JP2019161375A - Modulator, parametric speaker, simulator of ultrasonic emitter, and parametric speaker development system - Google Patents

Abstract

To provide a parametric speaker having a high sound quality and high efficiency and a parametric speaker development system capable of free design development.SOLUTION: The modulator for modulating a carrier signal in an ultrasonic band with an audible signal from an external sound source to generate a modulated signal in the ultrasonic band, includes: an ultrasonic transmitter for generating a carrier signal; and a processor. The processor is configured to prepare digital sound source data, S1; convert the digital sound source data to text data, S2; generate an ultrasonic signal by performing a numerical calculation on the carrier signal generated based on the digital sound source data and the ultrasonic transmitter included in the text data, S3; and reversely convert a result of the numerical calculation into digital sound source data, S4.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a modulator, a parametric speaker, an ultrasonic emitter simulator, and a parametric speaker development system.

  A parametric speaker is a highly directional speaker having a modulator, an amplifier, and an ultrasonic emitter as basic components. The modulator is a device for modulating carrier ultrasonic waves with an audible band signal supplied from a sound source, that is, for up-converting an audible band signal into an ultrasonic band signal (hereinafter referred to as “ultrasonic signal”). . The amplifier is a power amplification circuit for transmitting an ultrasonic signal generated by the modulator from a subsequent ultrasonic emitter at a high sound pressure. The ultrasonic emitter is a device for emitting ultrasonic waves, and is generally composed of a plurality of ultrasonic elements having a resonance frequency near the frequency of a carrier signal (hereinafter referred to as “carrier frequency”).

  The ultrasonic signal emitted from the ultrasonic emitter is naturally demodulated into an audible sound range in the process of propagating through the air. This is because an equivalent circuit of a square detection circuit (a circuit for converting a modulated wave into an audible sound range) is configured by a nonlinear effect generated when high-sonic pressure ultrasonic waves propagate in the air. It is described as self-demodulation due to non-linear action.

  Patent Document 1 describes an invention made by the present inventors prior to the present invention. In paragraph 33 of the same document, this parametric speaker is composed of a “main circuit” composed of “modulation means”, “amplifier”, and “ultrasonic radiation device” and “sub-circuit” for constituting audible signal correction means. It is described that it is comprised with the electronic circuit which consists of these. Patent Document 2 discloses a method of generating a signal equivalent to an amplitude-modulated ultrasonic signal by a computer.

JP2015-084499A JP 2003-153370 A

  There are various methods such as an AM method, an SSB method, an FM method, and a PWM method as the modulation method of the parametric speaker, and the circuit configurations thereof are also different. However, regardless of which modulation method is employed, a modulator composed of hardware by an electronic circuit as described in Patent Document 1 is generally used.

  On the other hand, as in the invention described in Patent Document 2, attempts have been made to replace a demodulator composed of an electronic circuit with a computer. However, among the newly added samples that are not in the audible sound signal by oversampling the audible sound at a sampling frequency that is an integer multiple of 2 or more of the upper limit frequency of the audible sound in order to generate an ultrasonic wave that becomes the carrier signal By replacing at least a part with a zero value, aliasing distortion is generated, and an electroacoustic transducer is driven by a signal including the aliasing distortion component to generate ultrasonic waves.

  However, in this method, since the carrier signal is continuously output even when the input sound source signal is silent, it is impossible to avoid the problem of ultrasonic exposure that is inferior in power efficiency and has a concern about influence on the human body. Furthermore, since the sampling frequency of the carrier signal is defined as an integer multiple of 2 or more of the upper limit value of the audible sound frequency, there is a problem that the degree of freedom of design is greatly limited.

  On the other hand, development of parametric speakers is not limited to modulator design. The ultimate in improving the sound quality of the audible range that will ultimately be obtained as a result of self-demodulation due to non-linear propagation in air, as well as safety issues based on reduced power efficiency and ultrasonic exposure. It can be said that there is a purpose.

  For these reasons, the development of parametric speakers tends to be prolonged, and it has been difficult to reduce development costs.

  The present invention has been made in view of the above, and it is a main technical problem to provide a parametric speaker with high sound quality and high efficiency, and to provide a parametric speaker development system capable of free design development. To do.

The modulator according to the present invention comprises:
A modulator for modulating a carrier signal in an ultrasonic band with an audible signal from an external sound source to generate a modulated signal in the ultrasonic band,
The modulator includes an ultrasonic transmitter for generating a carrier signal and a processor,
By the processor,
Step S1 for preparing digital sound source data;
Step S2 for converting digital sound source data into text data;
Step S3 for generating an ultrasonic signal by performing a numerical operation on a digital sound source data included in the text data and a carrier signal generated based on the ultrasonic transmitter;
Step S4 for inversely converting the result of the numerical calculation into digital sound source data;
Is executed.

  In this specification, the “carrier signal” means a signal in an ultrasonic band generally carrying several tens of kHz or more that carries an input sound source signal. For example, in the case of the SSB modulation method, an upper sideband ( This is a method for generating a signal obtained by adding a USB) signal or a lower sideband (LSB) signal.

The digital sound source data may be high resolution sound source data.
A high resolution sound source means a sound source that exceeds the specifications of an audio CD, specifically, a sound source that is recorded / reproduced with a sampling frequency of 44.1 kHz or a discretization characteristic that exceeds 16 bits of quantization. Say. The high resolution sound source is hereinafter abbreviated as “high resolution sound source”. In the current high-resolution environment, since the sampling frequency corresponds to 192 kHz, an ultrasonic band signal up to approximately 96 kHz can be handled. This high-resolution recording / reproducing device is also provided as a general-purpose audio device that supports high-resolution sound source output, such as a portable audio player and a smart phone.

  The numerical calculation by the processor is at least one selected from level adjustment in the process from the input sound source signal to the output signal, adjustment of frequency characteristics and phase characteristics as filter functions, sound quality improvement for distortion reduction, and power saving. An operation processing step for providing a numerical operation function for realizing one function can be further included.

  The converter having the above configuration further has a function of accepting an audible signal file from an external sound source, a streaming function of sequentially receiving an audible signal from an external sound source in a buffer memory, and directly calculating and outputting it to the outside, a signal file of a high-resolution sound source At least one of the functions for generating

  It is possible to configure a parametric speaker including the modulator configured as described above, an amplifier that amplifies the modulated signal, and an ultrasonic emitter that emits ultrasonic waves based on the amplified modulated signal. .

  The simulator of an ultrasonic emitter according to the present invention is characterized in that an ultrasonic signal is handled as a high-resolution sound source unified by, for example, a sampling frequency of 96 kHz and a quantization of 16 bits, and modeled as an RLC series resonance circuit. .

  By including any of the above-described modulators and the above-described ultrasonic emitter simulator, a parametric speaker development system can be configured.

  In the above configuration, the modulator and the ultrasonic emitter as hardware and the subsequent demodulation process can be simulated by the calculation function of the software, so that the product development period can be shortened and the manufacturing cost can be reduced. In addition, since it is easy to adjust the circuit configuration and parameters, it is possible to develop and realize a system for improving the sound quality without providing a separate circuit for sound quality improvement such as distortion reduction and input / output linearity improvement. .

FIG. 1 is a conceptual diagram showing the signal flow of a parametric speaker in units of functional blocks for each process. FIG. 2 shows functional blocks of the converter shown in FIG. FIG. 3 is a diagram showing a processing flow until the ultrasonic signal x (t) is output from the input sound source signal s (t) shown in FIG. FIG. 4A shows a 2 kHz sine wave signal. 4B and 4C show output signal waveforms when the sine wave signal shown in FIG. 4A is the input sound source signal s (t) (time waveform of the ultrasonic signal x (t)). Is shown. FIG. 4B shows a time waveform, and FIG. 4C shows its frequency spectrum. FIG. 5A shows an audio signal of about 4 seconds. FIG. 5B shows an output signal waveform (time waveform of the ultrasonic signal x (t)) when the sound signal shown in FIG. 5A is used as the input sound source signal s (t). FIG. 6A shows the appearance of the ultrasonic emitter used in the experiment. FIG. 6B is a diagram showing frequency characteristics of sound pressure radiated from the ultrasonic emitter. FIG. 7 shows the frequency response characteristics of the SPL of audible sound measured at a sound pressure level (SPL) of the ultrasonic emitter of 130 dB, distances of 1 m, 4 m, 10 m, and 20 m. FIG. 8A shows a time waveform showing an example of an input sound source signal and a demodulated signal obtained by simulation, and FIG. 8B shows a frequency spectrum of FIG. 8A. It is a block diagram of the parametric speaker of a 3rd embodiment. A portion indicated by a broken line is a modulator. Batch processing is represented by a one-dot chain line, and real-time processing is represented by a solid line.

(First embodiment)-Basic system configuration and basic processing flow-
Hereinafter, parametric speakers according to embodiments will be described with reference to the drawings.
FIG. 1 is a conceptual diagram showing the signal flow of a parametric speaker in units of functional blocks for each process. In order to build a computer simulator, the processing process is divided into three parts: (I) modulation process, (II) ultrasonic emitter driving process, and (III) self-demodulation process, and each process is modeled and formulated. Think about what to do.

  The physical functional block is composed of the modulator 1, the amplifier 2, and the ultrasonic emitter 3, but the nonlinear propagation 4 is the fourth functional block in order to model and formulate the self-demodulation process. . Hereinafter, each functional block will be specifically described.

(I) Modulation process (up-conversion)
The modulator 1 has a basic function of modulating an ultrasonic carrier with an input sound source signal s (t) in an audible band, converting it to an ultrasonic signal x (t), and outputting it. As a modulation method of the modulator, amplitude modulation (AM) for adjusting the amplitude of the carrier signal to the waveform of the signal, frequency modulation (FM) for increasing or decreasing the carrier frequency in proportion to the signal size, A plurality of modulation schemes are known, such as phase modulation (PM) in which the phase of the carrier signal is shifted in proportion to the magnitude of. An amplitude modulation method is often used in a parametric speaker, and a double sideband (DSB) in which sidebands (sidebands) are generated on both sides of a carrier signal frequency component and a single sideband (SSB: Single). There are two methods of Side Band). The modulator that inputs the signal from the sound source side and outputs the signal to the amplifier is basically based on the above-mentioned modulation processing in the process in between, but as another function, the process from the input sound source signal to the output signal Is equipped with a numerical calculation function to achieve at least one of the functions selected from level adjustment, frequency characteristic and phase characteristic adjustment, sound quality improvement for distortion reduction, and power saving Also good.

  As described above, the modulator 1 configured as described above has been conventionally realized by hardware such as an electronic circuit (analog circuit or digital circuit). However, in the case of an analog circuit, it is difficult to optimize a filter. Fine adjustment to obtain the desired frequency characteristics is difficult. Furthermore, changes in parameters and circuit configuration may affect the entire system, so knowledge and experience were required for prototyping and manufacturing, such as ensuring consistency for each block. On the other hand, in the case of digital circuits, specialized knowledge and experience of digital circuits such as DSP (Digital Signal Processor) and FPGA (Field Programmable Gate Array) are required. In addition, although it is conceivable to use a circuit simulator (for example, LTspice), it cannot cope with a complicated simulation at a practical level and is not realistic.

  Therefore, in the present embodiment, a method for outputting a modulated signal by numerical calculation using a processor mounted on a computer or the like will be specifically described.

  FIG. 2 shows functional blocks of the converter shown in FIG. This converter shows an example in which a dynamic SSB modulation method obtained by improving the SSB modulation based on the Weaver method is adopted.

  “Dynamic SSB modulation” is a modulation method in which the amplitude of the carrier signal is changed in accordance with the envelope of the input sound source signal s (t), and the broken line portion 20 shown in FIG. 2 is a functional block that realizes this function. In the broken line part 20, after calculating the absolute value (ABS) and the square root (ROOT) of the input sound source signal s (t) that has passed through the band filter, the band is limited by an LPF having a cutoff frequency of several tens of Hz, and then the delay ( DELAY) to obtain an envelope signal e (t) by numerical calculation, multiply the carrier signal, add it to the SSB signal, and output an ultrasonic signal x (t).

  According to such dynamic modulation, the output of the carrier signal becomes zero when the input sound source is silent. For this reason, there exists an advantage which can reduce the power consumption and ultrasonic exposure at the time of the system drive of a parametric speaker.

The input sound source signal s (t) is passed through a bandpass filter (BPF) and then subcarriers in a state in which digital sound source data recorded in a general-purpose format represented by the wav format or mp3 is converted into a text file. after sinusoid of frequency _{f s = 4.1kHz sinω s t =} sin (2πf s t) and cosω _s t = cos a (2πf _s t) is multiplied, is band-limited by the LPF cutoff frequency 4.1KHz. A sine wave signal sin [(ω _c ± ω _s ) t], cos [(ω _c ± ω _s ) t] obtained by adding or subtracting the subcarrier angular frequency ω _s to the carrier angular frequency ω _c to the signal thus obtained. Multiply by. Note that sine wave of carrier frequency 25kHz as an example (sinω _c t) signal generated by the ultrasonic generator. Finally, an SSB signal is obtained by adding or subtracting the two signals. Here, when added, an upper sideband (USB) signal is obtained, and when subtracted, a lower sideband (LSB) signal is obtained. On the other hand, the amplitude of the carrier signal is determined according to the amplitude of the input sound source signal, and the SSB signal is added thereto. However, a known method can be used as a method for generating the SSB signal, and a method other than the Weaver method, for example, a simple filter method, a PSN (Phase Shift Network) method, a merry-go-round method, or a circuit method may be used. Is not limited.

  In the above signal flow, each filter used a Kaiser type FIR (finite impulse response). The FIR type filter is known to have a characteristic not found in an analog filter, in which only the frequency response can be changed with a constant delay amount depending on the frequency.

  The final output is an ultrasonic signal x (t) of digital sound source data in the same format as the input sound source signal (when the wav format is input, the output is usually also in the wav format).

  Here, it is possible to generate a high-resolution sound source file (for example, a wav file) by using a high-resolution sound source for the input sound source signal s (t) and performing inverse conversion to output it to the high-resolution recording / reproducing device.

  This can be realized because all multiplications and additions / subtractions performed in the analog circuit are performed by a processor mounted on a computer or the like.

  FIG. 3 is a diagram showing a processing flow until the ultrasonic signal x (t) is output from the input sound source signal s (t) shown in FIG. Digital sound source data (for example, wav format) is prepared as an input sound source signal (step S1), and converted to text data that can be easily numerically calculated (step S2).

  Next, in step S3, numerical data is read from the text file, modulation processing such as carrier frequency, FIR digital filter, amplitude modulation, input / output signal level adjustment, frequency characteristic adjustment with filter function, sound quality improvement, A processing function for realizing power saving or the like is calculated as necessary. In this way, with the function that realizes modulation processing such as DSB or SSB amplitude modulation system as the core, various processing functions for signal level adjustment, frequency characteristic adjustment by filter, sound quality improvement, power saving, etc. are analog circuits Regardless, it can be realized by programming. Needless to say, any modulation method such as frequency modulation and phase modulation can be used for the modulation as long as the signal can be expressed by a mathematical expression.

  Next, in step S4, the result obtained by the numerical calculation is inversely converted into digital sound source data, and an ultrasonic signal x (t) modulated by the sound source signal in the audible band is obtained as an output signal. The output signal is then sent to an audio interface or the like. If both the output signal and the audio interface are compatible with a high-resolution sound source, an ultrasonic signal modulated by the high-resolution sound source can be obtained.

  Most of the numerical operations performed in each of the above steps are addition / subtraction and multiplication (and simple multiplication of constants), and high-level operations are not necessary. In the experiment, programming was performed using the C language. However, programming may be performed using other languages such as Fortran, JAVA (registered trademark), and machine language.

  Conditions for embodying individual functions can be easily added or changed on the program, and are very simple. For example, the carrier signal frequency of the modulation system, the filter coefficient for frequency adjustment, the sampling frequency for generating the high-resolution sound source file, the condition of the number of bits, and the like can be freely set and changed to fine adjustment. In other words, it can be done by coding the program. With a hardware modulator, which is a conventional measure, processing functions created with analog circuits and digital circuits can be realized by desktop programming.

  The numerical calculation of FIG. 3 can be executed on a computer (such as a personal computer), but a dedicated numerical calculation device may be used. Furthermore, the software simulator in the present invention is not limited to these, and may be mounted on a device that can incorporate a numerical calculation function. For example, using a numerical calculation function installed in a mobile terminal such as a smart phone or a touch panel, or a device having an artificial intelligence (AI) function that applies advanced information processing (robot, AI speaker, smart speaker, etc.) Also good.

  A high-resolution sound source file generated as a result of numerical calculation is first saved in a storage device of a computer. The high-resolution sound source file is transferred to a high-resolution recording / playback device by an external recording medium or a wired or wireless connection method. And can be saved. On the other hand, when the processor for numerical computation can execute the playback software for the high-resolution sound source file, for example, via an audio interface for high-resolution, an amplifier (for example, a digital amplifier) and an ultrasonic wave without going through a high-resolution recording / playback device An ultrasonic signal can be supplied directly to the emitter with good alignment.

-Experimental example-
(Experiment 1)
First, the output signal waveform when a sine wave signal was input as an input sound source signal was examined.
FIG. 4A shows a 2 kHz sine wave signal. 4B and 4C show an output signal waveform (ultrasonic signal x (t)) when the sine wave signal shown in FIG. 4A is used as the input sound source signal s (t). Yes. FIG. 4B shows a time waveform, and FIG. 4C shows its frequency spectrum. Here, the magnitude of the input sound source signal was adjusted so that the amplitudes of the 25 kHz carrier signal and the 27 kHz USB signal were the same.

  In the observed output waveform, a balanced modulated signal of the same level of 25 kHz and 27 kHz was obtained, and it was confirmed that the desired numerical calculation was performed correctly.

(Experiment 2)
Next, the output signal waveform when actual sound data was input as an input sound source signal was examined.
FIG. 5A shows an audio signal of about 4 seconds. FIG. 5B shows an output signal waveform (time waveform of the ultrasonic signal x (t)) when the sound signal shown in FIG. 5A is used as the input sound source signal s (t).

  As the input sound source signal, a sound data file in wav format in which audio data read out from a passage of a famous novel is digitally recorded is used. It was confirmed that there is almost no carrier component in part of the modulated signal in the time domain where there is no audio signal. If the WSB file signal of the SSB signal recorded in the computer or audio device is supplied to the ultrasonic emitter through an amplifier connected to the high-resolution player, the original audio signal can be restored. At present, the uncompressed linear PCM format wav has been selected to have a sampling frequency of 96 kHz and quantization of 16 bits, so that the reproduced sound of the parametric speaker can be obtained with the same sound quality as the conventional method using a modulator based on an analog circuit. It was. According to this method, it is possible to produce an ultrasonic emitter having a resonance frequency of 25 kHz and a high-resolution sound source for driving an emitter having a resonance frequency of 40 kHz.

  However, after an optimal circuit configuration and parameters are obtained by numerical calculation, it can be configured by an electronic circuit that realizes an equivalent function. This is because hardware generally has a higher processing speed and can perform real-time processing.

(II) Driving Process In the driving process following the modulation process, the ultrasonic signal x (t) output through the modulator is amplified by the amplifier 2 shown in FIG. 1 and the ultrasonic signal is emitted from the ultrasonic emitter 3. Is done. Many of the power amplifiers of parametric speakers have flat frequency characteristics, so in the modeling, it is only necessary to multiply by a constant according to the amplification degree. On the other hand, the ultrasonic emitter is an electroacoustic transducer for emitting an ultrasonic signal into the air. For example, it is common to have a large number of piezoelectric ceramic elements arranged in a plane. Since the frequency characteristic of the ultrasonic element affects the frequency characteristic of the demodulated sound, the frequency characteristic of the ultrasonic emitter is examined as a preliminary experiment.

FIG. 6A shows the appearance of the ultrasonic emitter used in the experiment.
<Ultrasonic emitter>
Ultrasonic element made by Nippon Ceramic Co., Ltd. (T4010B4, number of elements 271)

により表される。ここで、Ａは任意定数、ω_ｄ＝ω_０・（１−１／４Ｑ_０ ^２）^１／２、ω_０＝２πｆ_０）は共振周波数、Ｑ_０は共振の先鋭度、ｕ（ｔ）は単位ステップ関数である。この式の導出方法は、先ず、超音波エミッタに加える電圧をｖ（ｔ）、流れる電流をｉ（ｔ）と置き、ＲＬＣ直列共振回路の伝達関数Ｈ（ｓ）を計算する。そして、ラプラス変換を用いてインパルス応答を計算し、超音波は超音波エミッタから離れると回折が現れ、その効果を入れることは１階時間微分することに対応し、式１を得る。 FIG. 6B is a diagram showing the sensitivity frequency characteristic of this ultrasonic sensor. That is, it shows the result of driving with a voltage amplitude of 5.47 V, changing the frequency from 34 kHz to 46 kHz, and measuring the sound pressure at a position 4 m on the sound axis from the ultrasonic emitter at that time. However, the horizontal axis of this figure is the resonance frequency of about 40.2 kHz, and the vertical axis is dimensionless at the sound pressure amplitude at that time 21.2 Pa (SPL 117.5 dB).
Here, a method of modeling and formulating as an RLC series resonant circuit will be described. The mechanical resonance is equivalent to a series circuit of a coil (mass), a capacitor (spring), and a resistance (mechanical resistance) in the vicinity of the resonance. Assuming that the ultrasonic wave is a plane wave, it can be modeled that the ultrasonic wave is radiated in proportion to the current flowing through the circuit. The impulse response h (t) of the ultrasonic emitter is expressed by the following equation (1).
(Formula 1)

It is represented by Here, A is an arbitrary constant, ω _d = ω ₀ · (1-1 / 4Q ₀ ² ) ^1/2 , ω ₀ = 2πf ₀ ) is the resonance frequency, Q ₀ is the sharpness of resonance, and u (t) is Unit step function. In the derivation method of this equation, first, the voltage applied to the ultrasonic emitter is set to v (t), the flowing current is set to i (t), and the transfer function H (s) of the RLC series resonance circuit is calculated. Then, the impulse response is calculated using Laplace transform, and when the ultrasonic wave is separated from the ultrasonic emitter, diffraction appears, and adding the effect corresponds to first-order time differentiation, and Equation 1 is obtained.

The solid line in FIG. 6B is a fitting curve when Q ₀ = 20, and the dotted line is a fitting curve when Q ₀ = 30. As in this fitting, the value of Q ₀ is different in both frequency regions across the resonance frequency. Unfortunately, it has been found that the values of f ₀ and Q ₀ tend to decrease as the drive voltage increases. Originally, frequency characteristics must be measured near the emitter in order to suppress ultrasonic diffraction. Ultimately, it is necessary to determine the frequency characteristics of the ultrasonic emitter including the secondary effects as described above, but this is not included in this modeling.

The drive voltage of the ultrasonic emitter is the output signal x (t) from the modulation simulator (when the amplification factor of the power amplifier is included in the signal, x (t) is multiplied by the amplification factor) and is therefore emitted from the emitter. The ultrasonic sound pressure p _c (t) is given by a convolution integral (Expression 2) of x (t) and h (t).
(Formula 2)

(III) Demodulation process (down-conversion)
Parametric sound pressure signal p is generated as a result of self-demodulation process of the ultrasonic signal p _c. A simple model of the process is represented by Berktay's equation (Equation 3) and Merkinger's equation (Equation 4).

（式４）

(Formula 3)

(Formula 4)

とおいている。また、ρ_０、βは、それぞれ媒質の密度、非線形係数、またｃ_０は音速、α_ｃはキャリア超音波の音波吸収係数である。
Ｂｅｒｋｔａｙの式３は、包絡線Ｅ（ｔ）を時間ｔで２階微分したものがパラメトリック差音に比例することを示しており、物理的には２乗検波を行っているにすぎない。特に、キャリア超音波の音圧ｐ_０が低くてδ＜＜１が満たされるような場合は、式４を低次で近似するとＢｅｒｋｔａｙの式３が得られる。従って、Ｍｅｒｋｌｉｎｇｅｒの式は、Ｂｅｒｋｔａｙの式を包含した広範囲の１次波駆動に対して有効といえる。 In these expressions 3 and 4, the relationship between the sound pressure signal p _{0 of} the ultrasonic wave emitted from the emitter, the envelope function E (t), and the carrier angular frequency ω _c is
(Formula 5)

I keep it. Further, ρ ₀ and β are the density and nonlinear coefficient of the medium, c ₀ is the sound velocity, and α _c is the sound wave absorption coefficient of the carrier ultrasonic wave.
Berktay's equation 3 shows that the second derivative of the envelope E (t) with respect to time t is proportional to the parametric difference sound, and is only physically square-detected. In particular, when the sound pressure p ₀ of the carrier ultrasonic wave is low and δ << 1 is satisfied, Berktay's equation 3 is obtained by approximating equation 4 with a low order. Therefore, it can be said that the Merklinger equation is effective for a wide range of primary wave driving including the Berktay equation.

  By the way, the magnitude of the parametric difference sound is given by a second-order time derivative of a function having envelope information. This differentiation means that the parametric difference sound is proportional to the square of the frequency in the frequency domain. Strictly speaking, the frequency characteristics of sound pressure depend on the distance of the ultrasonic emitter. FIG. 6B shows the difference in sound pressure when the aperture radius a = 100 mm, one of the primary waves, that is, the carrier frequency is fixed at 40 kHz, and the other primary wave frequency is changed to less than 40 kHz (LSB). The frequency response is shown.

Here, the KZK equation was used for the theoretical calculation. The sound wave surface sound pressures p ₀ of the two primary waves are fixed at 128 dB. From this theoretical result, when the distance is 1 m, the frequency f is 1.6 to the power of 1.6, from 4 m to 10 m is from the 1.7 power to the 1.8 power, and when it is further 20 m, it approaches the square. However, the farther the distance is, the greater the distance from the quadratic curve is affected by sound wave absorption. Incidentally, since the sound absorption at 4 kHz is about 0.03 dB / m, the attenuation is 1 dB or less even when propagating 20 m.

FIG. 7 shows the frequency response characteristics of SPL measured with an ultrasonic sound pressure signal p ₀ = 130 dB, distances of 1 m, 4 m, 10 m, and 20 m. The time differentiation was substituted with a 7th-order numerical differentiation.

The fact that the difference sound is proportional to the square of the frequency (f ² ) indicates a far field. At long distances, the effect of sound wave absorption at high frequencies is seen. The near field is proportional to the 1.5th power of the frequency.

(Summary)
(1) The modulator described in the first embodiment of the present invention receives an audible signal from an external sound source by a software program, converts the audible signal into numerical data, and further performs various numerical operations on the numerical data. To generate and output numerical data of the modulated signal.
(2) According to the modulator, it is easy to handle a high-resolution sound source of a modulated signal. (3) According to the modulator, in addition to generation of a modulated signal, an input sound source signal is changed to an output signal. Any calculation function to achieve at least one function selected from level adjustment in every process, adjustment of frequency characteristics and phase characteristics that are filter functions, sound quality improvement for distortion reduction, and power saving It can also be provided.
(4) The above converter has a function of accepting an audible signal file from an external sound source, or a streaming function of sequentially accepting an audible signal from an external sound source in a buffer memory and directly outputting it to the outside. You may provide at least any one of the function which produces | generates the signal file of a sound source.
(5) The converter can replace all or part of the function described in the item (4) with a physical device configured by hardware.

  Since a high-resolution sound source is used, it is easy to send a signal via Bluetooth (registered trademark) and perform remote control. In addition, since the high-resolution sound source includes a band signal of audible sound, it also includes a parametric speaker signal and a conventional speaker (an AI speaker is also this type), and since it is a sound source that can be driven together, it is excellent in versatility. .

(Second Embodiment)
In the series of processes described above, a method of simulating on the computer the flow of all signals from the step of modulating the input sound source signal to the time of self-demodulation will be described.

-Ultrasonic emitter simulator-
Usually, the ultrasonic signal emitted from the ultrasonic emitter is down-converted to a frequency in the audible band by self-demodulation. Since the parametric speaker drive signal has a frequency component of approximately 40 kHz, this signal is handled as a high-resolution sound source, and the ultrasonic emitter is modeled and formulated as a simple RLC series resonant circuit. Can be reproduced. In the experiment, all signal flows were high-resolution sound sources (sampling frequency 96 kHz, unified with 16-bit quantization), and input and output were both wav format sound source data files, which were recorded and played back on a personal computer or smart phone. I was able to do it. Real-time processing will be described in detail in the third embodiment.

  In the experiment, display was performed according to Bekta's equation 4, but Meklinger's equation may be used. The derivative with respect to time is replaced with a seventh-order numerical derivative.

  FIG. 8A shows a time waveform showing an example of an input sound source signal and a demodulated signal obtained by simulation, and FIG. 8B shows a frequency spectrum of FIG. 8A. The input sound source signal was an audio signal s (t) = sin ωt + sin 3ωt (ω = 2π × 1000 rad / s) obtained by adding sine waves of equal amplitude of 1 kHz and 3 kHz, and the signal was input to the modulation process. Here, the amplitude of the modulated ultrasonic carrier 40 kHz and the 37 kHz and 39 kHz components of the LSB are all equal and input to the ultrasonic emitter.

  By using the ultrasonic emitter simulator as described above, the process from modulation to demodulation was performed on a computer. As the amplitude modulation method, the dynamic SSB modulation improved so that dynamic modulation can be performed based on the Weaver method is used this time, but other modulation methods may be used as long as the modulation method is expressed by a mathematical expression.

  As described above, the parametric speaker development system described in the present embodiment can simulate everything from an input sound source signal to a demodulation process on a computer, and a parametric signal having a desired quality (such as low distortion and input / output linearity). Can be realized on software. According to the present embodiment, since the demodulated sound obtained as a result of self-demodulation of the ultrasonic signal emitted from the ultrasonic emitter is reproduced on the computer, the ultrasonic emitter is not necessary and the system is adjusted. At this time, it is not necessary to consider the consistency of each block as in an analog circuit. In addition, troublesome and time-consuming field experiments are unnecessary, and costs and manpower associated with the experiments can be reduced.

  Although the self-demodulation process is modeled according to Berktay's equation, it goes without saying that other equations such as Meklinger can be used.

(Third embodiment)-Realization method of real-time processing-
In both the first and second embodiments, an example in which data in a general-purpose digital sound source format is used as an input sound source signal has been described. In this case, however, a file conversion process and input / output to a high-resolution recording / playback device are performed. Since the accompanying process is required, the flow becomes a batch process that is discontinuous in time. However, real-time processing is possible if the algorithm is devised and programmed with software capable of high-speed processing.

  For example, a continuous audio signal (in the case of non-digital data, through an A / D converter) is input, the above steps are executed continuously, and a signal corresponding to high resolution is supplied to an amplifier and an ultrasonic emitter. The method can be considered.

  If the input sound source is an analog signal, it is sufficient to continuously convert the input sound source signal of the digital sound source through the A / D converter into text. Also, the output format can be changed in a timely manner according to the design of the system, such as outputting to a D / A converter if necessary.

  FIG. 9 is a block diagram of a parametric speaker according to the third embodiment. A portion indicated by a broken line is a modulator. Batch processing is represented by a one-dot chain line, and real-time processing is represented by a solid line.

  In the case of batch processing, the input sound source signal s (t), which is a sound data file in wav format, is sent from the input sound source 90 to the converter 92 and subjected to numerical calculation by the processor 93, as will be described with a dashed line in FIG. Thus, an ultrasonic signal is generated. The output file of the converter may be stored in the recording medium 95 as necessary. The output data of the converter is sent to the high resolution recording / reproducing device 94. Further, an ultrasonic signal is continuously emitted from the ultrasonic emitter 100 in real time via the amplifier 96.

  On the other hand, real-time processing as described with a solid line in FIG. 9 is also possible. The input sound source signal in the audible band provided in real time from the input sound source 90 is subjected to numerical calculation to generate an ultrasonic signal, and then the output signal via the audio interface 98 is output by the audio interface that can handle the ultrasonic band. / A (digital / analog) conversion into an analog signal, which is then sent to the amplifier 96, where the ultrasonic signal is continuously emitted from the ultrasonic emitter 100 in real time. In this case, both the hardware configuration and the software need to use a configuration capable of performing necessary processing within a time that can be processed in real time. Specifically, for example, a function that accepts an audible signal file from an external sound source, a streaming function that sequentially accepts an audible signal from an external sound source into a buffer memory and directly outputs it to the outside, and generates a signal file of a high-resolution sound source Functions.

  Alternatively, everything from the sound source output signal to the internal processing of the amplifier can be incorporated digitally, ensuring the consistency of the entire system and realizing downsizing. On the other hand, in the audio interface or the high-resolution recording / reproducing device 94 of FIG. 9, if the D / A converter is equipped with an amplifier, or if it is a complete digital amplifier that can handle the ultrasonic band, D / A conversion, etc. When the digital signal of the software simulator is directly input to the amplifier without passing through the analog converter, the output can be supplied as an analog signal to the ultrasonic emitter. In FIG. 9, the block having the function and incorporating the amplifier is described as a high-resolution digital amplifier 97. On the other hand, if the software simulator has a D / A conversion function and can output an analog signal, the output of the software simulator may be passed to an amplifier.

  The present invention is not limited to the above embodiment, and can be appropriately changed within the scope described in the claims.

20 Broken line (functional block for realizing dynamic SSB modulation)
90 input sound source 92 modulator 93 processor 94 high resolution recording / reproducing device 95 external recording medium 96 amplifier 97 high resolution digital amplifier 98 audio interface 100 ultrasonic emitter

Claims (7)

A modulator for modulating a carrier signal in an ultrasonic band with an audible signal from an external sound source to generate a modulated signal in the ultrasonic band,
The modulator includes an ultrasonic transmitter for generating a carrier signal and a processor,
By the processor,
Step S1 for preparing digital sound source data;
Step S2 for converting digital sound source data into text data;
Step S3 for generating an ultrasonic signal by performing a numerical operation on a digital sound source data included in the text data and a carrier signal generated based on the ultrasonic transmitter;
Step S4 for inversely converting the result of the numerical calculation into digital sound source data;
A modulator characterized by causing   The modulator according to claim 1, wherein the digital sound source data is high-resolution sound source data.   The numerical calculation by the processor is at least one selected from level adjustment in the process from the input sound source signal to the output signal, adjustment of frequency characteristics and phase characteristics as filter functions, sound quality improvement for distortion reduction, and power saving. The modulator according to claim 1, further comprising a calculation processing step for providing a numerical calculation function for realizing one function.   One of the function of accepting an audible signal file from an external sound source, the streaming function of sequentially receiving the audible signal from the external sound source into the buffer memory and directly calculating the result, and the function of generating a signal file of a high-resolution sound source The modulator according to claim 1, further comprising: a modulator. A modulator according to claims 1 to 4;
A parametric speaker comprising: an amplifier that amplifies the modulated signal; and an ultrasonic emitter that emits an ultrasonic wave based on the amplified modulated signal.   An ultrasonic emitter simulator that treats an ultrasonic signal as a high-resolution sound source unified with a sampling frequency of 96 kHz and quantization of 16 bits, and is modeled as an RLC series resonance circuit.   A parametric speaker development system comprising the modulator according to claim 2 and the ultrasonic emitter simulator according to claim 6.

Images