (1201) operates such that the limited output drive circuit is selected when the
amplitude of the input digital signal is small, and controls such that all the output drive circuits
are selected when the amplitude is large. Do.
As a result, since only one output drive circuit is selected when the signal of small amplitude is
used, it is possible to suppress the influence of variations between the drivers at the time of small
amplitude, and at the time of small amplitude. Since sound is emitted only from a specific drive
device, localization of the sound image is improved. Further, by dynamically controlling the
number of drive of the drive device according to the amplitude of the input signal, it becomes
possible to optimize the power consumption consumed by the drive device. As described in the
first to fourth embodiments, the feature of the present invention is that noise shaping is
11-04-2019
19
performed on the digital input signal by the ?? modulator, then division is performed by the
formatter and mismatch shaping is performed by the post filter. The driving circuit is driven to
perform addition in an analog manner.
[0069]
As a result, even if the power for driving the plurality of driving devices is small, it is possible to
obtain a large output by adding in an analog manner.
[0070]
As described in the first to fourth embodiments, the digital-to-analog converter for converting
digital signals into analog signals converts the digital audio signals into a plurality of digital
signals and analogizes the outputs of a plurality of driving devices It is possible to apply to all the
devices which add.
[0071]
Fifth Embodiment FIG. 13A shows a fifth embodiment in which the digital-to-analog conversion
devices shown in the first to fourth embodiments are configured to be added by a current.
In this embodiment, the drive buffer and the resistance element which are the components of the
previous embodiments are respectively digitalized with a current source (1300), a switch circuit
(1302) provided between the current source and the output, and the switch. A configuration in
which a buffer circuit (1301) controlled by signals is replaced is shown.
[0072]
Sixth Embodiment FIG. 13 b shows a sixth embodiment in which the digital-to-analog conversion
devices shown in the first to fourth embodiments are added by sound pressure in FIG. 13 a. Is
shown.
In this embodiment, the drive buffer and the resistance element, which are components of the
previous embodiments, are respectively controlled by a speaker device (1310) and a voice coil
(1312) for driving the speaker device and the voice coil by digital signals. The configuration is
11-04-2019
20
shown in which the drive circuit (1311) is replaced.
[0073]
Seventh Embodiment Furthermore, FIG. 13 c shows the seventh embodiment in the case where
the digital-analog conversion devices shown in the first to fourth embodiments are added by light
in FIG. 13 a. It shows. In this embodiment, the drive buffer and the resistance element which are
the components of the previous embodiments are respectively replaced with a light emitting
element (1320) and a drive circuit (1321) for controlling the light emitting element by digital
signals. There is. In the present embodiment, as the light emitting element, any device capable of
emitting light by electric power such as a lamp or an LED can be used.
[0074]
Eighth Embodiment FIG. 14A shows an eighth embodiment in which the digital-to-analog
conversion devices shown in the first to fourth embodiments in FIG. 13A are configured by
addition using piezoelectric elements (piezo elements). An example is shown. In this embodiment,
the drive buffer and the resistive element which are the components of the previous
embodiments are respectively replaced with a piezoelectric element (1400) and a buffer circuit
(1401) for controlling the piezoelectric element with digital signals. It shows. Since the
piezoelectric element can convert an electrical signal into a physical displacement force, by
arranging a plurality of piezoelectric elements on a plane as in Example 14b (1410), the physical
displacement amount is converted to a sound wave in space. The present invention can be used
for applications such as combining, driving by adding a common vibration plate, and laminating
(1420) and adding a plurality of piezoelectric elements as in Example 14c.
[0075]
Since each piezoelectric element is driven by a 1-bit signal, it is possible to improve the power
efficiency and to reduce the influence of non-linearity of the piezoelectric element.
[0076]
In the present embodiment, the piezoelectric element is shown as means for converting an
electrical signal into a physical displacement force. However, any element capable of converting
an electrical signal into a physical displacement force can be used.
11-04-2019
21
[0077]
When a plurality of piezoelectric elements are stacked (1420) and added as in Example 14c
above, the strength of the physical displacement generated by each piezoelectric element can be
measured using another piezoelectric element It is.
That is, since variations in the strength of physical displacement generated by a plurality of
piezoelectric elements can be measured, the physical force generated by the plurality of
piezoelectric elements can be obtained by adjusting the driving force of the piezoelectric
elements according to the measured variations. It is also possible to improve the accuracy of
combined physical displacement by adding various displacements.
[0078]
Ninth Embodiment FIG. 15A shows a ninth embodiment in which the digital-analog conversion
devices shown in the first to fourth embodiments are configured to add magnetic fields
generated by coils.
In this embodiment, the drive buffer and the resistance element, which are components of the
previous embodiments, are respectively replaced with a buffer circuit (1501) for controlling the
coil (1500) and the front coil with digital signals. Since coils can convert electric signals into
magnetic field force, a plurality of coils are stacked and arranged (1510) as in Example 15b to
add magnetic fields or as in Example 15c, It can also be used in applications where the magnetic
field is added by simultaneously winding 1520 coils.
[0079]
As described above, since the magnetic fields can be added, the present invention can also be
applied to a digital-to-analog converter that reproduces an audio signal using a speaker drive
device using a plurality of voice coils. .
[0080]
11-04-2019
22
As in the above embodiments, it is also possible to measure the strength of the magnetic field
generated by each coil using another coil.
That is, since variations in the strength of the magnetic field generated by the plurality of coils
can be measured, the driving force of the coils is adjusted according to the measured variations
to combine the magnetic fields of the plurality of coils. It is also possible to improve the accuracy.
[0081]
Tenth Embodiment FIG. 16A shows a tenth embodiment in which the digital-to-analog converter
shown in the first to fourth embodiments is applied to a speaker driving device using a plurality
of voice coils. In this embodiment, a configuration is shown in which the resistive element which
is a component of the previous embodiments is replaced with a voice coil (1600). Since the voice
coil can convert an electrical signal into sound pressure by means of a cone (1601) or a dome,
the sound pressure is added by overlapping and arranging a plurality of coils (1610) as in the
embodiment 16a. Becomes possible. Further, in the method using the voice coil, the portion
emitting the audio signal is one cone (1601) or dome, so that the localization of the sound image
is improved.
[0082]
Further, as shown in FIG. 16b, the present invention can also be used in applications where
sound pressure is added by bundling and winding a plurality of voice coils (1620). By bundling
and winding a plurality of voice coils, respective voice coil characteristics can be made uniform.
Thereby, the error of the characteristic between voice coils is reduced, and it becomes possible to
reproduce a high-quality sound signal.
[0083]
As in the above embodiments, it is also possible to measure the strength of the magnetic field
generated by each voice coil using another voice coil. That is, since the variation in the strength
of the magnetic field generated by the plurality of voice coils can be measured, the driving force
of the voice coil is adjusted according to the measured variation to combine the magnetic fields
11-04-2019
23
of the plurality of voice coils. The accuracy of the voice signal is improved, and the voice signal
can be reproduced with high sound quality.
[0084]
In the above fifth to tenth embodiments, a plurality of driving devices are added in an analog
manner and output using the formatter and the post filter with the n-bit output from the
cascaded ?? modulator. Since the n-bit signal is converted into a thermometer code of m = 2
<n> by the formatter, 2 <n> post-filters and drive circuits are required. Here, by setting m = 2 <n>
= 16 or less, it is possible to suppress an increase in the circuit scale of the mismatch shaping
circuit and the swapping circuit. Similarly, by setting m = 2 <n> = 16 or less, as shown in the
embodiment shown in FIG. 14c, variation in the difference in the characteristics due to the
difference in the stacking order of the elements when the piezoelectric elements are stacked. Can
be reduced. Further, as in the embodiment shown in FIG. 15b and FIG. 16a, it is possible to
suppress the variation in the difference in the characteristics caused by the difference in the
stacking order of the coils when the coils are stacked. Furthermore, even in the embodiment of
bundling the coils as shown in FIG. 15c and 16b, it becomes possible to suppress the difference
in the characteristics of each coil.
[0085]
Eleventh Embodiment An eleventh embodiment according to the present invention is shown in
FIG. 17 using a modulator having another configuration than the cascaded ?? modulator of the
digital-to-analog converter shown in the first to fourth embodiments. Show. In order to transmit
the input signal to the output (1711) of the first stage of the cascaded delta-sigma modulator
(1700), a configuration using coefficients B0 (1720) and B1 (1730), respectively, is also possible.
When such a connection is made, part of the input signal is output also from the output of the
second and subsequent ?? modulators, so the number of stages of the cascade ?? modulator
may be increased, and a plurality of speaker devices may be used. In applications where sound
pressure is added in space by using it, it becomes possible to improve output sound pressure as
the number of speaker devices increases.
[0086]
(Twelfth Embodiment) The sound pressure is added in space by using a plurality of driving
11-04-2019
24
devices as described in the sixth embodiment, the eighth embodiment and the tenth embodiment
in the digital-to-analog converter shown in the first to fourth embodiments In such an
application, FIG. 18 shows a twelfth embodiment in which a block (1802) for digitally delaying a
signal for driving a driving device (1801) is inserted. It is possible to change the directivity of the
acoustic signal radiated in space by controlling the phase shift of the signal to each drive by thus
digitally delaying from the ?? modulator and the formatter. Become.
[0087]
For example, assuming that the distance between the speakers is d, the wavelength of the signal
is ?s, and the declination is ? when the front of the speaker is 0 radian, the phase of SP2 is
delayed by (2?d sin ?) / ?s with respect to SP3. By setting the phase of SP1 to (4?d sin ?) /
?s, it becomes possible to give the directivity characteristic on the SP1 side by only ?.
[0088]
In order to control the phase of a plurality of speakers in this manner, conventionally, an analog
phase shifter having a complicated structure is required, but since the input / output signal is a
digital signal, a digital delay (such as DFF) It is possible to easily control the exact phase shift
using
[0089]
(Example 13) Sound pressure is added in space using a plurality of driving devices as described
in Example 6, Example 8, and Example 10 for the digital-to-analog conversion devices shown in
the first to fourth examples. In such an application, FIG. 19a shows a thirteenth embodiment for
feeding back ambient noise as an input to a digital to analog converter.
The feedback control circuit (1900) is necessary to generate a signal whose phase is 180 degrees
out of phase with noise that cancels out ambient noise based on ambient noise information from
the microphone (1901) to which ambient sound is input. Calculate sound pressure and phase.
According to the present invention, it is possible to directly control the speaker with a digital
circuit, so it is possible to configure a precise noise reduction device. Further, as shown in FIG.
19b, since it is possible to generally detect the direction of the noise source by using a plurality
of microphones, the phase to each speaker driving device is controlled using the technique of the
twelfth embodiment. This allows the noise reduction speaker to have directivity. That is, not only
11-04-2019
25
the front direction of the noise reduction speaker but also noise in other directions can be
reduced.
[0090]
When noise reduction is performed in a car, a plurality of external noise sources may vary and
noise sources may vary, but a plurality of noise reduction speakers can be easily arranged by
using this embodiment. In addition, noise in directions other than the front can be reduced by
using a plurality of speakers, so that the interior of the vehicle can be muffled efficiently.
Furthermore, using a piezoelectric speaker makes it possible to realize a thin noise suppressor, so
noise can be silenced without reducing the space in the vehicle.
[0091]
(Example 14) The sound pressure is added in space using a plurality of driving devices as in
Example 6, Example 8, and Example 10 for the digital-to-analog converter shown in the first to
fourth examples. In such an application, FIG. 20 shows a fourteenth embodiment in which the
drive device (2000) is configured of a switching amplifier. As the switching amplifier, an analog
class D amplifier, a digital class D amplifier, an analog ?? modulator, a digital ?? modulator,
or the like can be used. Since the input digital signal is converted to a switching signal (binary
signal or ternary signal) by the switching amplifier, it is possible to improve the efficiency and
the output power.
[0092]
(Fifteenth Embodiment) The sound pressure is added in space using a plurality of driving devices
as described in the sixth embodiment, the eighth embodiment, and the tenth embodiment in the
digital-to-analog converter shown in the first to fourth embodiments. A fifteenth example of a
loudspeaker (2100) in such an application and a method of arranging electrical elements capable
of converting an electrical signal into a physical displacement force is shown in FIG.
[0093]
FIG. 21a shows an embodiment in the case of a grid arrangement.
11-04-2019
26
With such an arrangement, when covering a rectangular, square, etc. casing, subunits can be
arranged efficiently, and the horizontal and vertical directions are similar, and equivalent phase
characteristics can be realized. It can. In addition, when a rectangular or square speaker is used,
it is possible to arrange the rectangular surface most closely, and to maximize the radiation
sound pressure per unit area. Those arranged in this way are visually beautiful.
[0094]
FIG. 21b shows an embodiment in which the arrangement position is shifted by half for each row.
By arranging in a staggered manner in this manner, the surface density can be improved as
compared to the lattice arrangement. In particular, when a large number of speakers are
arranged, it is possible to increase the sound pressure per area. Furthermore, if hexagonal shapes
are used for the shape of the speaker, they can be arranged in a staggered arrangement and
without gaps. In this case, since it becomes possible to arrange without gaps, a high sound
pressure level can be realized. When the mismatch shaping technology is used, the distance
between the speakers is short, so that the mismatch shaping effect can be effectively realized.
[0095]
FIG. 21b shows an embodiment in which the loudspeakers are arranged concentrically. Thus,
since the distance of the speakers arranged on each concentric circle is equal from the central
axis of the whole speaker, the phase characteristics from the same concentric circle to the central
axis become equal, and the addition of the acoustic signals in the front is ideal. To be done. For
this reason, it becomes possible to improve acoustic characteristics.
[0096]
(Sixteenth Embodiment) The sound pressure is added in space using a plurality of driving devices
as described in the sixth embodiment, the eighth embodiment, and the tenth embodiment in the
digital-to-analog converter shown in the first to fourth embodiments. In such an application, FIG.
22 shows a sixteenth embodiment in which the digital-to-analog converter is configured in
stereo. Here, channel (2201) is a digital-to-analog converter for R signals, and channel (2202) is
a digital-to-analog converter for L signals. As described above, by providing the digital-to-analog
converter according to the present invention as a plurality of channels in parallel as well as
11-04-2019
27
reproducing the stereo sound of the digital signal, it becomes possible to reproduce an arbitrary
sound field generated by digital processing.
[0097]
(Seventeenth Embodiment) As shown in the sixteenth embodiment, sound pressure is added in
space using a plurality of digital-to-analog converters using a plurality of driving devices as in the
sixth, eighth, and tenth embodiments. A seventeenth embodiment relating to a speaker (2100) in
such an application and a method of arranging electric elements capable of converting an electric
signal to a physical displacement force is shown in FIG.
[0098]
FIG. 23a shows how to arrange speakers driving stereo L and R signals.
By arranging L and R symmetrically as described above, it is possible to enhance the stereo
effect. In the figure, L represents the left channel and R represents the right channel. FIG. 23b
shows a method of arranging speakers driving C signal in addition to stereo L and R signals. C in
the figure indicates a center channel. In the present invention, the assignment of channels to a
plurality of speakers can be easily changed dynamically, but it is possible to dynamically change
the assignment of channels by the music source to be reproduced and the sound field effect to be
realized. It is possible to perform stereo effects and sound field effects more effectively. FIG. 23c
shows a speaker arrangement method in the case of dynamically controlling assignment of a
plurality of speakers to channels. It shows how to arrange speakers driving C signal in addition to
stereo L and R signals. In the figure, L / C shows a speaker capable of driving both L signal and C
signal, and in the figure, R / C shows a speaker capable of driving both R signal and C signal.
[0099]
(Example 18) The sound pressure is added in space using a plurality of driving devices as in
Example 6, Example 8, and Example 10 with the digital-to-analog converter shown in the first to
fourth examples. In such an application, FIG. 24 shows an eighteenth embodiment in which a
digital filter processed digital signal is made into a stereo configuration of digital analog
conversion devices of a plurality of channels. Here, a plurality of digital signals obtained by
dividing the frequency band by the digital filter signal processing block (2401) are represented
by a plurality of channels (2402) and (2204) as digital-to-analog converters. For example, the
11-04-2019
28
digital-to-analog converter according to the present invention is provided in parallel as a plurality
of channels in order to divide the digital signal for high band and the digital signal for low band
in the digital filter signal processing block and reproduce each signal. Thus, it is possible to
reproduce with a speaker device that is optimal for the frequency band generated by digital
processing.
[0100]
(Example 19) The sound pressure is added in space using a plurality of driving devices as
described in Example 6, Example 8, and Example 10 for the digital-to-analog converter shown in
the first to fourth examples. In such an application, as shown in FIG. 25, the signal for driving the
driving device is once transmitted to the transmission line by the digital signal transmitter (2501)
and then received by the digital signal receiving device (2502) to drive the speaker by the driving
device. Shows an embodiment of the invention.
[0101]
As described above, by transmitting digital signals from the ?? modulator and the formatter by
the digital signal transceiver, it is possible to transmit a signal for driving the distributed
speakers as a digital transmission signal.
Since the digital signal is oversampled by the ?? modulator, even if there is an error in the
transmission line, the influence can be reduced. As the transmission line, it is possible to use any
transmission line that transmits digitally, such as a digital wired transmission line, a wireless
transmission line, and an optical transmission line.
[0102]
Also, when applied to a noise reduction device, it is necessary to distribute a plurality of noise
reduction speakers, but by using this embodiment, it is possible to transmit drive information to
easily separated sub speakers using a digital transmission line. Is possible.
[0103]
(Twenty-Sixth Embodiment) The sound pressure is added in space using a plurality of driving
devices as described in the sixth embodiment, the eighth embodiment, and the tenth embodiment
in the digital-to-analog converter shown in the first to fourth embodiments. In such an
application, FIG. 26 shows a twentieth embodiment in which an ultra-low frequency signal is
11-04-2019
29
superimposed on a signal for driving a drive device.
[0104]
Generally, the audio frequency is 20 to 20 KHz, and the sound of 20 Hz or less, which is the
lower limit frequency, is called an ultra-low frequency.
Although the sound in this band can not usually be recognized by human hearing unless it has a
considerable sound pressure, studies are being conducted regarding it as being related to health
and mental stress.
[0105]
In order to generate ultra-low frequency using conventional analog speakers, it is necessary to
drive the speakers with very slow signals, and there are many power consumption problems with
analog speakers with poor power efficiency.
If the configuration of the digital speaker according to the present invention is used to generate
an ultra-low frequency, it is possible to drive the electroacoustic transducer with a 1-bit signal,
thereby reducing the power efficiency and the influence of nonlinearity of the electroacoustic
transducer. It is possible to efficiently generate an ultra low frequency signal.
[0106]
In general, since the very low frequency signal is not included in the signal source (broadcast
signal or recording medium), when the very low frequency signal is generated, it needs to be
generated by the very low frequency generator (2600). The ultra low frequency generator should
use a digital circuit (2600) to generate an arbitrary frequency pattern, for example a pseudo
random signal of the digital circuit to generate an ultra low frequency signal with 1 / f
fluctuation It can be easily generated by using. Since the generated very low frequency signal can
be easily added digitally to the digital voice signal, it becomes possible to easily superimpose the
very low frequency signal.
11-04-2019
30
[0107]
(Twenty-first Embodiment) Physical displacements are synthesized using a plurality of
piezoelectric elements as in the eighth embodiment, using the digital-to-analog conversion
devices shown in the first to fourth examples. In such an application, FIG. 27 shows a twenty-first
embodiment in which a plurality of piezoelectric elements are used to drive a reflecting mirror.
27b, a device for driving a plurality of driving devices (2701) using a plurality of piezoelectric
elements (2702) to drive the stacked piezoelectric elements (2700) as shown in FIG. 27a. (2711)
A plurality of pieces are arranged on the upper substrate (2712) so that the support part (2714)
is a fixed shaft. The base of the upper places a reflector (2713). As shown in FIG. 27c, by driving
the piezoelectric element (2700), it is possible to change the reflection angle of the reflecting
mirror (2713) by deforming the upper base (2712) centering on the support part (2714). A
device in which such a piezoelectric element and a reflecting mirror are combined is applicable to
a compact projector device, but since the reflection angle can be controlled by a digital signal by
driving with a digital analog conversion device and a plurality of piezoelectric elements, the
device is compact Suitable for projector devices.
[0108]
(Example 22) A physical displacement is synthesized using a plurality of piezoelectric elements
as in the eighth example using the digital-to-analog conversion devices shown in the first to
fourth examples. In such an application, FIG. 28 shows another twenty-second embodiment in
which a plurality of piezoelectric elements are used to drive a reflecting mirror. As shown in FIG.
28a, a plurality of piezoelectric elements (2800) arranged in parallel on the substrate are used.
As shown in FIG. 28b, this piezoelectric element (2800) is placed around the reflecting mirror
(2701) whose center is fixed and driven along the X-Y axis to drive the reflecting mirror (2801)
with the center of the support as the base point. Can change the reflection angle of A device in
which such a piezoelectric element and a reflecting mirror are combined is applicable to a
compact projector device, but the reflection angle of a digital signal can be obtained by driving
with a digital analog conversion device and a plurality of piezoelectric elements arranged in
parallel. It is suitable for thin and small projector devices because it can control
[0109]
(Twenty-Third Embodiment) FIG. 29 shows a twenty-third embodiment where a bandpass ??
11-04-2019
31
modulator is used as the cascade ? ? ? modulator of the digital-to-analog converter shown in
the first to fourth embodiments. ing. In general, a bandpass ? ? ? modulator can be realized by
performing Z ? ?Z <2> conversion. This conversion converts the integrator into a resonator. In
this embodiment, the output of the internal ? ? ? modulator of the second stage is connected
to a two-clock delay that realizes Z <?2>, and a changeover switch is connected to the input and
output of the delay. This changeover switch switches the two inputs as shown in the figure in
accordance with a signal of 1/2 frequency of the clock. With such a configuration, even when
there is a mismatch in the elements constituting the DAC 21 and the DAC 22, it is possible to
reduce noise at a frequency of 1?4 of the clock frequency.
[0110]
As shown here, by performing frequency conversion, it is possible to realize arbitrary noise
shaping characteristics including band pass characteristics.
[0111]
Twenty-Fourth Embodiment FIG. 30A shows a twenty-fourth embodiment of the present
invention.
In this embodiment, it is assumed that the ?? modulator has an output of n bits. The n-bit
output of the ? ? ? ? ? ? modulator (2401) is converted into m sets of p-bit codes by the
formatter (2402), mismatch shaping and frequency selection are performed by the post filter
(2403), The output is converted to an analog signal by an internal digital-to-analog converter
(2404), and is added in an analog manner by an adder (2405). With this configuration, it is
possible to obtain a high-precision analog signal even using a multi-level internal digital-toanalog converter.
[0112]
FIG. 30 b shows an embodiment of the post filter. In order to reduce the influence of the
mismatch inside the internal digital-to-analog converter, the selection circuit (2410) selects the
output of the selection circuit (2410) according to the value of the output signal of the filter
circuit (2411). It is characterized by operating. Here, in the filter, the filter operation is
performed for each output level of the internal digital-to-analog converter. For example, using a
filter in which integrators or integrators are connected in multiple stages, selection is performed
11-04-2019
32
in ascending order of filter output, and selection is performed so that an output corresponding to
the input signal can be obtained by t
Документ
Категория
Без категории
Просмотров
0
Размер файла
54 Кб
Теги
description, jp2011211725
1/--страниц
Пожаловаться на содержимое документа