Ultrasonic control device for an electrical or electronic device

Abstract

The invention relates to a control device (7) for an electrical or electronic apparatus (3) comprising: - a microphone (13) arranged to convert an acoustic signal into an electrical signal, - a processing circuit (19) arranged to detect a sinusoidal ultrasonic component within an acoustic signal from the corresponding electrical signal, - a linear frequency modulation chirp demodulator (21) arranged, in response to the detection of a sinusoidal ultrasonic component, to switch from a standby mode to a demodulation mode in which the demodulator (21) is arranged to detect a linear frequency modulation chirp ultrasonic component within an acoustic signal from the corresponding electrical signal and, where applicable, to extract an on or off instruction, and - a communication module (23) arranged to transmit an on or off control signal to the apparatus (3). [Fig. 1]

Description

Title of the invention: Ultrasonic control device for an electrical or electronic device

[0001] The field of the invention relates to the ultrasonic control of an electrical or electronic device.

[0002] The latest technological advances in the field of home automation and connected devices have coincided with the development of wireless control, to which users have become accustomed. Wireless control offers many advantages, such as flexibility and comfort in the remote management of devices, which can be controlled without having to physically move.

[0003] The gradual integration of wireless control into everyday life is encouraging users to turn away from electrical or electronic devices which, despite technical progress, can only be controlled by wired means. The lack of wireless control makes the use of such devices less convenient and less intuitive for users.

[0004] It should also be noted that devices without such functionality are more prone to excessive electrical energy consumption, due to the fact that they cannot be switched off under all circumstances, for example when the user is not nearby.

[0005] There is therefore an interest in providing wireless control to electrical or electronic devices that do not have it.

[0006] To do this, it is known to use an intermediate device, that is to say a device accessible remotely and capable of being connected to an electrical or electronic device. A user can then send, for example by means of a mobile terminal such as a touchscreen tablet or a smartphone, instructions to the intermediate device, which interprets these instructions and then controls the device to which it is coupled accordingly.

[0007] However, the electrical energy consumption of the intermediate device inherent in its operation tends to reduce the real benefit of using such an intermediate device, or even to cancel it out.

[0008] In this respect, noise detectors are, among the intermediate devices of the prior art, renowned for their low electrical energy consumption.

[0009] US patent US 6,756,700 B2 thus describes an input device equipped with a noise detector and configured to switch to energy saving mode when its components remain inactive for a predetermined period of time. The noise detector is arranged to detect physical contact with a user or friction with the surface on which the input device is placed and, if necessary, to interrupt the power saving mode. Although equipped with a filter - typically a bandpass filter - the robustness of the noise detector against ambient noise or the Doppler effect is quite limited. Furthermore, the noise detector only allows the input device to be switched on, not switched off.

[0010] To improve resistance to sound disturbances such as ambient noise or the Doppler effect, it is possible to use linear frequency modulation (LFM), also known as chirp modulation. Such modulation makes it possible to generate a pseudo-periodic signal - called chirp - modulated in frequency around a carrier frequency and modulated in amplitude by an envelope whose variations are slow compared to the phase oscillations.

[0011] This modulation, which is particularly robust, has the disadvantage that the demodulation is complex and has a high cost in terms of calculation time and electrical energy consumption, which a priori limits its interest for an intermediate device.

[0012] The present invention improves the situation.

[0013] In this respect, the invention relates to a device for controlling an electrical or electronic device comprising: - a microphone arranged to convert an acoustic signal into an electrical signal, - a processing circuit arranged to detect a sinusoidal ultrasonic component within an acoustic signal from the corresponding electrical signal produced by the microphone, - a linear frequency modulation chirp demodulator arranged to operate in a standby mode and a demodulation mode, which demodulator is arranged, in the demodulation mode, to detect a linear frequency modulation chirp ultrasonic component within an acoustic signal from the corresponding electrical signal produced by the microphone and, if appropriate, to extract therefrom an on or off instruction, and - a communication module arranged to transmit a control signal representative of an on or off instruction extracted by the demodulator to the electrical or electronic device.

[0014] The device is configured such that the demodulator switches from standby mode to demodulation mode in response to the processing circuit detecting a sinusoidal ultrasonic component within an acoustic signal.

[0015] In one or more embodiments, the device further comprises a bandpass filter configured to have a bandwidth within the ultrasound frequency range.

[0016] The bandpass filter is for example an active biquadratic filter.

[0017] In one or more embodiments, the processing circuit comprises an envelope detector arranged to deliver, from an electrical signal produced by the microphone, an electrical voltage representative of an envelope of the electrical signal.

[0018] In one or more embodiments, the processing circuit further comprises a hysteresis comparator arranged to receive an electrical voltage delivered by the envelope detector and to deliver an output electrical voltage whose value is substantially equal to a first value if the value of the received electrical voltage is greater than or equal to the value of a high voltage or to a second value if the value of the received electrical voltage is less than or equal to the value of a low voltage. The first value and the second value respectively reflect the presence and absence of a sinusoidal ultrasonic component.

[0019] In one or more embodiments, the demodulator comprises a correlator arranged to perform a first cross-correlation between an electrical signal produced by the microphone and a reference turn-on signal and a second cross-correlation between the electrical signal and a reference turn-off signal.

[0020] Advantageously, the correlator is arranged to perform the first cross-correlation and the second cross-correlation in the frequency domain using a fast Fourier transform.

[0021] In one or more embodiments, the demodulator further comprises an envelope extractor arranged to apply a Teager-Kaiser operator to a first cross-correlation and a second cross-correlation performed by the correlator.

[0022] In one or more embodiments, the demodulator further comprises a comparator arranged to determine the respective maximum of a first cross-correlation and a second cross-correlation performed by the correlator, to compare the determined maxima, to detect the presence of a linear frequency modulated chirp ultrasound component on the basis of the comparison and, if appropriate, to extract therefrom an on or off instruction.

[0023] The invention also relates to a control system for an electrical or electronic device comprising: - a transmitter arranged to successively emit an acoustic signal comprising a sinusoidal ultrasonic component and an acoustic signal comprising a chirp ultrasonic component with linear frequency modulation, and - a device for controlling the electrical or electronic device as described above.

[0024] The invention also relates to a method for processing an acoustic signal implemented by the device described above. The method comprises the operations following: - convert, by the microphone, the acoustic signal into an electrical signal, - detect, by the processing circuit, a possible sinusoidal ultrasonic component within the acoustic signal from the electrical signal produced by the microphone, and, if applicable, - switch the demodulator from standby mode to demodulation mode.

[0025] The method further comprises, when the demodulator is in demodulation mode, the following operations: - detect, by the demodulator, a possible linear frequency modulated chirp ultrasonic component within the acoustic signal from the electrical signal produced by the microphone and extract an on or off instruction, and, if applicable, - transmit, via the communication module, a control signal representing the switching on or off instruction to the electrical or electronic device.

[0026] Finally, the invention further relates to a computer program comprising instructions whose execution, by at least one processor, results in the implementation of the method described above.

[0027] Other characteristics, details and advantages will appear on reading the detailed description below, and on analyzing the attached drawings in which:

[0028] [Fig.l] illustrates a control system for an electrical or electronic device according to the invention;

[0029] [Fig.2] illustrates a succession of acoustic signals emitted by a transmitter of the system of [Fig.l];

[0030] [Fig.3] illustrates in detail a control device for an electrical or electronic device electronics of the system of [Fig.l]; and

[0031] [Fig.4] illustrates a method of processing an acoustic signal according to the invention.

[0032] [Fig. 1] schematically illustrates a system 1 arranged to control an electrical or electronic device 3, which is referred to as device 3 in the remainder of the description for the sake of brevity. More precisely, the system 1 allows a user to turn the device 3 on or off remotely.

[0033] To do this, the system 1 comprises a transmitter 5 and a device 7 for controlling an electrical or electronic device - here the device 3.

[0034] An object of the present invention is to propose an architecture allowing a user, equipped with the transmitter 5, to turn on or off the device 3 remotely by means of the device 7, which plays both the role of receiver within the system 1 and of intermediary between the transmitter 5 and the device 3.

[0035] The device 7 has the particularity that it is capable of receiving signals acoustic signals from the transmitter 5 and consequently generate a control signal intended for the device 3, which is located nearby.

[0036] Such an architecture is advantageously applicable to electrical or electronic devices for which it is not usually possible to remotely control the switching on or off, and in particular those of these devices which are most often in standby mode (also known by the English expression "sleep mode"). It is indeed possible to achieve significant savings in electrical energy on such devices by switching them off as often as possible.

[0037] Furthermore, the use of elastic waves - namely here acoustic signals which are sound waves - makes it possible to avoid any interference with radiofrequency waves which are very present in a domestic or industrial context. Such radiofrequency waves are for example induced by third, fourth or fifth generation mobile telephone networks (3G, 4G, 5G) or a Wi-Fi network.

[0038] The transmitter 5 is arranged to emit acoustic signals to the device 7.

[0039] In the context of the invention, the transmitter 5 is arranged to first emit an acoustic signal comprising a sinusoidal ultrasonic component and then an acoustic signal comprising a chirp ultrasonic component with linear frequency modulation.

[0040] The sinusoidal ultrasonic component Ai typically takes the following form at time t: [0041 ] = Asin^2^ /

[0042] where: - A is the amplitude of the sinusoidal ultrasonic component xi, - fj is the frequency of the sinusoidal ultrasonic component ri, and - is the phase at the origin of the sinusoidal ultrasonic component %i.

[0043] The linear frequency modulated chirp ultrasound component %2 typically takes the following form at time t:

[0044] x2( / ) =Asin(^«))

[0045] The instantaneous phase 97 is expressed as follows at time t:

[0046] = ±^m2 + / 0

[0047] where: - B is the width of the frequency band swept by the linear frequency modulated chirp ultrasound component - T is the duration of the linear frequency modulated chirp ultrasound component a 2, and - / 0 is the start frequency of the sweep.

[0048] The starting frequency / 0 of the sweep generally corresponds to the center of the band swept frequency band. The frequency band swept by the linear frequency modulated chirp ultrasonic component xz then corresponds to the range [y B. y , B Ho 2'^o 2.

[0049] The instantaneous phase symbol reflects the fact that the linear frequency modulated chirp ultrasound component xz can be rising (+) or falling (-), which allows for two distinct and pseudo-orthogonal patterns. These two possible patterns correspond respectively to an instruction to turn on and off the device 3.

[0050] It should be noted that the acoustic signal comprising a sinusoidal ultrasonic component xi and the acoustic signal comprising a chirp ultrasonic component with linear frequency modulation xz can be separated by a delay A.

[0051] This succession of acoustic signals emitted by the transmitter 5 over time is illustrated in [Fig.2].

[0052] It can be considered equivalently that the transmitter 5 is arranged to emit an acoustic signal x comprising a sinusoidal ultrasonic component xi, a possible delay A and a chirp ultrasonic component with linear frequency modulation xz. The acoustic signal x can then be expressed as follows at a time t:

[0053] ( t ) if t E t J x(t) = < Osite ,x2(0

[0054] where: Ïq, ^i, O, and £3 are consecutive instants. In particular, the sinusoidal ultrasonic component is emitted between A) and ^1, a delay A elapses between I] and ^2 and the chirp ultrasonic component with linear frequency modulation x2 is emitted between ?2 and ^3. In the absence of delay A, we have: — f2-

[0055] Since the transmitter 5 is arranged to communicate by ultrasound with the device 7, it must be understood here that the frequency of the sinusoidal ultrasound component xi and at least part of the frequency band of width B swept by the linear frequency modulated chirp ultrasound component xz are included in the ultrasound frequency range, that is to say frequencies greater than or equal to 16 kilohertz (kHz).

[0056] Typically, the frequency range used is that of near ultrasound which extends from 16 to 30 kilohertz (kHz). It is advantageous to use more particularly frequencies between 16 and 25 kilohertz (kHz) since these correspond to the emission frequencies of transmitters existing on the market, in particular touch tablets or smart phones. For example, the frequency f} of the sinusoidal ultrasonic component xi can be located between 16 and 22 kilohertz (kHz) while the frequency band swept by the x2 linear frequency modulated chirp ultrasound component can cover frequencies between 17 and 20 kilohertz (kHz).

[0057] It should further be noted that a maximum frequency of 22 kilohertz (kHz) is less than half the conventional sampling frequency of 44.1 kilohertz (kHz), which allows it to be below the limit imposed by Shannon's theorem.

[0058] For the sake of brevity, the sinusoidal ultrasound component -L is referred to as the sinusoidal component in the remainder of the description, and the linear frequency modulated chirp ultrasound component x2 is referred to as the chirp component.

[0059] For the purposes of implementing, on the one hand, linear frequency modulation - or chirp modulation -, and, on the other hand, the emission of the acoustic signal, the transmitter 5 comprises, as illustrated in [Fig.l], a modulator 9 and a loudspeaker 11.

[0060] The modulator 9 is arranged to receive data representative of an ignition or extinction instruction and to generate an electrical signal, the reception of which causes the emission, by the loudspeaker 11, of an acoustic signal comprising a sinusoidal component then of an acoustic signal comprising a chirp component in accordance with the instruction received.

[0061] In other words, the chirp component has an instantaneous phase whose monomial has a sign corresponding to the instruction received. For example, the sign is positive (+) for an on instruction and negative (-) for an off instruction, or vice versa.

[0062] The loudspeaker 11 is arranged to receive an electrical signal generated by the modulator 9, convert it into a succession of acoustic signals and to emit these acoustic signals.

[0063] The loudspeaker has a bandwidth included in the ultrasonic frequency range such that it can transmit acoustic signals comprising respectively a sinusoidal component and a chirp component as described above.

[0064] Generally speaking, the loudspeaker 11 can be replaced by any electroacoustic transducer capable of emitting in the ultrasound frequency range.

[0065] It is understood from the above that the transmitter 5 can be formed by a dedicated remote control, a touch tablet, a smartphone, a portable or fixed computer, or, more generally, any device capable of sound diffusion.

[0066] The transmitter 5 allows a user, for example by means of a human-machine interface (also known by the English acronym HMI for “human-machine interface”), to turn on or off the device 3. Such a choice results in the generation of data representative of the switching on or off instruction which allows, as explained previously, the modulator 9 and the loudspeaker 11 to generate the appropriate acoustic signals.

[0067] The transmitter 5 therefore allows On / Off type control of the device 3.

[0068] The device 3 is typically an electrical or electronic device which has the following characteristics: particular that it is permanently plugged in but switched on occasionally or even rarely. Device 3 can be a domestic appliance such as a television, a computer screen, an air conditioner, a fan or even a roller shutter. Device 3 can also be an industrial machine.

[0069] It should be noted that the transmitter 5 can communicate with several devices such as the device 7, in which case the transmitter 5 offers the user the possibility of choosing the electrical or electronic device that he wishes to control remotely.

[0070] The device 7 is arranged to receive, as part of its role as receiver within the system 1, a succession of acoustic signals from the transmitter 5, to extract therefrom an ignition or extinction instruction and to emit, this time as part of its intermediary role, a control signal representative of the ignition or extinction instruction to the device 3.

[0071] The device 7 comprises a microphone 13, a bandpass filter 15, a signal duplication circuit 17, a processing circuit 19, a demodulator 21, a communication module 23, a memory 25 and a processor 27.

[0072] As detailed in the remainder of the description, the heart of the device 7 lies in the respective and combined operations of the processing circuit 19 and the demodulator 21. In particular, the processing circuit 19 and the demodulator 21 are configured to respectively detect a sinusoidal component and a chirp component within the received acoustic signals.

[0073] The microphone 13 is arranged to receive an acoustic signal and to convert it into an electrical signal. The microphone 13 is waiting for acoustic signals from the transmitter 5.

[0074] The bandpass filter 15 is configured to have a bandwidth included in the ultrasound frequency range.

[0075] The bandpass filter 15 is optional and makes it possible to eliminate at least the frequencies located outside the ultrasound frequency range.

[0076] The bandwidth of the bandpass filter 15 is typically centered on the frequency / 1 of the sinusoidal component and is advantageously as narrow as possible, which makes it possible to filter out any ambient noise. For example, the ends of the bandwidth of the bandpass filter 15 are each distant from the central frequency f { by a value of 3% of the latter. Furthermore, the bandwidth of the band-pass filter 15 may cover only part of the frequency band of width B swept by the chirp component. It is not necessary to recover the entire chirp component - 70%, or even 50%, of the chirp component is sufficient - to determine the instruction contained in it and it is preferable to limit the probability of a "false positive", that is to say the erroneous detection of a sinusoidal component.

[0077] The bandpass filter 15 may be an active biquadratic filter, which makes it possible to increase the power of the electrical signal and therefore to amplify it.

[0078] The signal duplication circuit 17 is arranged to duplicate the electrical signal at the output of the bandpass filter 15 or, in the absence thereof, directly at the output of the microphone 13, and to distribute each obtained version of the electrical signal respectively to the processing circuit 19 and to the demodulator 21.

[0079] The principle of circuit 17 is to allow, on the one hand, the processing circuit 19 to be able to proceed with the detection of the sinusoidal component and, on the other hand, the demodulator 21 to be able, if necessary, to proceed subsequently with the detection of the chirp component.

[0080] The circuit 17 can be implemented by mounting two operational amplifiers (often abbreviated AOP in the literature) as a follower. Each of these operational amplifiers then behaves, respectively for the processing circuit 19 and for the demodulator 21, as a voltage source.

[0081] In the example of [Fig.l], the bandpass filter 15 and the circuit 17 are distinct from the processing circuit 19. However, the bandpass filter 15 and the circuit 17 can be integrated into the processing circuit 19.

[0082] The processing circuit 19 is arranged to detect the presence of a sinusoidal component within an acoustic signal from the corresponding electrical signal produced by the microphone 13.

[0083] The processing circuit 19 is in operation most of the time, or even constantly. To do this, the processing circuit 19 can be supplied with electrical energy by a 1.8 volt (V) power source.

[0084] Referring now to [Fig. 3], the processing circuit 19 comprises an envelope detector 29 and a hysteresis comparator 31.

[0085] The envelope detector 29 is arranged to deliver, from one of the electrical signals distributed by the circuit 17, an electrical voltage representative of an envelope of the electrical signal.

[0086] The envelope detector 29 makes it possible to extract the successive peak values ​​of the electrical signal, which are interpolated to reconstruct the envelope of the electrical signal.

[0087] Typically, the envelope detector 29 is formed of a diode and a capacitor connected in series. Thus, when the voltage of the received electrical signal is positive, the diode is conductive and the capacitor charges. Conversely, when the voltage of the received electrical signal is negative, the diode is blocking and the capacitor discharges. It is possible to couple a resistor to the capacitor to form a charge so that the capacitor charges faster than it discharges.

[0088] The use of the envelope detector 29 is particularly appropriate here since the sinusoidal component has, by nature, an envelope that is easy to detect. Conversely, the chirp component, which results from a linear frequency modulation - or chirp modulation - and not from an amplitude modulation, is not likely to disturb the envelope detector 29.

[0089] The detection of a possible sinusoidal component within a received acoustic signal is carried out, strictly speaking, by the hysteresis comparator 31.

[0090] A hysteresis comparator is an operator whose output potential difference can take two stable and distinct values ​​depending on the input electrical voltage and a reference electrical voltage.

[0091] In the present case, the hysteresis comparator 31 is arranged to receive the electrical voltage delivered by the envelope detector 29 as an input electrical voltage. In other words, the envelope detector 29 is arranged to apply an electrical voltage to the input of the hysteresis comparator 31.

[0092] The reference electrical voltage makes it possible to define, within the hysteresis comparator 31, a high voltage and a low voltage. More particularly, the respective values ​​of the high voltage and the low voltage are distant from each other by a predefined step, and the interval whose limits correspond to the respective values ​​of the high voltage and the low voltage is centered on the value of the reference electrical voltage. In other words, the configuration of the hysteresis comparator 31 depends on the value Vref of the reference electrical voltage and the predefined step AV which define a value Vref + to V / 2 of high voltage and a value Vref- to V / 2 of low voltage.

[0093] The hysteresis comparator 31 is arranged to deliver an output electrical voltage whose value is substantially equal to a first value if the input electrical voltage is greater than or equal to the high voltage, or to a second value if the input electrical voltage is less than or equal to the low voltage. The first value reflects the presence of a sinusoidal component, while the second value reflects the absence of a sinusoidal component.

[0094] The value V ref of the reference electrical voltage and the predefined step AV can be defined experimentally.

[0095] The hysteresis comparator 31 may be configured differently and, for example, directly receive two reference electrical voltages which then correspond respectively to the high voltage and the low voltage. The respective values ​​of the two reference electrical voltages can also be defined experimentally.

[0096] The hysteresis comparator 31, and more generally the processing circuit 19, in fact makes it possible to determine whether the acoustic signal received by the device 7 is likely to have been emitted by a transmitter such as the transmitter 5, and therefore to precede an acoustic signal comprising a chirp component. Indeed, as explained previously, the relevant information, namely the switching on or off instruction, is carried by the chirp component.

[0097] Unlike the processing circuit 19 which is intended to be in operation most of the time, the demodulator 21 is intended to be almost always in standby mode and to only switch to demodulation mode exceptionally, that is to say when the processing circuit 19 has detected the presence of a sinusoidal component within an acoustic signal received by the device 7 and the assistance of the demodulator 21 is required to determine whether the next acoustic signal includes a chirp component carrying an on or off instruction.

[0098] Detecting a chirp component is more computationally expensive than detecting a sinusoidal component. The components of the demodulator 21 consume more electrical energy than those of the processing circuit 19.

[0099] The device 7 obeys the following principle: the processing circuit 19 is configured to “wake up” the demodulator 21, that is to say to make it switch from standby mode - which is the default mode of the demodulator 21 - to demodulation mode when a sinusoidal component is detected and there is therefore a strong presumption of the presence of a chirp component in a subsequent acoustic signal.

[0100] In particular, the transition from standby mode to demodulation mode of the demodulator 21 can be achieved by supplying the latter with electrical energy, for example by means of a 1.8 volt (V) power source.

[0101] The demodulator 21 can be implemented on a 32-bit microcontroller of the STM32 family (registered trademark) marketed by the company STMicroelectronics. The microcontroller used is for example the STM32G4 model.

[0102] Still referring to [Fig.3], the demodulator 21 comprises an analog-digital converter 33, a correlator 35, an envelope extractor 37 and a comparator 39.

[0103] As explained previously, the electrical signal at the output of the band-pass filter 15 or, in the absence thereof, directly at the output of the microphone 13, is duplicated by the circuit 17: a first version is distributed to the envelope detector 29 and a second version is distributed to the demodulator 21.

[0104] This second version is received by the analog-to-digital converter 33.

[0105] The analog-to-digital converter 33 (also known by the English acronym ADC for “analog-to-digital converter”) is arranged to convert the electrical signal transmitted by the circuit 17 into a digital signal.

[0106] The correlator 35 is arranged to apply a first and a second cross-correlation to the electrical signal transmitted by the processing circuit 19, and more precisely here to the digital signal produced by the analog-digital converter 33.

[0107] It should be noted that the correlator 35 can access the analog-to-digital converter 33 and therefore the digital signal by means of a direct memory access method (better known by the English acronym DMA for “direct memory access”).

[0108] More particularly, the correlator 35 is arranged to carry out a first cross-correlation between the digital signal and a reference ignition signal and a second cross-correlation between the digital signal and a reference extinction signal.

[0109] Cross-correlation is used here to measure the similarity between the digital signal and, on the one hand, the reference ignition signal, and, on the other hand, the reference extinction signal. Such a measurement reveals the presence of similar patterns or structures.

[0110] The first cross-correlation and the second cross-correlation may be performed in the time domain. Alternatively, the first cross-correlation and the second cross-correlation are performed in the frequency domain using a fast Fourier transform (FFT), which reduces the time required for the calculations.

[0111] The output of the correlator 35 takes the form of a first curve and a second curve corresponding respectively to the first cross-correlation and to the second cross-correlation. Each curve has one or more peaks, which correspond to the instants at which a chirp component carrying an ignition instruction - for the first curve - or an extinction instruction - for the second curve - is most likely to be present.

[0112] The envelope extractor 37 is arranged to apply a Teager-Kaiser operator to a first cross-correlation and a second cross-correlation performed by the correlator 35.

[0113] More specifically, the Teager-Kaiser operator is applied to the first and second curves at the output of the correlator 35. The Teager-Kaiser operator makes it possible both to refine the peaks of these curves and to increase their respective amplitudes. The peaks can thus be detected more faithfully and more quickly, and the envelope of each cross-correlation can be extracted.

[0114] It should be noted that optimization techniques other than the Teager-Kaiser operator can be used to extract the envelope of the first and second cross-correlations.

[0115] The envelope extractor 37 is optional and makes it easier to determine the respective maximum of the first cross-correlation and the second cross-correlation.

[0116] The comparator 39 is arranged to calculate a ratio of cross-correlation maxima and compare it to a predetermined threshold.

[0117] More specifically, the comparator 39 is arranged to calculate the ratio of the maximum of the first cross-correlation and the maximum of the second cross-correlation, and to compare this first ratio to the predetermined threshold. A first ratio greater than or equal to the predetermined threshold is interpreted by the comparator 39 as an ignition instruction.

[0118] On the contrary, if the first ratio is lower than the predetermined threshold, the comparator 39 is arranged to calculate the inverse ratio, that is to say the ratio of the maximum of the second cross-correlation and the maximum of the first cross-correlation, and to compare this second ratio to the same predetermined threshold. A second ratio greater than or equal to the predetermined threshold is interpreted by the comparator 39 as an extinction instruction. On the other hand, a second ratio lower than the predetermined threshold is interpreted by the comparator 39 as an absence of chirp component.

[0119] The comparator 39 can also proceed in reverse, that is to say first calculate the second ratio and compare it to the predetermined threshold then, if necessary, calculate the first ratio to compare it in turn to the predetermined threshold. It is then in the case where the first ratio is lower than the predetermined threshold that the comparator 39 concludes that there is an absence of chirp component.

[0120] It is understood from the above that what the comparator 39, and more generally the demodulator 21, implements is similar to a maximum likelihood estimation.

[0121] The first predetermined threshold and the second predetermined threshold can be determined experimentally.

[0122] The comparator 39 may be configured differently. For example, the comparator 39 may retrieve the maximum of the first cross-correlation and the maximum of the second cross-correlation, select the highest of the two, and compare it to a predetermined threshold. If the selected maximum is greater than or equal to the predetermined threshold, the comparator 39 interprets the result of this comparison as revealing the presence of an ignition instruction in the case where the selected maximum is that of the first cross-correlation, or of an instruction extinction in the case where the selected maximum is that of the second cross-correlation. Conversely, if the selected maximum is lower than the predetermined threshold, the comparator 39 interprets the result of this comparison as an absence of chirp component. Here again, the predetermined threshold used can be determined experimentally.

[0123] In the case where a chirp component is detected within the acoustic signal and an instruction to switch on or off the device 3 is extracted therefrom, the demodulator 21, and more precisely here the comparator 39, transmits the instruction to the communication module 23.

[0124] The communication module 23 is arranged to transmit a control signal representative of an ignition or extinction instruction extracted by the demodulator 21 to the device 3.

[0125] Such a control signal is typically binary, with logic level 1 corresponding to an ignition instruction and logic level 0 corresponding to an extinction instruction, or vice versa.

[0126] As explained previously, the device 7 is intended to be installed near the apparatus 3.

[0127] Consequently, the communication module 23 can use a short-distance communication technology, wired or not. Among the wireless technologies, the communication module 23 can communicate with the device 3 in near field (better known by the English acronym NFC for “near-field communication”), by the ZigBee protocol (registered trademark), by Bluetooth (registered trademark) or even by low-energy Bluetooth (better known by the English acronym BLE for “Bluetooth Low Energy”). Still by way of example, the communication module 23 can communicate with the device 3 by means of an electromechanical relay.

[0128] The memory 25 is arranged to store instructions whose implementation, by the processor 27, results in the operation of the device 7.

[0129] The memory 25 may designate any data storage medium arranged to receive and store digital data, for example a hard disk, a solid-state drive (better known by the English acronym SSD for “solid-state drive”) or more generally any computer hardware allowing the storage of data on flash memory. The memory 25 may also be a random access memory or a magneto-optical disk. A combination of several storage media may also be envisaged.

[0130] The processor 27 can be produced in any known manner, for example in the form of a microprocessor, a programmable logic circuit (better known by the English acronym PLD for “Programmable Logical Device”) or a dedicated chip of the FPGA type (English acronym for “Field Programmable Gate Array”) or SoC. (an acronym for "System on Chip"), a grid of computing resources, a microcontroller or any other specific form having the computing power necessary to detect a sinusoidal component, to detect a chirp component, to extract switching on or off information as well as to generate and transmit a control signal. One or more of these elements can also be implemented in the form of specialized electronic circuits of the ASIC type (an acronym for "Application-Specific Integrated Circuit"). A combination of processors and electronic circuits can also be considered.

[0131] A method for processing an acoustic signal implemented by the device 7 will now be described with reference to [Fig.4].

[0132] In the context of the implementation of this method, the device 7 receives an acoustic signal. At this stage, the device 7 is unaware of the origin and content of the acoustic signal. The acoustic signal may have been sent by the transmitter 5, in which case it may be an acoustic signal comprising a sinusoidal component or an acoustic signal comprising a chirp component. However, the acoustic signal may also be a parasitic signal, for example ambient noise.

[0133] The method aims to determine whether the received acoustic signal comprises a sinusoidal component and, if so, to switch the demodulator 21 from standby mode to demodulation mode. The method also aims to determine, when the demodulator 21 is in demodulation mode, whether the received acoustic signal comprises a chirp component and, if so, to extract therefrom an on or off instruction to control the device 3 accordingly.

[0134] During an operation 400, the microphone 13 converts the received acoustic signal into an electrical signal.

[0135] During an optional operation 410, the bandpass filter 15 filters at least the frequencies located outside the ultrasound frequency range.

[0136] As indicated above, the filtering can be finer; and the bandwidth of the band-pass filter 15 is advantageously as narrow as possible centered on the frequency / 1 of the expected sinusoidal component and comprises at least part of the frequency band of width B swept by the expected chirp component.

[0137] Furthermore, if the bandpass filter 15 is an active biquadratic filter, the electrical signal is amplified.

[0138] During an operation 420, the possibly filtered and amplified electrical signal is received by the circuit 17, which duplicates the electrical signal to generate a first and a second version.

[0139] The first version of the electrical signal is distributed to the processing circuit 19 to determine whether a sinusoidal component is present within the signal. acoustic. The second version of the electrical signal is distributed to the demodulator 21 to determine whether a chirp component is present within the acoustic signal.

[0140] We first consider the processing applied to the first version of the electrical signal.

[0141] During an operation 430, the first version of the electrical signal is received by the envelope detector 29.

[0142] The envelope detector 29 delivers an electrical voltage representative of an envelope of the electrical signal. Typically, the envelope detector 29 extracts the successive peak values ​​of the electrical signal and reconstructs the envelope of the electrical signal by interpolation.

[0143] During an operation 440, the electrical voltage delivered by the envelope detector 29 is applied to the input of the hysteresis comparator 31.

[0144] As detailed above, the operation of the hysteresis comparator 31 depends on a high voltage and a low voltage, which can be defined by a reference electrical voltage and a reference step or, alternatively, by two reference electrical voltages corresponding directly to the high voltage and the low voltage.

[0145] The hysteresis comparator 31 delivers an output electrical voltage whose value can take two possible values.

[0146] If the value of the input electrical voltage is greater than or equal to the value of the high voltage, the output electrical voltage takes a first value which reflects the presence of a sinusoidal component. Conversely, if the value of the input electrical voltage is less than or equal to the value of the low voltage, the output electrical voltage takes a second value which reflects the absence of a sinusoidal component.

[0147] The value of the input electrical voltage may be between the value of the low voltage and the value of the high voltage, in which case the value of the output electrical voltage remains unchanged.

[0148] Typically, at the instant preceding the application, by the envelope detector 29, of an electrical voltage to the hysteresis comparator 31, the input electrical voltage is zero and therefore lower than the low voltage. The output electrical voltage therefore takes the second value by default, and this as long as the input electrical voltage is lower than the high voltage, including, therefore, if the input electrical voltage is between the low voltage and the high voltage. The operation is the same in the case where the input electrical voltage is higher than the high voltage and then falls below the latter: the value of the output electrical voltage remains equal to the first value as long as the input electrical voltage is higher than the low voltage, including, therefore, if the input electrical voltage is between the low voltage and high voltage.

[0149] In the absence of a sinusoidal component within the acoustic signal (“KO” arrow at the output of operation 440), the value of the output electrical voltage is equal to the second value and nothing happens.

[0150] Conversely, in the presence of a sinusoidal component within the acoustic signal (“OK” arrow at the output of operation 440), the value of the output electrical voltage is equal to the first value and causes the demodulator 21 to switch from standby mode to demodulation mode during an operation 450.

[0151] As indicated above, the transition from standby mode to demodulation mode of the demodulator 21 can be achieved by means of a power source. For example, the output of the comparator 31 can be connected to such a power source so that an electrical output voltage value of the hysteresis comparator 31 equal to the first value triggers the power supply, by the voltage source, of the demodulator 21.

[0152] The “waking up” of the demodulator 21 allows it to be able to demodulate the next acoustic signal received by the device 7. Indeed, such an acoustic signal is likely to include a chirp component.

[0153] We now consider the demodulation applied to the second version of the electrical signal.

[0154] Demodulation is only implemented by the demodulator 21 if the latter is in demodulation mode (condition “WK?” at the output of operation 420).

[0155] When the demodulator 21 is in standby mode (“KO” arrow at the output of the “WK?” condition), the second version of the electrical signal is not received by the demodulator 21 and no demodulation is implemented.

[0156] On the other hand, when the demodulator 21 is in demodulation mode (“OK” arrow at the output of the “WK?” condition), the second version of the electrical signal is received by the demodulator 21 and the demodulation is implemented. This demodulation is described below.

[0157] During an operation 460, the analog-to-digital converter 33 converts the electrical signal into a digital signal.

[0158] During an operation 470, the correlator 35 performs a first cross-correlation between the digital signal and a reference ignition signal as well as a second cross-correlation between the digital signal and a reference extinction signal.

[0159] The correlator 35 thus measures the similarity between the digital signal and, on the one hand, the reference ignition signal, and, on the other hand, the reference extinction signal.

[0160] In an optional operation 480, the envelope extractor 37 applies a Teager-Kaiser operator to the first cross-correlation and the second cross-correlation.

[0161] The Teager-Kaiser operator has the effect of refining the peaks of the curves generated by the correlator 35 while increasing its amplitude, which makes it easier to detect peaks.

[0162] During an operation 490, the comparator 39 determines the maximum of the first cross-correlation and the maximum of the second cross-correlation.

[0163] When the envelope extractor 37 is used, we can speak of the maximum of the envelope of the first cross-correlation and the maximum of the envelope of the second cross-correlation.

[0164] The comparator 39 then compares the two determined maxima.

[0165] For example, comparator 39 calculates the ratio of the maximum of the first cor cross-relationship and the maximum of the second cross-correlation, then compares this first ratio to a predetermined threshold. If the first ratio is greater than or equal to the predetermined threshold, the comparator 39 concludes that there is a chirp component carrying an ignition instruction within the acoustic signal.

[0166] If the first ratio is lower than the predetermined threshold, the comparator 39 then calculates the inverse ratio, i.e. the ratio of the maximum of the second cross-correlation and the maximum of the first cross-correlation, then compares this second ratio to the predetermined threshold. If the second ratio is greater than or equal to the predetermined threshold, the comparator 39 concludes that there is a chirp component carrying an extinction instruction within the acoustic signal.

[0167] On the other hand, if the second ratio is also lower than the predetermined threshold, the comparator 39 concludes that there is no chirp component within the acoustic signal.

[0168] According to the variant described above, the comparator 39 can also select the one of the two maxima - among the maximum of the first cross-correlation and the maximum of the second cross-correlation - which is the highest and compare it to a predetermined threshold.

[0169] If the selected maximum is lower than the predetermined threshold, the comparator 39 concludes that no chirp component is present within the acoustic signal.

[0170] Conversely, if the selected maximum is greater than or equal to the predetermined threshold, the comparator 39 concludes that a chirp component is present. In particular, the comparator 39 extracts an ignition instruction if the selected maximum is the maximum of the first cross-correlation; and the comparator 39 extracts an extinction instruction if the selected maximum is the maximum of the second cross-correlation.

[0171] In the absence of a chirp component (“KO” arrow at the output of operation 490), the method is interrupted and the demodulator 21 switches from demodulation mode to standby mode during an operation 4100.

[0172] In the presence of a chirp component (“OK” arrow at the output of operation 490), the demodulator 21, and more precisely here comparator 39, transmits the extracted switching on or off instruction to the communication module 23.

[0173] Once the on or off instruction has been transmitted to the communication module 23, the demodulator 21 can switch from demodulation mode to standby mode.

[0174] Finally, during an operation 4110, the communication module 23 transmits a control signal to the device 3.

[0175] The control signal is representative of the instruction extracted by the demodulator 21, i.e. the switching on or off of the device 3.

[0176] The apparatus 3, which is typically located near the device 7, receives the control signal and is then switched off or on according to the instruction corresponding to the control signal.

[0177] It is understood from the above that, when the transmitter 5 emits an acoustic signal comprising a sinusoidal component and then an acoustic signal comprising a chirp component, the method of [Fig.4] is implemented twice to control the device 3.

[0178] Initially, the acoustic signal comprising the sinusoidal component is received by the device 7. The second version of the electrical signal produced by the microphone 13 is not demodulated by the demodulator 21 since the latter is a priori in standby mode. The first version of the electrical signal is processed by the processing circuit 19. The detection of the sinusoidal component within the acoustic signal then switches the demodulator 21 from standby mode to demodulation mode.

[0179] In a second step, the acoustic signal comprising the chirp component is received by the device 7. The first version of the electrical signal produced by the microphone 13 is, of course, processed by the processing circuit 19, but this does not result in anything in particular. The second version of the electrical signal is, itself, demodulated by the demodulator 21 since the latter is in demodulation mode. The detection of the chirp component within the acoustic signal and the extraction of the switching on or off instruction triggers the transmission, by the communication module 23, of a control signal representative of the instruction extracted by the demodulator 21 to the device 3.

Claims

Claims

1. Device (7) for controlling an electrical or electronic apparatus (3) comprising: - a microphone (13) arranged to convert an acoustic signal into an electrical signal, - a processing circuit (19) arranged to detect a sinusoidal ultrasonic component within an acoustic signal from the corresponding electrical signal produced by the microphone (13), - a linear frequency modulation chirp demodulator (21) arranged to operate in a standby mode and a demodulation mode, which demodulator (21) is arranged, in the demodulation mode, to detect a linear frequency modulation chirp ultrasonic component within an acoustic signal from the corresponding electrical signal produced by the microphone (13) and, where appropriate, to extract therefrom an on or off instruction,and - a communication module (23) arranged to transmit a control signal representative of an on or off instruction extracted by the demodulator (21) to the electrical or electronic device (3), the device (7) being configured such that the demodulator (21) switches from standby mode to demodulation mode in response to the detection by the processing circuit (19) of a sinusoidal ultrasonic component within an acoustic signal.,

2. Device (7) according to claim 1, further comprising a bandpass filter (15) configured to have a bandwidth included in the ultrasound frequency range.

3. Device (7) according to claim 2, wherein the bandpass filter (15) is an active biquadratic filter.

4. Device (7) according to one of the preceding claims, in which the processing circuit (19) comprises an envelope detector (29) arranged to deliver, from an electrical signal produced by the microphone (13), an electrical voltage representative of an envelope of said electrical signal.

5. Device according to claim 4, in which the processing circuit (19) further comprises a hysteresis comparator (31) arranged to receive an electrical voltage delivered by the envelope detector (29) and to deliver an output electrical voltage whose value is substantially equal to a first value if the value of the received electrical voltage is greater than or equal to the value of a high voltage or to a second value if the value of the received electrical voltage is less than or equal to the value of a low voltage, said first value and said second value respectively reflecting the presence and absence of a sinusoidal ultrasonic component.

6. Device (7) according to one of the preceding claims, in which the demodulator (21) comprises a correlator (35) arranged to carry out a first cross-correlation between an electrical signal produced by the microphone (13) and a reference ignition signal and a second cross-correlation between said electrical signal and a reference extinction signal.

7. A device (7) according to claim 6, wherein the correlator (35) is arranged to perform the first cross-correlation and the second cross-correlation in the frequency domain using a fast Fourier transform.

8. Device (7) according to claim 6 or 7, wherein the demodulator (21) further comprises an envelope extractor (37) arranged to apply a Teager-Kaiser operator to a first cross-correlation and a second cross-correlation carried out by the correlator (35).

9. Device (7) according to one of claims 6 to 8, wherein the demodulator (21) further comprises a comparator (39) arranged to determine the respective maximum of a first cross-correlation and a second cross-correlation carried out by the correlator (35), to compare the determined maxima, to detect the presence of a linear frequency modulated chirp ultrasonic component on the basis of said comparison and, where appropriate, to extract therefrom an on or off instruction.

10. System (1) for controlling an electrical or electronic device (3) comprising: - a transmitter (5) arranged to successively emit an acoustic signal comprising a sinusoidal ultrasonic component and an acoustic signal comprising a chirp ultrasonic component with linear frequency modulation, and - a device (7) for controlling said electrical or electronic device (3) according to one of the preceding claims.

11. Method for processing an acoustic signal implemented by the device (7) according to one of claims 1 to 9, said method comprising the following operations: - converting (400), by the microphone (13), said acoustic signal into an electrical signal, - detecting (430, 440), by the processing circuit (19), a possible sinusoidal ultrasonic component within the acoustic signal from said electrical signal produced by the microphone (13), and, if necessary, - switching (450) the demodulator (21) from standby mode to demodulation mode, the method further comprising, when the demodulator (21) is in demodulation mode, the following operations: - detecting (470, 480, 490), by the demodulator (21), a possible linear frequency modulated chirp ultrasonic component within the acoustic signal from said electrical signal produced by the microphone (13) and extracting therefrom an on or off instruction, and, where appropriate, - transmit (4110), via the communication module (23), a control signal representing the ignition or extinction instruction to the electrical or electronic device (3).

12. Computer program comprising instructions whose execution, by at least one processor (27), results in the implementation of the method according to claim 11.

Images