FI94287C - Method for noise attenuation and encoder construction for measuring a signal on the surface of solid material - Google Patents

Description

The invention relates to a method for attenuating a signal to be measured from a surface of a solid and to a sensor design applying this method. The invention can be applied to measuring an audio signal e.g. in technical and medical research and diagnostic tasks.

10 Active attenuation is already known from many applications in which the intended interference sound, i.e. noise, is sought to be attenuated by producing a counter-sound. Harmful vibrations are damped by the most active vibration damping means. The best way to dampen the so-called one-dimensional low-frequency (less than 300 Hz) sound, which typically propagates e.g. in ventilation ducts (U.S. Patent 2,043,416). Compensation for two- and three-dimensional sound is more difficult and is usually only possible in a small volume range, because the counter-sound must be in the opposite phase to the original, and thus requires good temporal accuracy. Systems in which noise is attenuated in the vicinity of the driver's head have been made e.g. for work machines and helicopters. The aim is to improve the comfort of passengers in the means of transport by actively compensating the noise of 20 countries by producing a resonance around the passenger's head. The systems are usually not very usable. The person is tied in place and already when turning the head, the level of damping varies. At worst, the systems fail and the attenuation is either non-existent or the noise may even increase.

.. 25

In certain e.g. in technical and medical registration situations, there is a need to measure the signal from a device or patient under interfering conditions. In this case, the useful signal is obtained by measuring from the surface of the body or device, but the interference (sound) signal simultaneously enters three-dimensionally through the air and adds to the signal to be measured. The problem is worst when the signal to be measured is weak. In this case, it is necessary to use sensitive sensors, such as a microphone, which also accurately registers the pressure fluctuations of the surrounding air, i.e. the interference signal.

In one application task, the recording of respiratory and cardiac sounds, 3 5 Good quality microphones and only passive noise protection are used. The noise protectors used, for example a hearing protector with a rather large microphone mounted through the rear wall, are too large in construction, especially for measurements from children, but also from the neck of adults. Even completely interference-free 94287 2 microphones are used, in which case the measuring room must be available for quiet registration. When using accelerometers to measure sound, interference from the air is only slightly coupled to the signal. However, the use of accelerometers is e.g. in medical sound studies more limited due to their insensitivity to 5 and the frequency range limited to lower frequencies.

In order to solve the problems described above, it is an object of the invention to provide an effective interference suppression solution for measurement needs in medicine and technology in situations where often rather weak signals, e.g. audio signals, need to be measured from, for example, a machine, device or patient. In the following, some known solutions which are somewhat close in solution or implementation to the solution of the invention are considered in the prior art.

15 Swedish Offenlegungsschrift SE-452 946 discloses an arrangement for detecting breathing, in which, in one example, one microphone measures the sound of breathing from the front of the human neck and the other microphone is placed on the side of the neck, neck or behind the ear. The signals are subtracted from each other, resulting in an indication that the patient is not breathing. The solution seems simple, 20 but it is clear to those skilled in the art of sound processing that when microphones are somewhat apart and no attention is paid to their protection, the solution requires signal preprocessing, which includes low-pass filtering, audio signal rectification, and signal averaging. The result is the purpose of the invention; 25 accordingly, only the voltage indicating respiration, and the actual audible signal no longer occurs. The solution is not suitable for purposes where the original signal is to be cleaned of ambient noise and kept as intact as possible in its original waveform.

WO-90/09083 describes a microphone for measuring body sounds. The solution is based on a fixed and closed chamber with a mass element suspended on a flexible membrane inside. The membrane divides the chamber into two parts, from which sound is conducted along two air ducts to two microphones, the signals of which are summed by electronic bridge connection. The signal is obtained from the opposite oscillation of the chamber and the mass inside it and, as a result, from the opposite pressure fluctuations formed in the two chambers. The problem is that the summation of the signals from the two opposite-phase pressure chambers via the two microphones results not only in noise compensation but also in compensation of 3,94287 low-frequency signals. The solution is also insensitive because the sound vibrations of the body must move the entire shell structure. In addition, the soft tissue and the sensor mass form a harmonic oscillator with a resonant frequency. With the exception of the fixed single-layer shell, passive noise protection has not been used, and the microphone elements, among other things, are free and thus susceptible to interference. Nevertheless, the patent does not require that the microphone elements should be located close to each other. There are no examples in the publication that would show in which frequency range and at what level noise control is successful.

U.S. Pat. No. 4,985,925 discloses an active noise reduction system designed primarily as an electronically operated hearing protector to be placed in the ear canal and to facilitate communication in noisy conditions. The system consists of an adapter placed in the ear canal and an electronic hearing protector placed inside it. The frame structure forms a passively attenuating shell part with a double transducer inside, which acts as a microphone and a speaker. The sound sensor speaker measures passively attenuated ambient sound as a microphone. It is acoustically summed by the difference sound obtained when the electrical signal generated by the microphone is subtracted from the useful signal. The distinction between gain and interference signals 20 is based in part on the fact that the gain signal is routed to the hearing protector on a separate line from the outside, and in part on frequency division: the gain signal, e.g., speech, is focused on higher frequencies than environmental interference, e.g., industrial and traffic noise. The payload signal is pre-filtered and amplified. The attenuation power, which is weighted at low frequencies, is based on the immediate detection of the mechanical movement caused by the sound pressure 25 and its compensation by the electromagnetic counterforce obtained by the negative feedback. The achievable upper limit frequency of active compensation is approx. 1 kHz. The compensation shown works moderately well in the ear canal, which is an approximately one-dimensional space.

30

U.S. Pat. No. 4,985,925 further mentions that its solution can also be applied to a stethoscope. The sensor head of the stethoscope is then conventional and the hearing protector according to the solution is located in the ear canal to attenuate the ambient noise while listening to the patient. Compensation is based on a single microphone in which the benefit-35 and interference signals are summed acoustically. The useful signal can be taken from an external source, such as a microphone, to which the signal must be measured more disturbed than the location of the ear. However, in the case of a stethoscope, the patient and the sensor head of the stethoscope are in the same room as the listener and his or her ears, so that the hearing protectors are not very useful because the noise is coupled to the signal through the sensor head.

In the present invention, the principle of the solution is that no echo is produced, but the aim is to obtain the purest possible signal in interfering conditions by measuring the interference acoustically similar with two microphones and then compensating them electronically at the line level. This can only be achieved successfully if the sensor design is acoustically properly constructed and dimensioned.

The solution according to the invention is characterized by measuring the useful signal from the surface of the solid with one microphone and simultaneously measuring the disturbances coming through the medium surrounding the measuring point (air, other gas, liquid) with two microphones. Interference can be removed from the useful signal by summing the inverted interference signal. The sensor design includes effective passive interference suppression. Two microphones are required inside the shield, which enables the implementation of active interference suppression. The distance between the microphones is as small as possible, which results in a good level of compensation and fairly high compensation frequencies.

20 The microphones are tightly seated in an acoustically good insulating material, in which acoustic chambers are made isolated from each other in front of each microphone, so that, despite the small distance, the useful signal is not only transmitted to the interfering microphone. The acoustic chambers have the same volume, they want to have the same (same spectrum) interference signal coupled to the microphones. The faces of the microphones are in direct acoustic communication with each other, which means that the distance between the microphones is as small as possible with respect to the interference signal.

The gain of the useful signal is enhanced by a wide and low acoustic chamber with a small volume, which means that the resonant frequency of the chamber is sufficiently high above 30 to 30 so that it does not distort the signal to be measured.

Thin symmetrical pressure equalization holes lead out of the chambers, which reduces very low frequency interference from the measurement object and improves the compensation result.

The microphones act as both sensors and physically insulating plugs at the same time.

35 5 94287

The signals of both microphones can be recorded from their own wires, in which case the compensation can be done either electronically by circuit technology or, for example, in a computer, where the compensation can be further enhanced by combining mathematical and signal processing methods.

5

The interference attenuation method according to the invention for improving the quality of the signal measured on the surface of a solid is characterized by what is stated in claim 1.

The sensor construction according to the invention is characterized by what is stated in claim 4.

Other claims describe embodiments of the interference suppression method and sensor construction according to the invention.

15

The method and device according to the invention provide decisive improvements over the existing technology presented above in connection with the review of the prior art. First, the invention achieves a clearly better compensation result than the proposed compensation solutions without substantially altering the original sig-20 signal. Thus, the interference-free signal can be used for accurate waveform and frequency analyzes after compensation. Second, the invention achieves better overall interference attenuation than known sensor solutions. Third, the signal capture efficiency of the sensor design under interfering conditions is better, especially at high frequencies, because the main microphone is in direct contact with the surface to be measured. In addition, the sensor construction according to the invention can be built significantly smaller than existing solutions, since only the size of the available microphone elements is limited.

30 The sensor design is simple because no electronic compensation circuits are included. The compensation of the two available signals is done afterwards, either in real time or for the stored signals later. It is essential that the signals be obtained from microphones for further processing. The upper limit of the compensation depends only on the size and the distance between the microphone elements and is higher than, for example, in the solution of U.S. Pat. No. 4,985,925, in which the gain and interference signals are separated in the compensation circuit by filtering, which means a significant modification of the gain signal. In the present invention, the signal is obtained directly from the surface of the solid as the original. The signal source becomes passively protected from noise when the sensor structure is physically placed on top of the measuring point and pressed tightly against the surface.

94287 6

The invention will now be described in detail with reference to the accompanying drawings, in which: Figure 1 shows an embodiment of a sensor structure according to the invention (a) from below, (b) from the side, (c) the decision maker (d) from the side of the line and Figures 2 and 3 show 1 results of frequency response measurements made to test the sensor design.

Figure 1 shows a disturbance compensating sensor construction measuring the surface of a solid, which is built for medical measurements, in particular breathing sound measurements 15, but is also suitable e.g. for measuring other sounds in the body, such as heart sounds, as well as many technical measurements in industry.

The capsule size of the microphones 1, 2 used is small and the distance between them is short, which means that the compensation band to be achieved is wide, starting from less than 50 Hz to 20 and exceeding 5000 Hz (Figures 2 and 3). At the same time, the construction is small, light and easy to use; therefore, the battery required to supply the microphones is also located at one end of the line 11. The implemented sensor structure comprises an effective passively damping multilayer shell part 8 consisting of nested hard and porous layers, which in the exemplary device; 25 is two pairs of layers. The more layers in the structure, the more effective the damping. The shell part is connected and the microphones are tightly implanted in an acoustically good insulating material 7, in which chambers 3 and 4, which are spaced apart in front of each microphone element, are machined, so that the useful signal does not pass into the interfering microphone. The rear sides of the microphones are in direct acoustic communication with each other via a chamber 5, the function of which is to realize the coupling of the interference signal coming through the passive shield 7, 8 in the same way to both microphones. The useful signal is obtained through wide and low, which characteristics amplify the signal, and through a sufficiently small acoustic chamber 3, which means that the resonant frequency of the chamber is so high that it does not distort the sig-35 signal. In front of the interference measuring microphone 2 there is an acoustic chamber 4, the function of which is to acoustically similarize the interference signal coupled to the microphones. The chamber 4 is approximately the same size as the chamber 3 in use, because the chamber 3 shrinks when the construction is pressed onto the patient's skin with the skin bulging slightly supported inside the chamber. Thin pressure equalization holes 10 lead out of the chambers and are placed very close to each other on the outer surface. The holes reduce the very low frequency pressure fluctuations caused by the measurement object and improve the compensation result. The interior containing the components 6 of the microphone preamplifiers is effectively acoustically acoustically absorbed by the attenuation material 9 to prevent echo and resonance.

In the present invention, the active attenuation is based on two identical microphones 1 and 2 arranged in close proximity to each other, implemented in a construction of two microphones 1 and 2, the signals of which can be combined so that interference in the payload signal is attenuated only by the interference measuring microphone 2. The function of the multilayer, effective passive noise shield 8 in the protective housing of the sensor structure is to attenuate all but primarily high frequencies. Passive attenuation affects the interference signal to both microphones in the same way, i.e. the interference is already passively attenuated before going into active attenuation. The function of active attenuation is primarily to reduce low interference frequencies, which are not very effectively affected by passive protection in a small structure for physical reasons. In order to obtain the best possible signal with the sensor design, the utility microphone must have good contact with the surface to be riveted. This connection takes place directly and efficiently via the acoustic chamber 3 when the sensor structure is pressed against the measuring surface. In order for the compensation to be as effective as possible, the interference signal must be connected as closely as possible to the two microphones. This is done more effectively the closer the microphones are to each other. Since the microphones cannot be at exactly the same point, the situation is improved if the acoustic conditions of the microphones 25, such as the chambers 3 and 4 in front of them, correspond acoustically well and the chamber 5 at the rear is similar for both microphones. The pressure equalization holes 10 leading out of the chambers must also be of the same size and length. In order for the external pressure fluctuations entering through the holes to come to both microphones simultaneously, the outlets of the holes 30 must be located on the outer surface very close to each other. However, the holes must not coincide, because then the useful signal is also transmitted to the interfering microphone along the channels 10 when weakened and results in partial compensation of the useful signal.

35 In order to be placed very close to each other, the microphone elements must be tightly embedded in a rigid acoustic insulating material, in the construction of the body 7. In addition to their main function, the microphone capsules act as acoustic plugs, as the useful signal through the chamber 3 must not pass behind microphones acoustically close to each other. The purpose of the connecting chamber is to make the interference signal coupled to both microphones the same so that the compensation result is as close to zero as possible so that only the useful signal remains pure. The chamber 4 of the interference measuring microphone 5 2 alone is not in communication with the acoustic chamber 3 of the useful signal, but there is a strong insulating wall of a sturdy body material 7 in between. The necessary electronic components 6 for the microphone preamplifier are placed inside the structure. The free space is filled with an acoustically attenuating material 9 in order to keep the level of interference sounds as low as possible, without echo and without resonating.

10 The pressure equalization holes 10 allow some pressure fluctuation in, i.e. the noise level connected to the microphone capsules increases, but at the same time the compensation improves. The effect is strongly dependent on the diameter and length of the holes and their ratio.

Figures 2 and 3 show the frequency response measurement results made for testing the exemplary sensor design. Figure 2 shows the plane curves in white noise and the attenuation comparison when the sensor design is tuned to active attenuation and Figure 3 shows the plane curves in white noise and the attenuation comparison when the sensor design is tuned to passive attenuation. It is characteristic of the embodiment that the sensor construction implements the active and passive attenuation curves shown in Figures 2 and 3 when riveting.

A good quality hi-fi speaker produced white noise from a test CD published by HIFI magazine and measurements were made in the frequency band 40 - 6000 Hz normal. : In 25 rooms. The distance between the speaker and the microphone was 25 cm. Figure 2 shows the plane curves when the sensor design has been tuned to active damping. The curve 'Noise directly' is measured with an open microphone. 'Utility channel' and 'Noise channel' are measured during normal operation of the device during background noise, but without a useful signal. The curves can be used to make an attenuation comparison of 30 lu in the frequency range 50 to 5000 Hz. It is observed from the curves that passive attenuation only has an effective effect from 2000 Hz upwards when active attenuation starts below 50 Hz and continues up to 5000 Hz. Compared to the payload channel, the ‘Compensated’ curve, i.e. active attenuation, is significant, i.e., more than 10 dB, in a wide band, 60 to 3000 Hz, when in sound production systems higher-35 dB values can be achieved, but in a much narrower band. A suitable way to compare the total result is to calculate the decibel values by tertiary band and sum these. The total attenuation result calculated from Fig. 2 by the invention is 231 tersecibels, while the Chelsean method 87,994287 tersideibles have been reached in the sound compensations and the counter-noise produced in the pilot's earphones is 161 teribels. In the frequency band 100-5000 Hz shown in U.S. Patent 4,985,925, a value of 207.70 tertiary sibels has been reached, while in the corresponding band of the present invention a value of 210.35 tertiary sibels has been measured.

5

In Figure 3, for ease of comparison, Figure 3 shows the same three curves as in Figure 2 ('Benefit channel' = 'Active tuning'), but in Figure 2 the noise channel is replaced by Figure 3 'Pass, tuning', measured with the sensor design tuned to passive attenuation. which occurred by closing the pressure equalization holes (10). Then the passive attenuation 10 changes in the 'Akt. tuning 'from the level of the curve to the' Pass, tuning 'level. Comparing the ‘Pass, Tuning’ and ‘Compensated’ curves in Figure 3, it is observed that active compensation attenuates the amount from 0 to 12 dB in the frequency band 100 to 2000 Hz. The total effect of the attenuation is 120 tertzides. This attenuation can be considered as an absolute improvement due to the active attenuation of the invention compared to the level achieved with passive attenuation. By tuning and developing the invention, it is probably possible to achieve an even better attenuation result.

Claims (12)

A method of noise attenuation to improve the quality of a utility signal measured on the surface of a solid, where the noise signal is transmitted via a surrounding medium and where a utility signal and a noise signal are measured with a first microphone (1) and only noise signal is measured. with a similar or corresponding second microphone (2), characterized in that the effects of passive and active noise attenuation are combined so that the first microphone (1) and the second microphone (2) are placed acoustically isolated from each other in a protective housing (7, 8). provided with an acoustic damping material (9) which prevents resonances and echoes, to passively attenuate a disturbance signal which arrives at each microphone (1,2) substantially with the same effect, the microphones (1,2) being simultaneously placed in the smallest possible distance from each other, in order to cause the interference signal to be connected to the microphones (1, 2) in the same form with substantial accuracy, the coupling of interference signals the signal to the microphones (1, 2) is furthermore compatible with a chamber arrangement, comprising one another corresponding to a first chamber (3), which primarily couples a utility signal to a first microphone (1), in front of the first microphone (1), and a second chamber (4) in front of the second microphone (2), and with an arrangement (g) (10) for equalization between the chambers, and a signal according to the signal from the second microphone (2) is removed from the signal. - from the first microphone (1). Method for noise cancellation according to claim 1, characterized in that a signal according to the signal from the second microphone (2) is removed from the signal from the first microphone (1) by subtracting the signal obtained directly from the second microphone (2) from the signal from the first microphone (1). Method for noise cancellation according to claim 1, characterized in that the signals from the first and the second microphone (1, 2) are passed separately for further processing, the removal of the signal according to the signal from the second microphone (2) from the signal from the the first microphone (1) occurs after processing and / or analysis of the signals. Sensor construction for measuring a utility signal on the surface of a solid, since an interference signal comes via a surrounding medium, comprising a first microphone (1) for measuring both the utility signal and the interference signal, and a similar or corresponding second microphone (2). ) for measuring only the interference signal, characterized in that it contains: a protective housing (7, 8) provided with an acoustically damping material (9) which prevents resonances and echoes, in which the first microphone (1) and the second the microphone (2) is installed at the shortest possible distance from each other and isolated from each other, to passively attenuate a noise signal to each microphone (1.2) having substantially the same power, and to enable the passively attenuated noise signal to be coupled to each other the microphone (1, 2) in the same form with substantial accuracy, as well as an arrangement for compatible connection of the interference signal (1.2) which comprises at least: a first chamber (3) which defines against the surface of the solid in front of the first microphone (1) where the transducer lies against the surface and whose main task is to connect the utility signal to the first microphone (1), a second chamber (4) formed in the housing (7, 8) and corresponding to the first chamber (3) in front of the second microphone (2) and an arrangement (10) for equalization between the chambers. Sensor construction according to claim 4, characterized in that the second chamber (4) is designed with substantially the same volume as the first chamber (3). 35 94287 Sensor structure according to claim 4 or 5, characterized in that said arrangement for compatibility of the coupling of a noise signal to the microphones (1,2) further comprises a symmetrical indentation (5) in the protective housing material (7) connecting the microphones (1, 2) acoustically to each other from opposite sides 5 to corresponding acoustic chambers (3, 4). Sensor construction according to any of claims 4-6, characterized in that the arrangement for pressure equalization between the chambers (3, 4) consists of respective thin pressure equalization heels (10) leading to the chambers (3, 4) through the outer surface of the protective housing (7). , 8) and which are equally long and at the same distance from the microphones (1,2) and are immediately adjacent to each other on the outer surface of the protective housing (7, 8). Sensor construction according to any one of claims 4-7, characterized in that the first acoustic chamber (3) is designed so that the measuring surface's ratio to the volume of the chamber is large, thereby providing an amplifying effect on the utility signal. Sensor structure according to any of claims 4-8, characterized in that the first microphone (1) simultaneously functions as a plug, ie. as part of the passive protection between the microphones (1,2), thereby preventing the utility signal 20 from reaching the second microphone (2). Sensor construction according to any of claims 4-9, characterized in that the protective housing (7, 8) consists of a multi-layered housing (8) and a supporting material (7) with good acoustic damping ability. . · 25 Sensor structure according to claim 10, characterized in that the body material (7) is a rubber material with good acoustic damping ability. Sensor structure according to claim 10 or 11, characterized in that the multilayer casing (8) consists of hardened plastic layers with inflated porous cell rubber.