WO2008135672A1

WO2008135672A1 - Hands-free car adaptor for mobile telephones with built-in fm transmitter, in which incoming requests are managed according to the frequency matching status in relation to the frequency of the associated radio set

Info

Publication number: WO2008135672A1
Application number: PCT/FR2008/000401
Authority: WO
Inventors: Guillaume Pinto
Original assignee: Parrot
Priority date: 2007-03-28
Filing date: 2008-03-26
Publication date: 2008-11-13
Also published as: FR2914526A1

Abstract

According to the invention, the device (10) relays signals from an audio source to a radio set, such as a car radio. The device comprises: means for producing a reference signal, means enabling the radio signal to be modulated by the reference signal or by the audio signals received by coupling means, a microphone (20) for capturing acoustic signals present in the environment of the device, and means for analysing the acoustic signals captured by the microphone. Upon receipt of a request, for example an incoming call, the device executes an algorithm for detection of a known signal in a noisy environment, in order to determine the presence or absence of the reference signal in the acoustic signals captured by the microphone. When the reference signal is present in the captured acoustic signals the device transmits a return acceptance signal and when the reference signal is absent from said acoustic signals the device activates an audible (22) and/or visible (30) alarm intended for the user, such that the user can react and perform an action enabling transmission frequency matching to be re-established, for example by adjusting a turn button (28) or pressing a push button (32) for frequency preselection.

Description

Car hands-free adapter for mobile phone with built-in FM transmitter, with incident request management according to the state of concordance with the frequency of the associated radio receiver

The invention relates to an interfacing device (hereinafter "adapter") for coupling, on the one hand, a source of audio signals with, on the other hand, a radio receiver.

The invention applies very advantageously to the coupling of a mobile phone to a car radio of a motor vehicle. This application is however not limiting of the invention, which can be used as well (i) with other types of audio sources, for example with an MP3 player or any other analog or digital source, (ii) with d other types of receivers than a car radio, for example a portable receiver or a stereo, and (iii) in other contexts than the passenger compartment of a motor vehicle, for example at home.

However, both the ergonomics and performance of the device of the invention are particularly suitable for use on board a moving vehicle, to ensure a "hands-free" operation of a mobile phone. If we take the particular example of a hands-free adapter that it is desired to couple to a car radio to allow the reproduction of the voice through the speakers of the vehicle, there are essentially two types of adapters. The first type includes permanently integrated adapters to the vehicle, original equipment or later. The adapter is in the form of a housing, generally inaccessible to the user, which is connected to a microphone and the car radio directly by a wiring harness. The wired connection then makes it possible to send control signals to the car radio to put the radio / audio operation on standby in the event of telephone communication, and relay the conversation to the loudspeakers of the vehicle.

The second type corresponds to an adapter that does not require permanent installation in the vehicle. The main advantage of this adapter is that it requires no connection or installation and can therefore be installed immediately, and by the user himself. Otherwise, being removable, the same adapter can be used on several different vehicles of the same user.

Due to the absence of any cabling between the adapter and the car radio, the coupling is effected by a low power FM radio transmitter built into the adapter, this radio signal being picked up by the car radio and reproduced on the radio. the speakers, in the same way as any station in the FM band.

It is of course necessary that the transmission frequency of the adapter and the frequency of reception of the car radio are identical. EP-A-0 920 170 A2 describes such an adapter in the form of a plug-in housing on the connector of a mobile phone and comprising an FM transmitter modulated by the signal from the mobile phone. The tuning of the frequencies with the car radio is performed by automatic scanning of the FM frequency, the carrier being modulated by a brief repetitive pulse. The adapter is also equipped with a "high frequency" microphone capable of sensing the presence or absence of the impulse in the sound environment of the passenger compartment of the vehicle. When the frequencies of the adapter and car radio match, then the pulse is recognized by the microphone, and the adapter stops scanning the FM frequency and locks to the frequency at which it is then flown. .

The principle of using an FM transmitter to couple to a car radio has the great advantage of not requiring any physical connection between the two devices and of ensuring a very good quality of sound reproduction, but suffers from a constraining disadvantage of the radio. makes the need for a frequency agreement between the two devices. To be able to use his phone (or other device) in optimal conditions, the user will choose on the car radio a frequency not occupied by a local radio station. On the other hand, it is unlikely that the user will condone the use of his car radio when the phone is not in use; the frequency of the car radio will then most often set to demodulate a radio station. In this way, in practice, the transmission frequency of the adapter and the frequency of reception of the car radio will be mostly different, and frequently modified. The technique described by EP-AO 920 170, previously implemented by the user, does not take into account the behavior of the user who frequently changes the settings of the car radio. With the device described in this document, once the calibration step has been carried out, the adapter remains locked on a given transmission frequency, and for the device to remain operational the tuning of the car radio must not be amended. In the opposite case - frequent in practice as just indicated - the adapter will transmit at the frequency on which it was locked and no signal will be restored by the car radio due to the change of setting. And the user will have no way to be warned why his car radio is silent and the speakers do not return the phone conversation. Another difficulty that does not take into account the system proposed by EP 0 920 170 is the particularly noisy environment of a moving motor vehicle.

If the calibration of the adapter described in this document is performed while stopped, the noise level will be relatively moderate and the microphone will be able to detect the presence of the pulse in the audio signal returned by the loudspeakers. This simple technique will become difficult or impossible to implement in the case of a moving vehicle, with an acoustic environment having both a high relative noise level, making it difficult to extract the useful signal embedded in the noise, and non-stationary spectral characteristics, that is to say, which evolve unpredictably according to the driving conditions (passage on deformed or paved roads, crossing of noisy vehicles, etc.). In addition, the closed and confined volume of a car interior generates a multitude of complex secondary propagations which are superimposed and strongly scramble the signal picked up by the microphone (multiple echoes phenomenon). The US 2004/204158 A1 describes an adapter operating on the same principle, that is to say with an automatic and prior scanning of the frequency of the adapter, and subject to the same disadvantages, namely the lack of robustness to the noise, poor ergonomics and the fact that the signal relayed by the adapter is superimposed on the signal already picked up by the radio on the frequency found by the adapter. Various other "hands-free" car phone adapters have been proposed, which furthermore provide wireless coupling between the telephone and the adapter via a Bluetooth-type connection, as is explained in particular by US 2007/0049197 A1 and US 2006/0094355 A1 (with an adapter / radio coupling by FM transmission), or by the US2002 / 0197954 A1 (with an adapter / car radio connection via a physical connection).

In general, none of the adapters described above provides an intuitive interface for reacting in a simple way to a request (for example an incoming telephone call, or the reading of an audio stream) involving a reaction of the user - typically, the acceptance or denial (explicit or implicit) of the incoming call. It should be noted that the invention is placed - unlike the techniques described in the aforementioned documents - in a context where (i) the adapter is tuned to a given frequency, where (ii) the car radio is also , tuned to a given frequency, but where (iii) this latter frequency is not necessarily the same as that of the adapter.

The reproduction chain (source -> adapter -> car radio) will therefore be operative, or inoperative, depending on whether or not the tuning frequencies of the adapter and the car radio match.

One of the aims of the invention is to have an adapter capable of an "intelligent" reaction in the event of an incidental request involving a reaction from the user (typically, the acceptance or refusal of an incoming call ), with or without matching of the transmission frequency of the adapter with the reception frequency of the car radio, and appropriate management of the situation as the case may be.

It should be emphasized in this regard that the context is that of the automotive environment, where the user is very solicited, and the time of concentration he can devote to the control of the adapter is very small. It is therefore necessary to provide an adapter capable of providing assistance that is as simple and intuitive as possible. The adapter of the invention is an interface device of the type disclosed in the aforementioned EP-AO 920 170, that is to say a device for relaying the audio signals from a source to a radio receiver for restitution by the means of acoustic reproduction of the latter, and comprising: means for coupling to the source of audio signals,

means transmitting a radio signal, tuned to a given radio transmission frequency,

generating means for producing a reference signal; means for modulating said radio signal by said reference signal or by the audio signals received by the coupling means;

a microphone, for picking up the acoustic signals present in the environment of the device, and

means for analyzing the acoustic signals picked up by the microphone.

The device of the invention is characterized in that:

said coupling means are bidirectional coupling means capable of receiving from the source a request for activation of the interfacing device and to issue in response a signal of acceptance, or not, of this request, and

- It includes control means capable of recognizing an action, acceptance or refusal, performed voluntarily by the user to the device and also able, upon receipt of said request, to perform the following operations: a) by said generating means, production of the reference signal; b) by the analysis means, execution of an algorithm for detecting a known signal in a noisy environment, able to recognize the presence or absence of the reference signal in the acoustic signals picked up by the microphone; c) if so, sending said acceptance signal to the source of audio signals, this emission being conditioned to the recognition of an acceptance action performed by the user voluntarily; d) if not, activation of a buzzer and / or visual signal for the user, and reiteration of step b).

In other words, when the request is sent (for example when the radiotelephone detects a cross-call), two things one: there is, or not, agreement between the tuning frequency of the adapter and that of of the car radio. In the first case (there is concordance), the device must first verify that this match is effective, and this verification must be performed by a process having a high robustness to noise. If the match is confirmed, the adapter authorizes the continuation of the process and then takes two actions:

1 °) it signals to the user the appearance of the request, and 2 °) it is waiting for the detection of a voluntary action of the user, who can accept or refuse the incidental call (in the case, for example, where the request is an on-call stall request). Note that in case of refusal, it can be explicit (pressing a button, voice response, etc.) or implicit, letting a timer run to completion.

The second case is where there is no match between the tuner frequency of the adapter and the tuning frequency of the car radio. The reproduction chain is therefore inoperative, and the adapter then takes two actions:

1 °) it warns the user of this situation, by a sound and / or visual feedback, so that he can react appropriately, and 2 °) he reiterates the process of checking concordance. The user, well informed, may take the initiative to change the frequency of the adapter or that of the car radio. There is thus initiation of a feedback loop involving both the adapter (automatic and conditional reaction) and the user (who has, alone, the initiative to change settings). This looping continues until the user realizes the frequency matching by changing the settings - which means that he accepts, implicitly, the request - or by running the timer until its end. Unlike known adapters which provide automatic frequency scanning, in the case of the present invention it is up to the user to change the settings in case of frequency mismatch; most often it will be enough to change the channel, by switching to a preset of the car radio corresponding to a frequency where no radio program is broadcast, and which it has chosen to match the frequency, fixed, okay of the adapter. This procedure has the double advantage:

- to always inform the user of the appearance of a request; and

- to then allow him, by a single and simple action on his part, to indicate his desire to establish the communication. In general, when the audio signal source is a radiotelephone, the request is sent on detection by the radiotelephone of an incoming call. In this case, according to various advantageous characteristics:

operation c) also comprises the production of a ring signal and the application of this ringing signal to the means for modulating the radio signal;

- The device is housed in a housing having at least one button available to the user, and said action performed voluntarily by the user is a maneuver of said button; - The device further comprises voice recognition means, and said action performed voluntarily by the user is a voice command spoken by the user;

- Operation c) also includes the production of a stall command, and the transmission of this command to the radiotelephone. In general, it has been seen above that the operation d) can comprise at least one reiteration of the algorithm of the operation b). In this case the device is advantageously housed in a housing comprising at least one button, available to the user, for changing said radio transmission frequency, and the control means are able to recognize a maneuver of the button to change the radio frequency. Frequency executed voluntarily by the user, and conditioning the repetition of the algorithm of the operation b) to the recognition of such a maneuver. Operation d) may furthermore comprise the production of a stall command and the transmission of this command to the radiotelephone. According to various advantageous subsidiary features:

the reference signal is a repetitively transmitted pulse signal, in particular a musical signal whose energy is concentrated in the frequencies above 2 kHz such that a signal chosen from a sinusoidal signal, a DTMF type signal, or a telephone ring type signal;

the algorithm of operation b) implements the evaluation of a correlation between the acoustic signal picked up by the microphone and the reference signal generated by the device;

this correlation is a statistical correlation, evaluated for a plurality of delay values applied to the reference signal produced by said generating means, with a search for a maximum among the different correlations and comparison of this maximum with a predefined threshold, to determine the presence or absence of the reference signal in the signals picked up by the microphone; the statistical correlations in question may be calculated by recursively evaluating the variances and covariances of the signals present;

the algorithm of the operation b) also implements the application to the signal picked up by the microphone of a noise reduction processing, in particular a processing including the determination of the weighted average energy spectrum of a noise signal captured between the periods of production of the reference signal, and subtraction of this spectrum of the signal picked up by the microphone during these periods of production of the reference signal;

the control means comprise means for terminating the generation of the reference signal of the operation a), and the execution of the detection algorithm of the operation b), at the end of a predetermined delay time.

0

An embodiment of the device of the invention will now be described with reference to the appended drawings in which the same reference numerals designate identical or functionally similar elements from one figure to another.

Figure 1 is a perspective view of a preferred example of an adapter according to the invention. Figure 2 is a top view of the adapter of Figure 1. Figure 3 shows schematically the various functional elements for interfacing between a mobile phone or a player of audio content, and the car radio of a motor vehicle. FIG. 4 shows, in the form of block diagrams, the functions implemented by the algorithm for extracting the signal embedded in the noise, the presence or absence of which is to be detected. Figure 5 is a general flowchart describing the sequence of the various operations and illustrating the ergonomics of the device of the invention, according to a first aspect of the invention. Figures 6 and 7 are timing diagrams illustrating the evolution over time of the different signals transmitted or received, respectively when the transmission and reception frequencies are granted and when they are not. Figure 8 is a counterpart of Figure 5 for a second aspect of the invention.

0

In Figures 1 and 2, there is illustrated in 10 an example of adapter 10 according to the invention. This adapter is in the form of an independent housing 12, plugged into the cigarette lighter socket of a vehicle by a plug 14 carrying supply contacts 16 and 18. This housing comprises wired coupling means or , very advantageously, wirelessly with a mobile phone, for example via a Bluetooth type radio link (registered trademark of Bluetooth SIG, Inc.). The Bluetooth specifications provide for the possibility of remote control over a bidirectional wireless link all functions of a mobile phone such as unhooking, hang-up, dialing, navigation menus and directories, etc., taking full control away from the mobile phone and completely replacing the keyboard of it. The presence of the mobile phone in the range of the adapter being detected fully automatically, it is not necessary for the user to take out the mobile phone, which can remain in a jacket pocket, in a bag or in a briefcase without the user having to do anything to activate the adapter. The latter becomes immediately operational as soon as the user enters his vehicle into the vehicle, all the usual commands that can be executed by means of the adapter.

The adapter also includes a low power FM radio transmitter for coupling to a car radio, so that the vehicle speakers can be used to reproduce the audio signals in the same way as any station in the band. FM. The adapter 10 includes a microphone 20, a loudspeaker or, at least, a buzzer 12, and two red and green telephone control keys 24 and 26 (especially the line / dial functions and hang up / cancel), and a multifunction button 28. This button 28 is a rotary knob that provides different functions according to the different states of the adapter, such as:

- adjusting the loudspeaker volume (if the adapter has one), - setting the transmit frequency of the FM transmitter,

- fast forward / rewind control when playing an audio stream.

The button 28 can be controlled not only by turning it, but also as a push button, to "click" and control, as the case may be: - the acknowledgment of a function or a choice in a menu,

the activation of a voice recognition mode,

- pause / play an audio stream.

Finally, the device comprises a display 30 (optional) which makes it possible in particular to display in clear the frequency of the FM transmitter incorporated in the adapter.

The selection of the frequency of the FM transmitter can be operated by the rotary knob 28 until the desired value is obtained. By pressing a pusher 32, it can also be recalled from among a series of transmission frequencies previously stored (at the factory or by the user). In addition to the digits 34 indicating the frequency in megahertz, the display 30 also comprises a number 36 for recalling the number of the memorized frequency.

These frequencies are generally chosen at both ends of the FM band, in ranges where the probability of finding a transmitter is low or zero in the geographical area considered. FIG. 3 illustrates the general structure of the various circuits of the adapter 10. This is mainly intended to be coupled to a mobile telephone 38, very advantageously via a wireless link such as a Bluetooth link provided by a module 40. It should be noted, however, that the invention would also be applicable to the case of a cable connection between the adapter 10 and, for example, a cradle on which the mobile phone 38 would be placed, the cradle then being provided with a cable. Plug-in connector for coupling the phone to the adapter. Furthermore, the adapter 10 is intended to be coupled to a car radio 42 by means of an FM transmitter module 44. This transmitter is tuned to a frequency which can, as indicated above, be selected by the rotary knob 28 or stored among preselected values selected by successive presses on the push button 32. The entire device is controlled by the central control unit 46, for example an ASIC designed and programmed to perform the various functions described below.

On the source side of the audio signals, the coupling is not necessarily a coupling to a mobile phone 38; it can also be provided with any other audio source, for example an MP3 audio player 48 connected to a socket 50 of the adapter 10.

The sources of audio signals (telephone 38, MP3 player 48 or any other source) are coupled to the car radio 42 via the adapter 10 by sending a signal on the FM band, to enable the reproduction of the telephone conversation (case of the telephone) or the audio stream (in the case of the MP3 player) on the loudspeakers 52 of the motor vehicle.

For this coupling to be effective, it is necessary that the transmission frequency of the FM transmitter 44, and the reception frequency of the car radio 42, are identical. In order to detect whether these frequencies are identical or different, the basic principle consists in causing the adapter 10 to transmit an audio signal having known characteristics, for picking up the acoustic signals present in the sound environment by means of the microphone 20, and for analyze these signals: if it is possible to recognize the reference signal in the sound environment, then we will consider that the transmit and receive frequencies match. If not, we consider will be different and it is necessary to warn the user to perform appropriate actions to readjust the frequencies and / or to inhibit the performance of such or such a feature that can not be correctly executed due to the impossibility of coupling the adapter to the car radio.

The very particular sound environment of the passenger compartment of a motor vehicle requires, however, for an effective recognition of the reference signal, to use algorithms that make it possible to efficiently extract a signal from a highly noisy environment, since otherwise Given the noise level, the reference signal would almost never be identified.

The algorithm used preferentially with the invention is illustrated by the block diagram of FIG. 4. A reference signal is emitted (block 54), this signal being chosen so as to have physico-acoustic characteristics very different from the noise. environment with respect to spectral power and predictability in particular.

In the case of an automobile the ambient noise is diffuse, rather concentrated in the low frequencies (typically on frequencies below 1 kHz) and little consistent over time. It will then be appropriate to choose as a reference signal a rather musical signal (that is to say rather coherent in time) and whose energy is rather concentrated in high frequencies (typically frequencies above 2 kHz). For example, it is possible to choose a sinusoidal signal or a DTMF type signal, produced repeatedly in the form of pulses transmitted in a loop. One can still choose a signal similar to a telephone ring, which will have the double advantage of meeting the above requirements and pass unnoticed by the user, the latter not making the difference between the signal sent to detect the existence pairing with the car radio, and a real ringing of the phone, especially if it is an incoming call that has triggered the detection process. The signal picked up by the microphone 20 is first digitized at 56, then filtered at 58 by a noise reduction algorithm. A first form of noise reduction may consist of a simple bandpass frequency filtering; this technique is particularly applicable with a simple reference signal (sinusoidal or DTMF) that is easy to extract from all surrounding sounds with a high rejection factor.

A more elaborate noise reduction technique can be implemented by applying an algorithm which will advantageously utilize the intrinsic differences between the reference signal and the surrounding noise: the noise picked up by the microphone 20 is analyzed to give a spectrum weighted mean energy of noise, continuously updated. The noise spectrum thus determined and stored will then be subtracted from the signal picked up by the microphone, in the frequency domain.

The most effective methods are based on the statistical model established by D. Malah and Y. Ephraim in:

[1] Y. Ephraim and D. Malah, Speech Enhancement Using a Minimum Mean-Square Error Short-Time Spectral Amplitude Estimator, IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol.

ASSP-32, No. 6, pp. 1109-1121, Dec. 1984, and

[2] Y. Ephraim and D. Malah, Speech Enhancement Using a Minimum Mean-Square Error Log-Spectral Amplitude Estimator, IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. ASSP-33, No. 2, pp. 443-445, April 1985.

More precisely, making the approximation that the speech and the noise are uncorrelated Gaussian processes and presupposing that the spectral power of the noise is a known datum, these two articles give an optimal solution to the problem of noise reduction in the present case. The noisy signal is divided into independent frequency components by using the discrete Fourier transform, an optimal gain is applied to each of these components and the signal thus treated is recombined. In [1], the applied gain is named STSA gain and minimizes the mean squared distance between the estimated signal (at the output of the algorithm) and the original speech signal (non-noisy), whereas in [2], a gain named LSA gain minimizes the mean squared distance between the logarithm of the amplitude of the estimated signal and the logarithm of the amplitude of the original speech signal. In all cases, the main idea is to reduce the energy of very noisy frequency components by applying a low gain while leaving intact (by applying a unit gain) those that are little or not at all. The next step of the processing consists in performing a statistical correlation calculation (blocks 60) between the captured, digitized and filtered signal and the reference signal, for different delay values (blocks 62) to reproduce the delays caused by the propagation paths between the loudspeakers and the microphone.

For the calculation of the statistical correlation, the classical correlation formula for centered X and Y signals can be chosen:

Thus, if FM (n) is the signal emitted by the FM transmitter at time n and Mic (n) is the signal acquired by the microphone after denoising, the recursive evaluations of variances and covariances can be defined by:

/ (n) = a.γ (n - 1) + (1 - a) FM (n - S) .Mic (n)

The coefficient α is a smoothing coefficient chosen appropriately. The calculation takes as final result (block 64) the maximum value of these different correlations:

It can be shown that, if we neglect the secondary paths of propagation of the signal emitted by the loudspeakers and acquired by the microphone in front of the main acoustic path and if we assume that the signal-to-noise ratio is very large in front of 1 (which will be the case after noise reduction), the quantity defined above will be close to 1 when the reference signal is present and 0 otherwise.

This is compared (block 66) with a predefined detection threshold (block 68): if the maximum statistical correlation is sufficiently high to determine the effective presence of the reference signal in the acoustic environment, then it will be considered that the frequencies of transmission and reception are identical (and they are not, if not).

Figure 5 illustrates the general ergonomics of the adapter. FIGS. 6 and 7 also illustrate, in case of agreement and disagreement of the transmission and reception frequencies, the chronology of the different signals:

FMout: modulation of the FM signal transmitted by the adapter,

- BUZZER: activation, or not of the buzzer, - Microphone: identification of the different signals picked up by the microphone.

According to a first aspect of the invention, it is for the device to react to a request produced by the microcontroller, typically in case of incoming call from the mobile phone coupled to the adapter by Bluetooth link.

When this request occurs (step 70), then the adapter transmits the REF reference signal and activates the microphone (step 72). The signal picked up by the microphone is analyzed by the algorithm described above with reference to FIG. 4, so as to determine the presence or absence of the reference signal in the acoustic environment of the device (steps 74 and 76). This analysis is performed in a loop for a maximum duration T ₀ given (step 78).

If, after this analysis, the device detects the presence of the reference signal, and therefore concludes the agreement between transmission and reception frequencies, the transmission of the reference signal is stopped, and the FM signal is then modulated by a signal imitating a ringing, so as to signal to the user the incoming call (step 80, and RINGER period in Figure 6). This ringing signal is transmitted to the car radio and reproduced on the vehicle's speakers. The adapter then waits for a user action, for a maximum duration T ₁ (steps 82 and 84). This action can be an acceptance or a refusal of the call, by interaction with the adapter, either physically (pressing a key) or by voice recognition. If the user accepts the call (FIGURE 6 digest reference), the adapter acknowledges the request that was originally issued (step 88), causing the stall and establishment of the telephone call (step 90, and period COMM in Figure 6). If no action occurs after the period Ti, the adapter sends a negative acknowledgment to the microcontroller, and disables the microphone (step 86).

We will now examine the situation when at the end of the period T ₀ the adapter does not recognize the reference signal REF in the acoustic signal picked up by the microphone, and therefore concludes that the transmission and reception frequencies do not do not agree, or more.

A signaling is produced by the device, by activation of the buzzer, and possibly flashing of the display 30. Due to the disagreement of the frequencies, this signaling can not be relayed by the car radio, and it is therefore the adapter itself. the same which must produce the signaling (step 92, and BUZZER period in FIG. 7).

The device then starts waiting for an action by the user during a maximum delay T ₂ (steps 94 to 96). If the user reacts to the signaling issued by the buzzer, for example by pressing a button of the adapter and / or trying to change the transmission frequency thereof by pressing the frequency selection buttons 28 or 32, then the adapter sends to the phone a stall signal (step 100): indeed, even if it is not yet certain that the frequencies match, we can assume that the user wishes to answer the call, and for this reason the adapter automatically triggers the stall, in anticipation.

The algorithm searches for the presence of the reference signal for a maximum duration T ₃ (steps 102, 104 and 106). It should be noted that this search can be conditioned either on the detection of an action by the user as in the illustrated flowchart, or in parallel with the detection of this action, independently of the latter. If the analysis algorithm recognizes the reference signal, this means that the user has succeeded in tuning the transmission and reception frequencies, either by changing the settings of the adapter (button 38 or preselection by the pushbutton 32). ), or by changing the frequency of reception of the radio, for example by selecting on the latter a particular frequency that the user has reserved for receiving signals from the adapter.

The buzzer is then deactivated, and an acknowledgment is sent back to the microcontroller in response to the initial request (step 110). The telephone communication can then be established (step 90).

This is illustrated in the right zone of the timing diagrams of Figure 7 where we see that the communication (COM) is established, and reproduced by the speakers of the car radio (and thus picked up by the microphone), soon after that the microphone has detected the appearance of the reference signal REF in the sound environment of the adapter.

If the frequency tuning fails, the signal REF is still not recognized and, after the timer T ₃ , the adapter abandons the procedure, commands the hang-up (step 108), sends a negative acknowledgment to microcontroller and disables the microphone (step 86). Figure 8 is a counterpart of Figure 5 for a second aspect of the invention.

According to this second aspect, the adapter seeks to prevent the user from making use of a feature requiring the matching of the transmit and receive frequencies, when this match is not effective.

In contrast to FIG. 5, in the case of FIG. 8 the request that triggers the process is a request initiated voluntarily by the user - and not a request from the microprocessor generated by an external cause such as an incoming call whose The user is not at the origin and is wanted to react.

This request may be the request to send an outgoing call by the radiotelephone 38, or the playback of an audio stream by the MP3 player 48. This user request is relayed by the microcontroller of the adapter (step 112), triggering the sending of the reference signal and the activation of the microphone (step 114). The signal picked up by the microphone is analyzed so as to detect the presence or absence of the reference signal during a maximum delay T ₀ (steps 116, 118 and 120).

If the reference signal is recognized, then the adapter stops transmitting it, and disables the microphone (step 122). It can then fully activate the requested function (step 124) so as to perform the desired action: dialing the call, listening to the music, etc. Otherwise, that is to say if no signal The reference signal is detected before the time limit T ₀ , the detection of the reference signal is stopped (step 126) and a signaling is sent back to the user, for example by activation of the buzzer (step 128).

Simultaneously, the requested function is inhibited: the numbering is not done, the music does not start, ... It avoids to start actions that could not be properly completed, because of the impossibility to obtain the reproduction of an audio signal, because of the absence of concordance between the transmission frequency of the adapter and the frequency of reception of the car radio.

Claims

1. An interfacing device between a source of audio signals (38, 48) and a radio receiver (42) comprising a tuned circuit at a predetermined radio reception frequency and that means (52) for acoustic reproduction of the signal received by said tuned circuit, this interfacing device (10) being a device for relaying the audio signals from the source to the radio receiver for restitution by the sound reproduction means of the latter, and comprising:

means (40, 50) for coupling to the source of audio signals; means (44) transmitting a radio signal, tuned to a given radio transmission frequency,

generating means (54) for producing a reference signal (REF),

means for modulating said radio signal by said reference signal or by the audio signals received by the coupling means,

a microphone (20) for picking up the acoustic signals present in the environment of the device, and

means for analyzing the acoustic signals picked up by the microphone, this interfacing device being characterized in that:

said coupling means (40) are bi-directional coupling means, able to receive from the source a request for activation of the interface device and to deliver in response a signal of acceptance, or not, of this request, and - it comprises control means (46) capable of recognizing an action, of acceptance or of refusal, performed voluntarily by the user to the device and also able, upon receipt of said request, to perform the following operations: a) by said generating means, producing (72) the reference signal; b) by means of analysis, execution (74, 76, 78) of an algorithm for detecting a known signal in a noisy environment, suitable for recognizing the presence or absence of the reference signal (REF) in the acoustic signals picked up by the microphone (20); c) if so, transmission (88) of said acceptance signal to the source of audio signals, this emission being conditional upon recognition (82) of an acceptance action performed by the user voluntarily ; d) in the negative, activation (92) of a buzzer (22) and / or visual (30) to the user, and reiteration of step b).

2. The interfacing device of claim 1, wherein said audio signal source is a radiotelephone (38), and wherein said request is transmitted on detection by the radiotelephone of an incoming call.

3. The interfacing device of claim 2, wherein the operation c) also comprises producing (80) a ring signal and applying this ring signal to the means for modulating the radio signal.

4. The interfacing device of claim 1, wherein the device (10) is housed in a housing (12) having at least one button (24, 26,

28, 32) available to the user, and wherein said action performed voluntarily by the user is a maneuver of said button.

5. The interfacing device of claim 1, wherein the device further comprises voice recognition means, and wherein said action performed voluntarily by the user is a voice command uttered by the user.

6. The interfacing device of claim 1, wherein the operation c) also comprises the production (100) of a stall control, and the transmission of this command to the radiotelephone.

7. The interface device of claim 1, wherein the device (10) is housed in a housing (12) comprising at least one button (28), available to the user, for changing said frequency of program radio, and in which the control means are able to recognize a maneuver of the frequency change button performed deliberately by the user, and to condition said reiteration (102, 104, 106) of the algorithm of the operation b) to the recognition (94) of such a maneuver.

8. The interfacing device of claim 2, wherein the operation d) also comprises the production (100) of a stall control and the transmission of this command to the radiotelephone.

The interfacing device of claim 1, wherein said reference signal (REF) is a repetitively transmitted pulse signal.

10. The interfacing device of claim 9, wherein said reference signal (REF) is a musical signal whose energy is concentrated in frequencies above 2 kHz.

11. The interfacing device of claim 9, wherein said reference signal (REF) is a signal selected from a sinusoidal signal, a DTMF type signal, or a ring-type telephone signal.

12. The interfacing device of claim 1, wherein the algorithm of operation b) implements the evaluation of a correlation between the acoustic signal picked up by the microphone and the reference signal generated by the device.

The interfacing device of claim 12, wherein said correlation is a statistical correlation (60), evaluated for a plurality of delay values (62) applied to the reference signal (REF) produced by said generator means (54). ), searching for a maximum (64) among the various correlations and comparing (66) this maximum to a predefined threshold (68) to determine the presence or absence of the reference signal in the signals picked up by the microphone .

14. The interfacing device of claim 13, wherein said statistical correlation is calculated by recursively evaluating the variances and covariances of the signals in the presence.

The interfacing device of claim 13, wherein the algorithm of operation b) also implements the application to the signal picked up by the microphone from a noise reduction processing (58).

16. The interfacing device of claim 15, wherein said noise reduction process (58) comprises determining the weighted average energy spectrum of a noise signal captured between the reference signal producing periods, and subtracting this spectrum of the signal picked up by the microphone during these periods of production of the reference signal.

17. The interfacing device of claim 1, wherein the control means comprises means (78) for terminating the generation of the reference signal of the operation a), and the execution of the algorithm. detecting operation b) after a predetermined delay time (T ₀ ).