HK1238785B - Transmission method of arbitrary signal using acoustic sound - Google Patents

Description

Method for transmitting arbitrary signals using sound

Technical Field

The present invention relates to a method for transmitting an arbitrary signal using sound, which is capable of adding an arbitrary signal in an audible frequency band to a sound such as music without significantly affecting the quality of the original sound.

Background Art

Conventionally, various electronic watermarking technologies have been developed that embed arbitrary signals in a frequency band that is indiscernible or difficult for humans to recognize in sounds such as music and transmit the arbitrary signals together with the sounds.

Prior art literature

Patent Literature

Patent Document 1: Japanese Patent Application Laid-Open No. 2011-530939

Summary of the Invention

Problems to be solved by the invention

However, the above-mentioned conventional method of transmitting an arbitrary signal using sound has the following problems.

(1) When using an arbitrary signal in a frequency band that humans cannot recognize, humans cannot hear the sound at all, so arbitrary signals can be transmitted without affecting the quality of the sound. However, in the case of this method, data in frequency bands other than the sound data in the human audible frequency band (the frequency band that humans can recognize) is required, which increases the amount of data. In addition, transmission and reception using general-purpose sound equipment designed based on the audible frequency band are limited, so sometimes dedicated equipment must be newly produced.

(2) When using an arbitrary signal in a frequency band that is difficult for humans to recognize, the frequency of the arbitrary signal, while difficult to recognize, is still within the human audible range and may therefore affect the atmosphere of the original sound (noise, distortion, discomfort in the scale, etc.). In addition, the temporal and frequency positions of the insertion may also be limited by the sound pressure level and frequency components of the original sound.

The present invention has been made in view of the above-mentioned points, and its object is to provide a method for transmitting an arbitrary signal using sound, which is unlikely to affect the atmosphere (quality) of the original sound even if an arbitrary signal in the audible frequency band is added to the sound such as music.

Means for solving problems

The present invention is a method for transmitting an arbitrary signal using sound, in which an arbitrary signal is implanted in a sound composed of multiple sounds and transmitted. A separable sound is found from the multiple sounds that constitute the sound, or a separable sound is inserted into the multiple sounds that constitute the sound. The separable sound has an essential part and an incidental part in a manner that can be separated on the time axis or the frequency axis. The essential part makes a main contribution to the identification of the separable sound, and the incidental part makes an incidental contribution to the identification of the separable sound. The incidental part of the found or inserted separable sound is changed into a signal graph of the arbitrary signal, and the arbitrary signal is transmitted through the changed sound.

Here, the so-called incidental part that makes an incidental contribution to the recognition of the sound can be said to be the sound produced when the energy of the vibration is transferred to the resonator (resonator). For example, if we use a guitar as an example, the sound at the moment when the string is plucked with a pick is equivalent to the incidental part. It is not the essential sound of the guitar itself. In the case of clapping or hitting the drum, the sound at the moment when the palm, drumstick, etc. collides with the palm or drum of the other party is equivalent to the incidental part. These sounds are hereinafter referred to as "hitting sounds".

On the other hand, an object (a resonant body, such as a musical instrument or hand) that receives vibrational energy from the striking sound produces a sound according to its own resonance pattern. The sound inherent to the object is based on this resonance pattern, and this is the essential part that contributes to the recognition of the sound.

Through experiments, the inventors of this application discovered that, in the separable sound, the essential portion is the part that gives the sound its characteristic features, while the incidental portion is only perceived as the portion that imparts the sound's atmosphere. Furthermore, they discovered that when the incidental portion is removed from the sound, humans can still recognize the original characteristics of the sound through the essential portion, but the sound's atmosphere is altered. Therefore, by embedding a signal pattern of an arbitrary signal in the incidental portion, it is possible to transmit an arbitrary signal while maintaining the same atmosphere as the original sound. In other words, according to the present invention, even if an arbitrary signal within the audible frequency band is added to a sound such as a musical piece, humans can perceive the sound as being substantially identical to the original sound, including the atmosphere.

Furthermore, the present invention provides a method for transmitting an arbitrary signal using the aforementioned sound, characterized in that the signal pattern of the arbitrary signal has a duration identical to that of the incidental portion, a frequency component similar to that of the incidental portion, or a sound pressure level similar to that of the incidental portion. According to the present invention, the signal pattern of the arbitrary signal is substantially identical to the signal pattern of the original incidental portion. Thus, the effect of the signal pattern of the arbitrary signal imparting an atmosphere to the sound is similar to that imparted by the original incidental portion, allowing humans to perceive the sound as being substantially identical to the original sound.

Furthermore, the present invention is a method for transmitting an arbitrary signal using the aforementioned sound, characterized in that the separable sound is any of clapping, drumming, percussion instrument sounds, cymbals, gongs, car horns, telephone calls, doorbells, whistles, or various sound effects. Thus, a person can perceive these sounds as being substantially identical to the original sound.

Furthermore, the present invention provides a method for transmitting an arbitrary signal using the aforementioned sound, characterized in that, when there are multiple incidental components, one of the incidental components is changed to a signal pattern of the positive signal of the arbitrary signal, while the other incidental components are changed to signal patterns that are complementary to the positive signal of the arbitrary signal. If the positive and complementary relationships do not match on the side receiving the arbitrary signal, the positive and complementary signal patterns are compared and the signal pattern with the higher S/N ratio is adopted. This allows for effective utilization of all of the multiple incidental components, enabling transmission of arbitrary signals with higher accuracy.

Furthermore, the present invention provides a method for transmitting an arbitrary signal using the aforementioned sound, characterized in that the intervals before and after the essential portion that produce a temporal masking effect are used as the incidental portion, and the incidental portion is modified into the signal pattern of the arbitrary signal. This method effectively utilizes the temporal masking effect to transmit an arbitrary signal.

Effects of the Invention

According to the present invention, even if an arbitrary signal in the audible frequency band is added to a sound such as music, it is possible to make a human perceive the sound as being substantially the same as the original sound including its atmosphere (quality).

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram showing the results of an experiment in which single tones of 4 kHz and 8 kHz with different durations were listened to and compared to determine whether they were perceived as intervals.

FIG. 2A is a waveform diagram showing an example of applause sound.

FIG2B is a waveform diagram showing an example of applause sound.

FIG3 is a frequency analysis diagram of the applause sound after decomposing it according to frequency bands.

FIG4A is an approximate waveform diagram of a da-on (corresponding to Japanese: da-on) in the original applause sound.

FIG. 4B is a diagram showing an approximate waveform of an arbitrary signal that is modified into the striking sound of FIG. 4A and then implanted.

FIG. 5 is a waveform diagram showing an example of a clapping sound after an arbitrary signal is embedded.

FIG. 6 is a waveform diagram showing an enlarged signal pattern b1 - 1 of an arbitrary signal.

FIG. 7 is a diagram schematically showing the arbitrary signal b1 shown in FIG. 5 .

FIG. 8 is a diagram schematically showing an example of the data content of the data series portion c3.

FIG. 9 is a diagram showing another data series c3 - 2 .

FIG. 10 is a diagram showing another data series c3 - 3 .

FIG. 11 is a diagram showing a waveform diagram of a portion of a certain music piece and a waveform diagram of a clapping sound embedded in the waveform diagram.

FIG12 is a block diagram showing the system configuration of a detection device 10 for extracting an arbitrary signal from a sound into which an arbitrary signal has been embedded.

FIG13 is a block diagram of the detection unit 17 .

FIG. 14 is a diagram showing an example of a case where the first signal series is set to positive data values and the second signal series is set to the complementary values thereof.

FIG. 15 is a diagram illustrating a determination method when the positive data value and the complementary data value have a different positive and complementary relationship.

FIG. 16 is a schematic block diagram showing an example of a rhythm synchronization system.

DETAILED DESCRIPTION

The present invention provides a method for seamlessly inserting an arbitrary signal into a sound such as music. "Never again," as used herein, "never again" means that the characteristics of the original sound remain unchanged when the arbitrary signal is inserted, and that the same sound atmosphere as the original sound is achieved.

First, let's explain the masking effect used to implement the above method. The masking effect is a characteristic of the human brain that causes multiple tones to become difficult to hear when certain conditions are met. There are two types of masking effects: simultaneous masking and temporal masking.

Simultaneous masking is the masking between two simultaneous tones, also known as frequency masking. For example, while 440Hz and 450Hz sine waves can be distinguished when listened to separately, they are in nearly the same frequency band and therefore cannot be clearly distinguished when sounding simultaneously. However, when the frequency difference between the two tones is less than a few hertz, this frequency difference causes a beat, which needs to be considered separately from masking.

Temporal masking is a phenomenon in which, when a loud sound is suddenly emitted, the sounds preceding and following it become inaudible. The masking of the preceding sound is called backward masking, while the masking of the following sound is called forward masking. This effect decreases exponentially as the time interval between the masking sound (masker) and the masked sound (masked sound) increases. For backward masking, the limit is approximately 20 milliseconds, and for forward masking, the limit is approximately 100 milliseconds.

Next, the recognition of intervals will be explained. Humans are sensitive to intervals, but the clarity of intervals decreases in the high-pitched range. For example, rhythm instruments (drums, percussion instruments, etc.) are sounds that clearly have frequencies (intervals), but due to the high-pitched range, the interval elements such as harmony in music theory become fewer. That is, the human ear can hear sounds of about 20Hz to about 20kHz, but the upper limit that can be distinguished as an interval is about 4kHz at most. When it is above about 4kHz, it becomes a sound that can be heard without a sense of interval. However, even in the high-pitched range above 4kHz, when it is played continuously for a long time, it is of course not clear but can be heard as an interval, that is, there is a limit to the playing time when it cannot be felt as an interval.

Figure 1 shows the results of an experiment comparing the perception of interval lengths at 4kHz and 8kHz single tones (sine waves) with different durations (50 milliseconds, 25 milliseconds, 12 milliseconds, 6 milliseconds, 3 milliseconds, 2 milliseconds, and 1 millisecond). As shown, the perception of interval lengths at 4kHz and 8kHz begins to diminish around 2 milliseconds, while the perception of interval lengths at 8kHz begins to diminish around 3 milliseconds, transforming into a short sideband sound like "fu" (corresponding to "フッ" in Japanese). This experiment concluded that in the high-pitched range, intervals of continuous tones of less than a few milliseconds are difficult to discern.

Next, we will illustrate the example of an clapping sound (simulated clapping) using the human body as a percussion instrument. Clapping sounds are originally the sound of a person's palms clapping together rapidly. However, in modern music, waveforms are synthesized electronically and adjusted to create easily usable instrumental melodies. Figures 2A and 2B are waveform diagrams illustrating examples of clapping sounds. As shown, the clapping sound has a waveform that lasts for two or three impulse responses of approximately 11 milliseconds, followed by a longer waveform. Furthermore, when the clapping sound waveforms shown in Figures 2A and 2B are decomposed, listening to only the first half of waveform a1 reveals a short high-pitched sound "ka" (corresponding to Japanese: カッ) or a short high-pitched sound "pu" (corresponding to Japanese: プッ). This is a sound unique to percussion instruments (clapping is also a percussion instrument using the human body). Furthermore, the characteristic 11- and 12-millisecond periods are due to the physical shape and size of the human hand, the viscosity of the skin, and other factors. The continuous time is about several milliseconds, and as mentioned above, it is a sound that is difficult to hear as an interval. Hereinafter, in this specification, the portion of waveform a1 is referred to as a "hitting sound."

On the other hand, by listening to only the second half of waveform a2 of the waveforms shown in Figures 2A and 2B, it is understood that this is the main tone that constitutes the interval of the applause sound. In other words, it is understood that the second half can be said to be the dominant tone that characterizes the applause sound, which is the attenuation of the natural vibration frequency of materials such as the palm under the resonance conditions around the palm. Hereinafter, in this specification, waveform a2 is referred to as the "essential tone."

Furthermore, the three hits in waveform a1 shown in Figure 2A were listened to and compared when the sounds were eliminated one by one, starting with the earliest hit. This experiment revealed that the greater the number of hits, the thicker the applause sound appeared (i.e., these hits enhanced the atmosphere (quality) of the applause sound). However, regardless of the number of hits, there was essentially no significant change in the pitch of the applause sound, regardless of whether there were one, two, or three hits. This is believed to be due to the effects of temporal masking (particularly reverse masking). In particular, the two hits, which occur approximately 20 milliseconds from the true tone, are masked, resulting in a completely dominated, inconspicuous sound (however, as mentioned above, these hits contribute significantly to the atmosphere). In addition, it can be confirmed that when only the essential sound part of the applause sound in Figure 2A is attenuated to half the level (that is, the amplitude of only the waveform a2 part in Figure 2A is attenuated to half the amplitude (-6dB)) and compared with the unattenuated applause sound, the atmosphere of the applause sound remains unchanged and the overall sound volume is reduced.

Based on the above, it can be seen that the applause sound is recognized as the essential sound in the second half being dominant both in terms of continuous sound and masking sound, while the applause sound in the first half is a sound that only enhances the atmosphere (hereinafter referred to as "ambient sound").

Therefore, the inventors of this application used the second half of the applause sound (the essential sound portion) as the existing applause sound, and modified the first half (the hitting sound portion) to use the signal pattern of an arbitrary signal. The signal pattern to be embedded in this process preferably uses a duration identical to that of the original hitting sound, frequency components similar to those of the original hitting sound, and a sound pressure level similar to that of the original hitting sound. This allows the arbitrary signal to produce an ambient sound effect similar to that of the original hitting sound. Consequently, a sound nearly identical to the original applause sound can be created.

Next, the analysis results of the applause sound are explained. Figure 3 is a frequency analysis diagram that decomposes the applause sound shown in Figure 2B into corresponding frequency bands. Each extracted frequency band approximates the interval frequency band (Octave Number) with an integer value. Octave #3 is the main melody frequency band (near the center of the piano 88 keyboard). As can be seen, in the case of the applause sound shown in the figure, the components are dispersed throughout the entire frequency band in the mid-high range, but when focusing on and analyzing the shape of the impulse response of the striking sound, the components at the positions (three positions) surrounded by the dotted lines are particularly significant. In other words, the center of the applause sound is around 4 to 8 kHz.

Based on the fact that the center of such applause sound is around 4 to 8 kHz and the conclusion that the interval of continuous sound of several milliseconds in the frequency band above 4 kHz is difficult to recognize as described above, it is considered that the frequency of any signal implanted as a substitute (simplification) for the applause sound is preferably a single tone (sine wave) of Octave #7 (4 to 8 kHz).

Therefore, as the frequencies of the arbitrary signals used in this embodiment, 4kHz and 8kHz, and then 16kHz, were selected as a representative example of Octave #7. The reason for choosing 4kHz and 8kHz was to consider the performance of the bandpass filter used for frequency extraction. In general, increasing the order of a bandpass filter increases complexity and cost. Therefore, as a reference standard, a second-order filter, easily implemented using analog passive circuits, was used to consider the detection circuit. Its attenuation rate is -12dB/oct. Therefore, even accounting for component variations, the effect of adjacent tones is reduced to approximately 1/4 to 1/2 (the transmission band is from a single tone to an octave), resulting in relatively low waveform distortion. Furthermore, since it is an overtone, waveform distortion can be simplified (easily understood) if phase management is possible, making actual design easier. Furthermore, 16kHz was chosen because it is close to the inaudible range, yet is an overtone of 8kHz, and its signal-to-noise ratio (S/N) is improved, making it a signal with little impact on the music.

FIG4A shows an approximate waveform (envelope) of a striking sound of the original applause sound (a striking sound in a waveform a1 of FIG2A or FIG2B), and FIG4B shows an approximate waveform (envelope) of an arbitrary signal that is changed into the striking sound and implanted. However, each arbitrary waveform only shows the positive side portion (i.e., as a half-wave). The approximate waveform of the striking sound shown in FIG4A decays in a time period of about half the period of the striking sound (more than ten milliseconds). On the other hand, as the waveform of the arbitrary signal, as shown in FIG4B, an intermittent waveform of the same amplitude and the same waveform is used. Moreover, the energy accumulation of the waveform of the striking sound shown in FIG4A and the waveform of the arbitrary signal shown in FIG4B is roughly the same (i.e., the area is roughly the same). The waveform of the arbitrary signal is set to this waveform because signal transmission using the same attenuation waveform as the original striking sound is predicted to be difficult for signal detection on the receiving side. Therefore, by setting the waveform of the arbitrary signal to a waveform obtained by dividing a simple waveform of the same amplitude into multiple stages, signal detection on the receiving side is facilitated. As mentioned above, the hitting sound is not recognized as a pitch, so even if it is changed, it has little effect on the pitch of the applause sound. On the other hand, as mentioned above, the waveform of the original hitting sound and the waveform of the arbitrary signal are configured to have almost the same energy accumulation, so they have the same effect as ambient sound. Therefore, the arbitrary signal can also be called a simulated hitting sound.

FIG5 is a waveform diagram showing an example of an applause sound in which an arbitrary signal (simulated applause sound) is implanted instead of the original applause sound (two positions in the first half). As shown in the figure, the signal patterns b1-1 and b1-2 of the arbitrary signal are implanted in the two positions in the first half of the waveform. As for the essential tone b2 in the second half, the same sound as the original essential tone of the applause sound is used. In addition, FIG6 is a waveform diagram showing an enlarged representation of the signal pattern b1-1 (or b1-2) of an arbitrary signal. As shown in the figure, the signal pattern b1-1 (b1-2) is configured such that a data series portion (signal series) c3 is arranged between detection marks (pre-synchronization code portion and post-synchronization code portion) c1 and c2 provided at the front and rear. The waveform of the arbitrary signal b1-1 (b1-2) is formed by adding together all the sine waves of the same amplitude at 4kHz, 8kHz, and 16kHz. Basically, data transfer is complete with the first-stage signal pattern b1-1, so it's best to use the second-stage signal pattern b1-2 for subsequent processing (e.g., unique interpolation, error correction, etc.) of the first-stage signal pattern b1-1. Various methods are possible for subsequent processing, but in this example, the same data sequence c3 is repeated. As described below, adding complementary data (e.g., 0 for 4, 1 for 3, etc.) to the second-stage signal pattern b1-2 to improve detection accuracy can have a beneficial effect on the S/N ratio.

Figure 7 schematically illustrates an arbitrary signal (simulated beeping sound) b1-1 (b1-2) shown in Figure 5 . As shown in the figure, the preamble c1, which indicates the beginning of the data of arbitrary signal b1-1 (b1-2), consists of a valid and invalid period of 1.5 milliseconds, followed by an invalid period of more than 2 milliseconds. Furthermore, the postamble c2, which indicates the end of the data, consists of a valid and invalid period of 1.5 milliseconds. After being received by the receiving end, arbitrary signal b1-1 (b1-2) passes through bandpass filters for each frequency band (4 kHz, 8 kHz, 16 kHz), undergoes envelope detection (full-wave rectification), and then passes through a lowpass filter at approximately 2 kHz (-12 dB/Oct).

Figure 8 schematically illustrates an example of the data content of data sequence portion c3 shown in Figures 5 and 7. This data sequence portion c3 utilizes a data count method, where the data content is based on the number of valid counts within the sequence. This method, as shown in Figure 8, limits transmission to only five types of content, but does not detect bit positions, making it highly effective for signal-to-noise ratios. Furthermore, the two arbitrary signals b1-1 and b1-2 shown in Figure 5 both become Data 4 in Figure 8.

Figures 9 and 10 show other data sequence parts c3-2 and c3-3. The data sequence shown in Figure 9 uses a conventional digital bit (4-bit) method. This method can transmit 16 types of content, but requires bit position detection. The data sequence shown in Figure 10 uses a digital bit (7-bit) method that does not include invalid bits between bits. This method can transmit 128 types of content, but requires sufficiently accurate bit position detection.

The waveform diagram shown in the upper section of FIG11 is a waveform diagram of a part of a certain music, and a plurality of applause sounds of the present invention are implanted. That is, the waveform shown in the middle section of FIG11, i.e., two applause sounds (qiaqia (corresponding to Japanese: チャッチャッ)) and the waveform shown in the lower section of FIG11, i.e., four single applause sounds (qia (corresponding to Japanese: チャッ)) are added. If there are no applause sounds in the original music, the applause sounds are newly implanted. In addition, if there are already applause sounds in the original music, they are replaced with the applause sounds shown in FIG5 above, so that the original music is maintained. Moreover, the striking part of each applause sound becomes an arbitrary signal, so by taking it out on the receiving side, the arbitrary signal can be utilized. For example, the LED of the portable terminal that receives the music can be made to light up in response to the applause sounds.

However, separable sounds that have an essential tone (essential part) that makes a major contribution to the recognition of a sound and a percussion tone (incidental part) that makes an incidental contribution to the recognition of the sound in a separable manner are not limited to the above-mentioned applause sounds, but can also be sounds from various other rhythmic instruments. That is, various rhythmic instrument sounds have an essential tone and a percussion tone in one sound. And among these rhythmic instrument sounds, in addition to sounds that have percussion and essential tone that can be separated on the time axis, there are also sounds that can be separated on the frequency axis, similar to the above-mentioned applause sounds. The so-called sounds that can be separated on the time axis are sounds in which the essential tone and the percussion tone exist at different times, such as the above-mentioned applause sounds. The so-called sounds that can be separated on the frequency axis are sounds that can be separated according to frequency due to the different frequencies of the essential tone and the percussion tone. In particular, separation by frequency is effective when the percussion tone and the essential tone overlap in time, but it can also be used even when the two sounds do not necessarily overlap. In the case of such separable sounds that can be separated by frequency, it is sufficient to replace the percussion tone that exists at a frequency different from the essential tone with an arbitrary signal.

Furthermore, the present invention can be applied not only to various rhythm instrument sounds but also to various other sounds, as long as the separable sound has an essential part and an incidental part in a manner that can be separated on the time axis or the frequency axis.

Furthermore, in the embodiment using the applause sound described above, the arbitrary signal pattern used had a duration identical to that of the incidental portion, a frequency component similar to that of the incidental portion, and a sound pressure level similar to that of the incidental portion. In other words, the arbitrary signal pattern fully satisfied the above three conditions. This allowed the sound after the arbitrary signal was implanted to be nearly identical to the original sound, not only in terms of pitch but also in terms of ambiance. However, even if the arbitrary signal pattern satisfied only one of the above three conditions, it would still produce an effect that approximated the atmosphere of the original sound.

〔Detection of Implanted Sound〕

FIG12 is a block diagram showing the system configuration of a detection device 10 for extracting an arbitrary signal from a sound into which an arbitrary signal has been embedded. As shown in the figure, the detection device 10 comprises a microphone 11, an ADC/AGC 13, an input segment envelope detector 15, a detector 17, and a controller 19.

ADC stands for Analogue/Digital Converter, and AGC stands for Automatic Gain Controller. The analog sound signal input from speaker 21 to microphone 11 is sampled by the ADC and converted into digital data. Meanwhile, the AGC adjusts the gain so that the time-integrated DC signal input from input segment envelope detector 15 is equal to the reference voltage.

Figure 13 is a block diagram of the detection unit 17. As shown, the digital audio signal input from the ADC/AGC 13 passes through three bandpass filters 171, 173, and 175 to extract a 4kHz, 8kHz, and 16kHz extracted signal, respectively. Each extracted signal then passes through envelope detectors 189, 191, and 193, each consisting of absolute value circuits 177, 179, and 181, and lowpass filters 183, 185, and 187, respectively, to generate a 4kHz, 8kHz, and 16kHz amplitude signal, respectively, consisting of the amplitude envelopes of the extracted signals. These amplitude signals are then summed, binarized, and input to a discrimination circuit 195. Based on the binarized data, the discrimination circuit 195 verifies the accuracy of the detection flag and data sequence and outputs the final detected data.

Regarding the signal patterns b1-1 and b1-2 of the first and second simulated beating sounds shown in Figure 5 above, they are set to the same signal pattern. That is, as shown in Figure 6, when the data series c3 of the first and second times is set to the data counting mode, it becomes "4". On the other hand, compared with continuously embedding the same content as in Figure 5, embedding supplementary content is more advantageous for the S/N ratio. Figure 14 shows an example in which the data series of the first simulated beating sound is set to positive data values and the data series of the second simulated beating sound is set to its complementary value (the sum of the two is 4). The reason for the advantage in S/N ratio is that if the invalid interval when the number of digits in the data series is expanded, the detection position becomes clearer accordingly, and the accuracy is improved. Therefore, in the case where the positive and complementary relationship between the positive data values and the complementary data values is inconsistent, as shown in Figure 15, giving priority to the data with fewer numbers can further improve the data accuracy.

12, the detection data output from the detection unit 17 is input to the control unit 19, which issues various control instructions. The control unit 19 is an application layer and can be considered to have various controls. Here, rhythm synchronization control will be described.

If a receiver receives any signal from a piece of music and performs any action related to the music, then extracting the rhythm of the music is essential. For example, even if an LED on a mobile device on the receiver side is flashing in time with the rhythm of the music, if it deviates from the rhythm of the music, the enjoyment will be lost.

Figure 16 is a schematic block diagram showing an example of a rhythm synchronization system. As shown in the figure, in this example, rhythm synchronization is performed if data of a certain value is received from any signal detected in the music. The rhythm counter is a full-count and limited counter. When the rhythm counter 50 receives data of a certain value as a rhythm synchronization instruction signal (reset signal), the counter is reset. Furthermore, if a rhythm synchronization instruction signal arrives when the rhythm counter is not fully counted, the value is saved as a preset value, thereby performing the flywheel function. Therefore, it is necessary to continuously send signals twice at the timing required for rhythm synchronization. Moreover, if there is a sufficient amount of data to be transmitted, complex operations such as syncopation can be performed. In other words, if the syncopation is an eighth note, even if the rhythm synchronization instruction arrives at the timing of the syncopation, the counter value can be returned to the past by the amount of the eighth note to determine the start position of the measure.

The above describes the embodiments of the present invention, but the present invention is not limited to the above embodiments, and various modifications can be made within the scope of the technical ideas described in the claims, the specification, and the drawings. In addition, even if any structure is not directly described in the specification and the drawings, as long as it plays the role and effect of the present invention, it is also within the scope of the technical ideas of the present invention. For example, in the above embodiments, music is shown as the sound into which an arbitrary signal is implanted, but the present invention can not only implant music, but also various other sounds (such as car horns, telephone calls, doorbells, whistles, various sound effects, etc.).

Explanation of symbols

a1: Waveform (sound)

a2: waveform (essential tone)

b1-1: Signal graph

b1-2: Signal graphics

b2: Essential tone (essential part)

c1: Preamble (detection mark)

c2: post-synchronization code (detection mark)

c3: Data series part (signal series)

c3-2, c3-3: data series

10: Detection device

11: Microphone

13: ADC/AGC

15: Input segment envelope detection unit

17: Testing Department

171, 173, 175: Bandpass filters

177, 179, 181: Absolute value circuit

183, 185, 187: Low-pass filter

189, 191, 193: Envelope detection unit

195: Discrimination Circuit

19: Control Department

Claims (6)

1. A method for transmitting arbitrary signals using sound, comprising embedding and transmitting arbitrary signals into a sound composed of multiple tones, characterized in that, The method involves identifying separable tones from the plurality of tones constituting the sound, or inserting separable tones into the plurality of tones constituting the sound. These separable tones have an essential component and a secondary component, which are separable in a manner that allows them to be separated along a time axis or a frequency axis. The essential component makes a primary contribution to the identification of the separable tones, while the secondary component makes a secondary contribution. The incidental portion of the identified or inserted separable tone is transformed into a signal pattern of the arbitrary signal, and the arbitrary signal is transmitted through the altered sound. 2. The method for transmitting any signal using sound as described in claim 1, characterized in that, The signal pattern of the arbitrary signal is the same period as the period of the attached portion, or a frequency component similar to the frequency component of the attached portion, or a sound pressure level similar to the sound pressure level of the attached portion. 3. The method for transmitting any signal using sound as described in claim 1, characterized in that, The separable sound is any sound from various effects, such as clapping, percussion, car horn, telephone call, doorbell, whistle, or other sound effects. 4. The method for transmitting any signal using sound as described in claim 3, characterized in that, The sounds of the percussion instruments include any of the sounds of drums, cymbals, and gongs. 5. The method for transmitting any signal using sound as described in claim 1, characterized in that, In the case of multiple accompanying portions, any accompanying portion is transformed into a signal pattern of the positive signal of the arbitrary signal, and any other accompanying portion is transformed into a signal pattern that has a complementary relationship with the positive signal of the arbitrary signal. When receiving any signal, if the positive and complementary signals do not match, the signal patterns of the positive and complementary signals are compared, and the signal pattern with the higher S/N ratio is adopted. 6. The method for transmitting any signal using sound as described in claim 1, characterized in that, The intervals that produce time-delayed masking effects existing before and after the essential part are taken as the incidental part, and the incidental part is changed into the signal pattern of the arbitrary signal.