JPH06168382A - Method for analysis of received signal from sensor and alarmdevice - Google Patents

Abstract

PURPOSE: To effectively distinguish an actual intrusion from another intrusion sequence which can disturb a standing wave and to detect intrusion when it occurs while an alarm system is in an on state. CONSTITUTION: The alarm system includes a sensor 7 for detecting a pair of standing waves generated within the cabin of a vehicle and a processor for analyzing a received signal from the sensor 7. The standing wave is generally an ultrasonic wave and the processor detects the illegal intrusion to the wave of the vehicle by detecting the change of the pattern of the standing wave. The alarm system includes a microcontroller M, which generates an adapting signal from a receiver 2 during an adapting stage for previously adapting the sensitivity of the system.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle alarm system, and more particularly, to a method of generating an alarm signal, and for analyzing a received signal from a sensor for detecting a standing wave set generated in a cabin of a vehicle. The present invention relates to an alarm device that implements the above method. The standing wave is generally an ultrasonic wave, and the device of the present invention detects intrusion into the cabin of the vehicle or destruction of the vehicle by detecting a change in the set of standing waves. ing.

[0002]

2. Description of the Prior Art At the present state of the art, the level of the signal received by a sensor is measured and then a simple shock or disturbance (not due to vehicle intrusion and destruction) is distinguished from entry into the cabin. The analysis method consisting of This latter step is performed by measuring the signal level at various frequencies.

[0003]

The problem with this type of method, however, is that, in practice, little intrusion occurs and, if at all, it is extremely short. In addition, the intrusion detection signal may be very similar to the signal that results from other phenomena, such as those that cause the vehicle to easily shake due to external transient effects. Another example of such a phenomenon that can disturb the standing wave of the ultrasonic standing wave is a phenomenon caused by heat or humidity. An example of a mechanical shock that can activate the vehicle's alarm system is a shock when the alarm system is on and the pneumatic drill starts to operate, very close to a stationary vehicle. .

These circumstances make it possible to better distinguish between actual intrusions or destruction and other events, such as heat-induced movement of air or mechanical shocks from a remote location, than ever before. A method is desired.

[0005]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method which overcomes the above drawbacks.

In the prior art, methods have been proposed which allow a processing device or circuit for processing the received signal to adapt to the main conditions, in particular with regard to the characteristics of the circuit components themselves and / or the various physical parameters. ing. Examples of physical parameters include temperature, relative humidity and pressure within the vehicle cabin. Such adaptation, as a general expression, is described herein (including the claims) as
For convenience, this is simply called “adaptation”. Presently known methods of this type are particularly adapted to suppress false alarms, which can occur, for example, due to a large increase in the amount of sunlight entering the vehicle.

However, known adaptation methods measure the average level of the received signal by integrating the signal and maintain it through a feedback loop in order to keep that level as high as possible and below a certain threshold. It relies on an automatic gain correction method that increases the gain of the processor. If the signal level becomes too high, at least in its early stages, the total gain can be reduced, so that the output signal from the processor will be below a certain threshold.

When the automatic gain control is performed at the node of the standing wave (and this is performed for a long time), the gain is maximized by the automatic gain control. The total gain of the processing circuit becomes high, and the intrusion that occurs at that time is in a short time.If this intrusion causes an undesired displacement in the standing wave pattern, the detection circuit saturates and the intrusion is no longer detected. Can not be.

Another object of the present invention is to provide an adapting means for a processing device which can detect an intrusion even under the above-mentioned unfavorable conditions.

According to the first aspect of the present invention, a detection signal from at least one sensor, for example, an ultrasonic sensor, which detects the displacement of the standing wave in the cabin of the vehicle is detected so as to detect the entry into the vehicle. The method of analysis includes: (a) pre-modifying away from the adaptation of the processor during repeated adaptation steps, and (b) repeated monitoring or investigation steps: Except for levels, it is characterized by: detecting non-routine events (eg shocks) that do not correspond to vehicle intrusions; and finally detecting actual vehicle intrusions.

According to a second aspect of the invention, at least one wave emitter (1) for generating a series of standing waves in the vehicle cabin when the system is on.
And a vehicular alarm device comprising at least one receiver (2) comprises means (M) for adapting the reception characteristics of the device, the output end of said adapting means being connected to the processing means. The processing means is characterized by including means for extracting a detection signal from noise, and thereafter means for extracting an intrusion signal from the detection signal together with an alarm means.

Other features and advantages of the present invention will be understood more clearly from a reading of the following description of the preferred embodiments, given with reference to the accompanying drawings.

FIG. 1 shows a preferred embodiment of an alarm device for implementing the method of the present invention. The alarm device comprises at least one ultrasonic wave generating module 1, which module 1 is caused by a vibration, for example 40 kHz, generated by an oscillator 5 connected to a wave emitter 1 via a connecting wire 6. When the electrodes of the are energized, they contain capsules for emitting ultrasonic waves. Oscillator 5
Is connected to the output port 10 of the microcontroller M and the oscillator 5 is adapted to supply control and / or power signals to the microcontroller.

The alarm device also includes at least one receiver 2. The receiver 2 includes a sensor capsule 14 for receiving ultrasonic waves, and an output terminal 16 of the capsule 14 is connected to one input terminal of a gain controllable amplifier 15. The amplifier 15 includes at least one gain control input terminal connected to at least one other output port 13 of the microcontroller M.

The receiver 2 includes a demodulation circuit 18 connected to an output terminal 17 of a gain controllable amplifier 15. The output 19 of the demodulator 18 is connected to the port of the microcontroller M for converting from analog mode to digital mode. The microcontroller also comprises at least one input port 9, which is connected to a sensor 7, for example a temperature sensor, a relative humidity sensor or an atmospheric pressure sensor, for measuring environmental conditions via a suitable connection 8. Connected.

The microcontroller M comprises at least one input port 22 connected to a control device 25 via a suitable connection 24. Control device 25
Can consist of, for example, at least one push button. The push button is activated by the user of the alarm system to activate the alarm system. The control device 25 may in particular consist of command execution means for reconfiguring the microcontroller M or for loading this microcontroller M with new operating programs or new data.

The microcontroller M also includes an output port 21 connected to the message device 23. This device 23 may be of the voice type using a voice synthesizer or of the audible alarm signal type using a siren or loudspeaker. It may also be of the visual type, such as an alphanumeric indicator with a console or liquid crystal display panel. Alternatively, the message device 23 may be adapted to visually display information in an illuminated and / or colored state using colored or non-colored electroluminescent diodes. The state of the alarm system is indicated by the emission or extinction of this diode.

The microcontroller M has at least one output port 26 connected to the control input 27 of a suitable alarm device, which alarm device
In particular to generate a signal for disabling alarm sirens and / or essential functions of the protected vehicle, eg a signal for disabling a fuel injection computer or ignition, or a signal for deactivating the alternator. May be included. Finally, the microcontroller M has a power supply input 11 connected to the power supply. The power supply comprises, for example, a vehicle battery, or other suitable power generation device, such as a voltage regulator connected to an uninterruptible power supply during activation of the alarm system or to respond to a power failure.

Referring now to FIG. This figure shows an embodiment of the receiver 2 of the system shown in FIG. Figure 2
In the figure, the ultrasonic receiving capsule is indicated by reference numeral 30, and this capsule 30 is connected to the ground and the first gain controllable amplifier 3
2 connected to a suitable input 31. The gain control input 33 of this amplifier is connected to an access terminal 46 of a module 49 containing a microcontroller M via a suitable connection 34. Access terminal 46 sends control signals. The value of this control signal, indicated by the unidirectional voltage level, can determine the gain required to operate amplifier 32.

The amplified output signal 35 of the amplifier 32 is
It is sent to the input terminal of the demodulator 36. The output terminal 37 of the demodulator 36 is connected to a low-pass type first filter 38. The output end 39 of the filter 38 is first connected to the input end of a high-pass type second filter 40, and then
It is connected to another access terminal 52 of the module 49 via the connecting portion 42. The output terminal 41 of the high-pass filter 40 has a second gain controllable amplifier 43.
Connected to the input end of the amplifier 43, and the gain control input end 4 of the amplifier 43.
4 is connected to the third access terminal 47 of the module 49 via a suitable connection 45. The amplifying output end of the second amplifier 43 is connected to the fourth access terminal 48 of the module 49 via an appropriate connecting portion.

Microcontroller M in module 49 includes two circuits 50 and 51 (designated CAN1 and CAN2, respectively) for converting from analog mode to digital mode. This circuit 50
Receives the amplified output signal from the amplifier via access terminal 48, while circuit 51 receives the output signal from low pass filter 38 via access terminal 52.

In the embodiment shown in FIG. 2, the circuit parts shown to the left of the access terminals 46, 47, 48 and 50 of the module 49 are of analog type, adapted to detect any vehicle type. It consists of a standard circuit. On the other hand, the parts shown to the left of these terminals, i.e. the parts within the module 49, are digital and unique to a particular vehicle type and for a particular application. If standardization of mass production of the alarm device is promoted, each module 49 can be dedicated by a logic circuit. This significant reduction in manufacturing cost is a major advantage of the system of the present invention.
This reduction in cost is directly related to the particular nature of the circuit architecture and the method used, as will be seen later.

This method consists essentially of a series of adaptation steps, followed by a monitoring or investigation (or inspection) step. Some of these steps may vary depending on the situation, but the continuation of these steps is predetermined.

When the alarm system is activated, the initial adaptation phase begins. During this phase, the launcher 1 of the system launches a series of waves of a predetermined waveform. The emission of this series of waves is carried out such that the receiver 2 detects a normally protected ultrasonic wave in a space-specific traveling mode at the start and / or end of each pulse.

The microcontroller that collects the data performs the adaptive operation assuming that the vehicle has not been invaded or destroyed. For this purpose, the microcontroller stores the minimum and maximum values of the received wave. Figure 3 (a)
(B) shows such reception. A firing with a specific envelope ENV-E obtained by suitable control of an oscillator 5 via an output port 10 of the microcontroller activated according to a specific program contained in a memory zone (not shown) of the microcontroller M. Wave row OE
Is received by the receiver 2, the receiver generates a response indicated by a curve OR representing the received wave (see FIG. 3A).

The signal received by the receiver 2 is the analog / digital conversion port 19 of the microcontroller M.
To the minimum value Vmin by a specific program in the memory zone (not shown) of the microcontroller M.
And the maximum value Vmax can be stored. In one embodiment, after Nt rows of similar waves OE have been fired,
This pair of values is obtained and calculated with the mean value represented by the equation below.

"Formula 1" <Vmax> = (total of (Vmax (i))) / Nt <Vmin> = (total of (Vmin (i))) / Nt

In particular, if the magnitude of the average value is increased and the relationship with the balance coefficient is determined, another average value can be obtained according to the waveform of the row of the wave OE.

In one embodiment, the resulting pair of values (Vmax,
If Vmin) is stable, i.e. does not change during the firing of the new wave train OE, the adaptation phase ends. Generally, this situation is
Hundreds of milliseconds last for a wave train of 50 milliseconds. As a result, another embodiment counts a predetermined number of acquisitions corresponding to a sufficiently long time for the ultrasound system to stabilize.

The microcontroller M estimates the detection values from the adaptation values of the processing device by means of a prerecorded table or by means of a table according to a prerecorded relational expression. In one embodiment, these adaptation values consist of gain control values that are also applied by the microcontroller M to the first amplifier 32 via connection line 34. Due to the gain of the first amplifier, the average voltage of the reception signal SR (see FIG. 3B) is
The value can be set to the following formula.

"Formula 2" Vm <Vsat-G

Here, Vsat is a saturation value of the processing circuit, and G is a protection value determined according to the characteristics of the receiving circuit.

The microcontroller M includes the second amplifier 4
A gain value of 3 is defined as the adaptation value. Since this value is sent to the input end 44 of the second amplifier 43 via the connection line 45, the maximum peak value of the output signal from the high-pass filter 40 of the processing circuit (where the received signal loses its one-way component) is Below the saturation value, the protection value corresponds in particular to the detection value which corresponds to the detection of the entry into the vehicle. For this reason, the microcontroller M guarantees:

"Formula 3" Vcm <Vsat-g

In the method of the present invention, the circuit of the receiver is adapted to the maximum detection sensitivity without considering the saturation when the interference with the vehicle is detected.

The adaptation step is repeated at predetermined intervals, for example in a predetermined cycle. When the intrusion into the vehicle is detected, the detection is stopped. In a further preferred embodiment, in order to prevent unwanted sensitivity variations, i.e. variations in the two values of said gain, these variations being adaptations of the processing circuit, several modulation steps, for example 500 mm. Only after seconds are modulation steps away from each other, it is admitted to change the previous adjustment of the adaptation value only if the adaptation value is confirmed.

Now referring to FIG. This figure shows that detection of the received signal is executed while noise is greatly suppressed.
FIG. 5 is a sequence diagram of some of the functions performed by the programmed microcontroller M. 5 and 6 show the spectrum reception characteristics of this method.

In FIG. 4, the microcontroller M
Has a routine or subprogram for the input 60 of the data X (p), where p is the second amplifier 43 (FIG. 2).
Shows the sequence of the numerical sequences available at the analog switching port 48 at the output of the. A series of sequence pieces of data X (p) are supplied to a memory 61 having an address, which stores these sequence pieces in a plurality of high-pass and / or low-pass filter passes F1, F.
2 ,. ． ． , FN (for number N). These filter paths operate in parallel and each path (typically path Fi
,), For example, the path F1 in succession includes: That is, the input means 62 receives an item of data X (p) and the output means 64 produces a filtered value X (p), and a single detection of the direct wave from the emitter. Filter paths F2 to FN are passed through a demodulating means 65 for recovering the wave and generating a demodulated output value X (p) *, and a path 68 to the output of the measuring means 69.
Of the input terminals 69-2 ,. ． ． 69-N
And level measuring means 67 for generating a signal at the output terminal 69-1 connected to the selecting means 70 having

The selection means 70 also has a first output 71 for an average value, which average value is selected, demodulated, filtered and then means 76 for processing and identifying levels and times. Sent to. The selection means 70 also has a second output 72, which sends the serial number of the filter path (for example F1 in this example) to be selected, as will be described later.

Next, referring to FIG. This figure shows the spectral characteristics identified by the method of the invention, with the thermally generated noise BR for the lowest frequencies. This noise varies between continuous or unidirectional states and frequencies of a few hertz, typically 10 hertz. The noise curve has a shape made up of a main portion and downward spectrally inclined portions Br1, Br2, Br3, Br4 that change with time. Since the level Av of the received signal is often lower than the level of noise, it is difficult to detect the received signal.

To deal with this situation, a high-pass type digital filtering with a spectral characteristic FPH adapted according to the specific situation is performed. By performing the high-pass type filter processing, the received signal can be output at the noise level even when the received signal level is lower than the signal level. It has been found that the signal spectrum indicative of vehicle entry is typically greater than the noise limiting frequency. Therefore, except for the noise level, the received signal can be detected even if the noise interferes with the signal to be detected.

To this end, several filter paths F1, F2 ,. ． ． , FN. The cut-off frequencies of these filter paths extend from a few hertz to above 10 hertz, which is the maximum limiting frequency of noise. However, it is necessary to be able to select a particular filter (or filter path) Fi that satisfies at least one of the following conditions. That is, the condition of the lowest cut-off frequency, and the condition of the lowest detection level when there is no entry into the vehicle, especially in the adaptation stage.

Next, according to the following formula, for example, PPH executed digitally over P0 cycle X (p)
The filtering process indicated by is executed.

"Formula 4" <X (p)> = {a (p-P0 + 1). X (p-P0 + 1) + a (p-P0 + 2). X (p-P0 + 2) +. ． ． + A (p). X (p)} / P0

Equations with equal balance factors a (i) and relations with cycles X (pk) increased to the series b of perfect numbers (ie the filtering is the magnitude of b) are available.

A predetermined period of time elapses before the first value X (p) is available, during which P0 consecutive cycles X
(P) occurs, during which the mean value X (p) is calculated. As a result, if the selected filter path Fi is appropriate in a state where the noise level is higher than the level of the filtered signal, and if an intrusion occurs at this time, the selection of the filter path Fi should be changed. May be too late.

It is a preferred feature of the present invention that an adapted filter path Fi0 which produces a minimum voltage value i0 at the output 69 can be preselected, since a bank of filter paths is available in the adaptation stage. This value i0
Is fed back by the selection means 70 via its output 72 and the connecting line 74 to the selection means, which selects the next item of data received in the next examination (or investigation) stage in a suitable filter path. It is possible to turn to Fi0.

FIG. 6 shows another spectral characteristic. With this spectral characteristic, all the filters have the same cutoff frequency fc, but the slopes are as large as 1, 2, and so on. The selected filter path should have the lowest slope (ie, the lowest magnitude). It must also be at the lowest detection level when there is no entry into the vehicle.

It is also possible to combine a bank of high-pass filters, which uses a combination of a method of changing the slope (size of filtering) and a method of changing the cutoff frequency.

In the preferred embodiment, a single subprogram is used to compute the filters one after the other. In the next adaptation stage, these filters are fed with a control signal which is equivalent to the selection control signal on line 74 and which adjusts the cutoff frequency and / or the magnitude of the filtering.

Returning to FIG. 4, the process for the level and the time, and the operation of the identifying means 76 will be described.

The inventors of the present invention have found that events that do not respond to intrusion or destruction situations and that generate false alarms in conventional systems have very similar spectral characteristics but differ in time and level. Discovered by experiment. Therefore, the processing means 76 determines that the selected reception level X
(P) The measurement of * is performed, and these are compared with a predetermined threshold value. If the period of the received signal at a level greater than a predetermined threshold value exceeds a predetermined value, the output 77 of the processing means 76 will be provided with an alarm device or other suitable means for indicating that an intrusion has occurred. It will be an active level, as will the momentum.

Next, referring to FIG. This figure shows an example of an algorithm used in the processing means 76. The item of data resulting from the filter means Fi selected as described above is denoted as X (n). These items are sampled (or collected) in step S1 at a predetermined frequency equal to 100 Hertz in the preferred embodiment. These items are then filtered.
This filtering operation consists of a combination of a first or second magnitude analog filtering operation and a subsequent second magnitude digital filtering operation. As a result, in the preferred embodiment, high pass filtering is performed with a cutoff frequency greater than the threshold noise frequency equal to 10 Hertz. To the demodulation means,
The output signal X1 (n) is sent, and the demodulation means performs step S
3. Calculate the absolute value of the filtered signal in 3.

Next, in step S4, the output X1
Smoothing is performed as (n) = X2 (n). In one embodiment, this smoothing operation is performed with a lowpass filter having a cutoff frequency equal to 5 Hertz and a slope of magnitude one.

The smoothed output X3 (n) is the value X3
Whether or not (n) is larger than a predetermined threshold value S is tested in step S5. This predetermined threshold value S is a predetermined function f of parameters linked to the sensitivity level of the system. That is, S =
f (N). This threshold value serves to at least partially define an intrusion threshold value that triggers the alarm function.

If the test result in step S5 is affirmative (YES), the system proceeds to step S11, where the first counter C is incremented by one unit and the second counter D is reset to zero. It

In the test step S12, it is detected whether or not the first counter C is set to a value lower than the predetermined value Cmax. This value Cmax is equal to 80 in one embodiment and corresponds to a time of 800 msec for a detected signal level greater than the detected threshold value. As a result, the intrusion can be reliably detected without giving a false alarm. If the response to the test in S12 is affirmative (YES), the alarm state is estimated and the alarm action is executed in step S13. On the other hand, if the response to the test at S12 is negative (NO), the system moves to the "end" routine. During this routine, in particular by testing the shifting of the disabling signal provided by the control means for accessing the vehicle (for example the central locking unit) to the active state,
A test is carried out whether the system is put into the adaptation phase (which is not shown in FIG. 7) or whether it remains in the investigation or monitoring phase. If nothing happens, the system returns to the signal sampling or acquisition stage (step 1).

If the response to the test in step S5 is negative (NO), the process moves to step S6, and a test is executed here to see if the current value of the first counter C is greater than zero. If the response to the test in S6 is negative (NO), the system immediately returns to the "end" above.
Immediately return to the routine. However, if the response to the test in S6 is affirmative (YES), the second counter D increments and measures the time when the level of the detection signal is lower than the intrusion threshold S.

Next, another test in step S8 is executed so that the current value of the second counter D is equal to the threshold value N.
Is greater than. This threshold value N is the detected threshold value S in step S5.
Also works to determine. In the embodiment embodying the present invention, the time Cmax of 800 milliseconds at the first counting time
A value of 30 corresponding to a shorter time of 300 milliseconds,
Select for threshold N. Therefore, except for the silent time, which is shorter than the maximum time of 300 milliseconds, the two counters C and D can be used to detect, for example, a phenomenon that was confused with an actual intrusion or destruction in the past, such as a false alarm due to repeated shock. It turns out that can be prevented with.

If the response to test S8 is negative (NO), the system moves to the "end" step S10. On the other hand, if the response to test S8 is affirmative (YES), then two counters C and D
Is reset to 0 and the system returns to the "end" routine again.

The present invention can be embodied in modes other than the above. In particular, the processing means may be wholly digital or wholly analog within the scope of the invention.

[Brief description of drawings]

1 is a block diagram showing a processing apparatus for carrying out the method of the present invention.

2 is a block diagram showing an embodiment of a receiver in the apparatus of FIG. 1. FIG.

FIG. 3 is a graph showing signals received and processed in the apparatus generally shown in FIG. 2 at an early stage of the method of the present invention.

FIG. 4 is a sequence diagram illustrating various functions of an apparatus for detecting a received (or detected) signal while significantly suppressing noise.

FIG. 5 is a diagram showing spectrum reception characteristics.

FIG. 6 is a diagram showing spectrum reception characteristics.

7 is a sequence diagram showing an embodiment of an algorithm used in the processing means in the method shown in FIG.

[Explanation of symbols]

1 wave emitter 2 oscillator 5 oscillator 7 sensor 8 connection part 9 input port 11 power input end 13 output port 14 sensor capsule 15 amplifier 16 output end 18 demodulator 19 output end 21 output port 23 message device 24 connection part 25 control device 26 Output Port 27 Control Input 30 Ultrasonic Optical Capsule 32 Amplifier 33 Input 34 Connection 35 Output Signal 36 Demodulator 37, 39 Output 38, 40 Filter 41 Output 42 Connection 43 Amplifier 46, 47 Access Terminal 48 Terminal 49 modules 50, 51 circuits

Claims (8)

[Claims] 1. A method for analyzing a detection signal from at least one sensor, for example, an ultrasonic sensor, which detects a displacement of a standing wave in a cabin of a vehicle so as to detect intrusion into the vehicle, comprising: ) During the repeated adaptation steps, pre-modification away from the adaptation of the processor, and (b) during the repeated monitoring or investigation steps: -except for the level of noise, especially noise due to heat-to the vehicle A non-routine event that does not correspond to the vehicle's intrusion (for example, a shock); and finally detecting the actual vehicle intrusion. 2. Detected data by emitting a train of waves of a predetermined waveform in at least one of said adapting steps to generate a value for adapting the processor so that the collected data is stable. The method of claim 1, wherein the items are collected. 3. A method as claimed in claim 2, characterized in that, after several adaptation steps, a modification of the previous adjustment of the adaptation value is allowed only if the adaptation value is confirmed. 4. The step of removing noise levels comprises a plurality of simultaneously operating filter paths (F1, F2, ... FN).
And / or intruding into the vehicle in response to at least one of the following conditions: having the lowest possible cutoff frequency, and / or having the lowest possible slope. 3. The method of claim 2 including selecting a filter path having a spectral characteristic responsive to the condition of minimum detection level in the absence of. 5. Method according to claim 4, characterized in that the step of selecting a filter path is carried out beforehand in the adaptation step. 6. A count (C) having a first period when a detected value becomes larger than a predetermined intrusion threshold value.
Then, when the detection value becomes smaller than the intrusion threshold value (S), the counting (D) having the second period is started, and thus only the suspended detection signal is: a predetermined threshold value ( A signal having a first period (C) that is at least equal to Cmax), and a signal that does not interfere with the detected value greater than the intrusion threshold value for a period (D) that is at least equal to a predetermined threshold value (N). The method according to claim 1, characterized in that it prevents false alarms. 7. A system comprising at least one wave emitter (1) and at least one receiver (2) for generating a series of standing waves in a vehicle cabin when the system is on. The vehicle alarm device includes means (M) for adapting the reception characteristic of the device, the output end of the adapting means is connected to the processing means, and the processing means extracts a detection signal from noise. An alarm device for a vehicle, comprising: means for doing so and thereafter means for extracting an intrusion signal from a detection signal together with an alarm means. 8. A first amplifier (32) having a gain control means (33) responsive to said adapting means (M), said adapting means (M) having a threshold value in the adapting step. Also includes a microcontroller which produces an output signal of a magnitude pre-adapted to a lower level, said output means being fed to a low pass filter (38), the output signal from which is a high pass filter (40).
Of the high-pass filter (40) to the analog / digital conversion port of the microcontroller (M), and the gain control means (44) controllable by the microcontroller. The input end of the amplifier (43) is connected, the output end of the second amplifier (43) is connected to another analog / digital conversion port of the microcontroller, and the microcontroller is connected to the first and second amplifiers. Device according to claim 7, characterized in that it comprises means for pre-adapting the device by adjusting the gain of the device.