DE3342928A1 - ACTIVE ACOUSTIC DAMPING DEVICE - Google Patents

Description

The present invention relates to a sound attenuation device according to the preamble of claim 1. You relates in particular to a device for attenuating a band of low sound frequencies as well individual frequencies propagating from a given source by introducing a cancellation sound .

In US Pat. No. 213,254, the multitude of are exemplary active acoustic damping arrangements with particular reference to physical systems including describes the propagation of sound in a channel. The devices and methods disclosed are simple Applicable to other acoustic cancellation environments such as closed rooms or free space. According to Particular aspects of the present invention are intended to be exemplary forms of active sound attenuation device can be described in a free space.

In limited enclosures, such as ducts, substantial reductions in sound pressure levels have presented an unsolved problem for many years. For example, in factories, the noise generated by machines and various manufacturing operations via the heating and ventilation ducts in these areas are transferred to ducts that are connected to offices and connected to other parts of the factory where low noise levels are desired. This sets poses a particular problem with low-frequency noises in the infrasound range up to 800 hertz, as it is passive Devices for attenuating such frequencies are costly, relatively ineffective and physically large, as they are impractical for use in most low frequency applications.

Since 1925 and thereafter at extremely high speed until today electronic developments have been made which use the concept of active noise damping and which represent not only a possible but also an attractive alternative to passive damping of low frequencies. The principle of so-called active damping is based on the fact that the speed of sound in air is much lower than the speed of electrical signals. In the time it takes for a sound wave to travel from a location where it is detected to a remote location where it can be attenuated, there is sufficient time to sample the propagating wave, the information it contains in an electronic circuit and generate a signal to drive a loudspeaker that introduces a 180 ° phase-shifted cancellation sound with the same amplitude as the propagating sound. While the method of active noise attenuation appears simple in concept, a review of the prior art in this area reveals that it is a complex problem to obtain good attenuation over a relatively wide band of low frequencies.

One of the first efforts in the field of active attenuators can be found in US Pat. No. 2,043,416 in FIG will. The known system is a monopole, consisting of a microphone, amplifier and loudspeaker. The microphone is on the source sound and converts it into an electrical signal that is fed to the amplifier. The one through the amplifier controlled loudspeaker is arranged behind the microphone in a place that provides the necessary time delay results and realizes a 180 ° phase inversion of the source sound. The speaker injects a mirror image of the source

Sound in the canal, so that when the source sound passes by at the location of the loudspeaker, a volume of either high or low air pressure with a phase relationship of 180 ° to the corresponding high and low air pressure of the Source sound is introduced. When the speaker is exactly in sync with the passing of the source noise, the mean of the pressure of the noise source and the loudspeaker is 0 and the noise is canceled.

An examination of the Lueg system clearly shows that the Attenuation occurs when the distance between the microphone (where the sound source is sampled) and the loudspeaker (where the cancellation sound is introduced) is such that the time delay of the electrical signal going to the amplifier is sent corresponds to 180 ° or an odd multiple "of 180 °. However, this condition only occurs for specific stationary acoustic signal that is not subject to any change over time. As a practical consequence of this the known system is only effective for a single frequency, since no means are provided for an adaptation to to achieve a phase change. This limitation of the known system results in two parameters that are important for good attenuation must be met. These concern the delay time that allows the acoustic wave to move from the point of detection to the point of attenuation and the phase that ensures that the attenuation occurs at the point of introduction of the canceling acoustic wave.

In addition to the limitation of the known system with regard to phase detection and adaptation, this results a problem with the generation of standing waves by the loudspeaker towards the microphone in front of it. Because of the standing line is the pressure of the sound field on the microphone artificially non-uniform, which means that

- ίο -

at a given frequency, the microphone can be placed in a node or belly of a standing wave. Therefore can cancel the signal that is generated by the speaker will be made too big or too small. As a result, came the sound field can be amplified by the standing waves in such a way that the resulting propagation is behind the loudspeaker is even larger than the sound produced by the source. In addition, the field of standing Waves intensify the sound pressure in the canal and more sound can pass through the walls of the canal a secondary problem is created.

In an effort to avoid the aforementioned standing wave problem and to expand the damping frequency range various active damping systems based on this system have been developed. A well-known system which is shown in Fig. 2 uses the combination of a monopole / dipole arrangement with the dipole on a wall of the Channel is arranged and the monopole is arranged in the manner shown on the opposite side of the channel is. This system was first introduced by M. Jessel and G. Mangiante and is in a work entitled "Active Sound Absorbers In An Air Duct ", JSV (1972) 23 (3,383-390). The dipole and the unipole of this system are out of phase so that they add upstream and subtract upstream, causing propagation in one direction is allowed with balanced sources. However, it has been found that those with this system achievable results are frequency-dependent and are related to half the wavelength of the dipole separation. Additionally the complexity of this system is not suitable for use in many practical applications.

To simplify this system and to achieve an improved The dipole systems according to FIGS. 3 and 4 were developed for performance. The system of FIG. 3 can be found in US Pat 4,044,203. There the monopole is removed and the phase characteristic of the dipole is changed, so that the propagating sound of both sources is added downstream and upstream towards the Microphone is compensated. The damping device according to FIG. 4 is described in US Pat. No. 4,109,108, the Microphone is placed between two loudspeakers in order to generate a minimum level at the location of the microphone, when the phase relationship between the speakers is correct. While this system is reflective and one standing wave is generated upstream of the dipole, the detector system (microphone) is not affected as it is between the speakers is arranged.

When looking at the performance of the dipole and dipole / monopole systems it has been found that each of these multi-source systems has restrictions in terms of geometry is afflicted. The physical distance between the loudspeakers and microphones creates an adaptation effect that changes the frequency the best performance and bandwidth. Although high levels of attenuation are possible, especially through the system according to FIG Fig. 3, such performance is only maximal over a relatively narrow bandwidth, on the order of approximately maximum 2 1/2 octaves achievable. Accordingly, the most recent solutions to active noise attenuation in ducts have focused on one Concentrated improvement of the first mentioned monopolar system. U.S. Patent No. 4,122,303 shows an example of a single pole system shown and described in which a relatively sophisticated electronic circuit in the basic microphone / Loudspeaker assembly is included. There will be a second one there

... pe-

Cancellation wave generated by a first electrical signal which is the primary sound wave detected by a microphone represents. The first electrical signal is combined with the second signal, which is identified by the system impulse response as a Progran m is derived from operations steps. This process represents the standard operation of an adaptive filter, as described in the following publications: "Feedback and Control Systems", foam * ε Outline Series, 1967, pages 179-185; "Adaptive Filters", Widrow, B., Aspects of Network and System Theory, (1971); "Principles and Applications of Adaptive Filters: A Tutorial Review ", McCool, J.M .; Widrow, B., Institute of Electrical Engineers, (1976). To avoid the problem of standing waves that are generated by the speaker and move upstream to the microphone, U.S. Patent 4,122,303 teaches the use of a unidirectional microphone without the liner the channel between the loudspeaker and the microphone with acoustically absorbing material is proposed. In addition, the possibility of including a circuit is suggested there to form a second fold, which is able to compensate for the feedback signal given by the upstream ones Waves is generated. As will be explained in more detail below, this so-called adaptation process is, however, time-consuming and required a time period of 5 to 30 minutes, which assumes that the signal source is substantially constant over this period remain. In addition, the control system works so slowly that in most practical cases it is manual must be switched off while the first convolution process is still effective to prevent pendulum oscillations or an oscillation to avoid between better and worse results.

It should also be noted that ^e ^ ö ^ ^there he aforementioned systems have a common problem that decreases in many practical applications its possible use or bans. In any case, the loudspeaker part of the system is placed directly in the vicinity of the duct in order to introduce the cancellation sound. It should be anticipated that in many applications the present loudspeakers will not be able to survive the temperatures, particles or foreign materials found in the vicinity of the duct or in unprotected areas outside the duct. In the duct environment, an active acoustic attenuator of the modified monopole type, which can be thought of as consisting of three separate components, which comprise a physical system, an electronic circuit and a coupling component between the physical system and the circuit, can be arranged. The physical system consists of the duct through which the sound waves travel from a given source, an acoustic mixer in series with the duct, a loudspeaker in a protective enclosure at a distance from the duct, and a waveguide that supports the loudspeaker connects to the acoustic mixer. The electronic circuit component of the active attenuator here consists of three different adaptive filters that use a modification of the Widrow-Roff LMS algorithm in a true adaptive acoustic cancellation configuration. A microphone assembly located in the channel at a location upstream of the acoustic mixer and a microphone within the acoustic mixer downstream of the waveguide provide the coupling means between the physical system and the electronic circuit.

As will be explained later, the microphone assembly detects the source sound in the duct and converts it to an electrical signal which is sent to a modified adaptive filter in the electronic circuit component. The adaptive filter generates a signal to drive the loudspeaker and generate a cancellation sound that is 180 ° out of phase with the source sound that propagates through the waveguide to the acoustic mixer where it is combined with the source sound. The microphone, which is located at a position downstream of the wave guide within the acoustic mixer _r detects the sound resulting from the combination of the source sound and the cancellation sound, and generates a signal that the error or the difference between the in the attenuation achieved by the acoustic mixer and the desired attenuation based on preset levels. This error signal is introduced into the adaptive filter, which then adjusts its signal driving the loudspeaker, so that the cancellation sound propagating in the acoustic mixer comes closer to the mirror image of the source sound. High levels of attenuation are thus achieved within the acoustic mixer and elsewhere in the far field.

In a preferred exemplary embodiment of the active acoustic damping device, an adaptive compensation filter is used in order to ensure a stable operation of an LMS algorithm ensure when creating the cancellation sound. This adaptive filter provides a phase correction and is done by hand instead of a phase correction filter must be calibrated. In an open space environment, it is advantageous to have the canceling sound generator as close as possible at the sound source generating the sound to be canceled to arrange. To help in completing the placement you want

To help the cancellation sound generator, the adaptive compensation filter should be set correctly in the active one acoustic damping device are arranged. In accordance with one aspect of the present invention, the adaptive compensation filter comes out of the cancellation loop removed from the attenuator, thereby reducing the delays in the cancellation loop. That allows a closer placement of the canceling sound generator to the one generating the sound to be canceled Source.

According to a further aspect of the invention, when using the above-described active acoustic Damping device in a free space environment several Cancellation sound generators spatially separated from the source generating the sound to be canceled in different Arranged directions of propagation of the sound.

Other objects and advantages of the invention and how it can be carried out will become apparent upon reading the following detailed description with reference to the drawing understandable. Show it:

1 is a view of a simplified version of the active acoustic damping device according to FIG of the present invention.

2 shows a partial block diagram of the digital implementation of an adaptive filter with the Widrow-Roff standard DiS algorithm.

3 is a partial block diagram of the digital implementation an adaptive filter with a modified Widrow-Roff UiS algorithm.

Fig. 4 is a block diagram of the electrical system according to the invention circuit component.

Figure 5 is a block diagram of the adaptive decoupling filter the electrical circuit component.

Figure 6 is a block diagram of the adaptive compensation filter of the electrical circuit component.

7 shows a schematic representation of the relative orientations of a sound to be canceled generating source and canceling sound sources.

FIG. 8 is a schematic illustration of the relative orientations of sound sources shown in FIG. 7 are shown, with additional scanning devices and an active acoustic sound attenuation furnishings.

9 is a block diagram of a modified form of the electrical circuit of FIGS. 4 and 4

Fig. 10-14 simulated far-field radiation curves for individual, specific positions of individual components the active acoustic sound attenuation device with respect to a sound source.

While the invention is susceptible to many modifications and alternative forms, certain illustrative embodiments will be made shown as an example in the drawing and described in detail below. However, it is understandable that it is not intended to limit the invention to the particular form disclosed, but on the contrary all modifications, equivalents, and alternatives that fall within the spirit and scope of the invention and are defined in the claims to include.

Referring to the drawings, and more particularly to FIG. 1, the active acoustic attenuator according to FIG Invention generally provided with the reference number 11. As previously explained, the active damping device 11

are considered to be composed of three different components and the explanations in this section are in primarily to the physical system component with general reference to the other components wherever it is necessary directed. The physical system comprises a channel 13 »through which sound from a predetermined Source propagates, one connected to channel 13 acoustic mixer 15, a loudspeaker 17 which is the source of the cancellation sound, and a wave guide 19, which connects the loudspeaker 17 to the acoustic mixer 15.

Although the acoustic mixer 15 is shown with a slightly larger diameter than the channel 13, is this geometric relationship is not required for the correct implementation of the active damping device 11 and shown here for the purpose of explanation only.

One of the immediately emerging advantages of the active damping device 11 over the known systems is that the loudspeaker 17 is placed at a location remote from the duct and with the acoustic mixer 15 is connected by an elongated wave guide 19, wherein the shaft guide can be provided with valve devices, not shown, in order to prevent dust particles, Corrosive material or other debris flowing through the channel 13 will damage the loudspeaker 17. According to the known active damping devices, the loudspeaker is placed directly in the duct, where the internal and external conditions that may be encountered in many use cases quickly damage it or put it out of operation. The speaker 17 is not only facing the internal environment of the channel 13 and the acoustic mixer 15 protected by the wave guide 19, but contained in a casing 21, the protects the loudspeaker 17 from the external environment of the channel 13. The envelope 21 must have high transmission losses and can also be lined with acoustically absorbent material to prevent the output signal of the loudspeaker 17 propagates in a different direction than through the wave guide 19. It will be understood that the wave guide 19 is completely separated from the channel 13 and the acoustic mixer 15; i.e., the wave guide 19 does not form a branch duct and thus no sound flow from the main duct 13 into the Wave guide 19 transferred. The wave guide 19 is solely for the purpose of transmitting the cancellation noise from the Loudspeaker 17 to the acoustic mixer 15 and to isolate and protect the loudspeaker 17 from the environment of the channel 13 is provided.

The electronic component of the active damping device 11 is shown in its simplest form in FIG. 1 for the purpose of the present explanation. A detailed description of the electronics according to the invention is given below and includes an explanation of the complete circuitry used herein. The simplest version of the electronic component of the active damping device 11 comprises an adaptive filter 23, an amplifier 25, a phase correction filter 29 and a direct current loop, which is indicated generally by the reference numeral 31. The coupling component according to the invention, which couples the physical system with the electronic system, consists of a microphone arrangement 33, which is arranged in the channel 13 upstream of the wave guide 19 for detecting the source sound, and of a microphone 35, which is in the acoustic mixer 15 is arranged downstream of the wave guide 19 for purposes that will be seen below.

In general, the active attenuator 11 operates in the cancellation mode as follows. A broadband one Noise propagates through the channel 13 and is detected by the microphone assembly 33 which is a signal which is sent to the adaptive filter 23. The adaptive filter 23 supplies an output signal for control of the loudspeaker 17, which sends a canceling sound through the wave guide 19 into the acoustic mixer 15 introduces. Because a sound wave consists of a series of compressions and expansions that deal with a given one Propagating phase and frequency, the pressure of such sound waves can be reduced or canceled by a second sound wave is generated, the compressions and expansions with the same amplitude and a phase shift of 180 ° to the primary sound waves.

The microphone 35, which is arranged downstream of the wave guide 19 in the acoustic mixer 15, detects this Degree of attenuation or cancellation of the source sound after the sound waves generated by the loudspeaker 17 with have been combined with this. The signal from the microphone 35 is sent to the adaptive filter 23 as an error signal, which actually represents an indication of the attenuation achieved within the acoustic mixer 15. That adaptive filter adjusts its output signal depending on the nature of the error signal, so that the loudspeaker 17 is activated to generate a cancellation sound, whose amplitude and 180 ° phase shift better match the Source sound is approximated.

The second primary component of the active acoustic attenuator 11 according to the invention provides the electronic control of the physical previously described System. A brief description of the prior art can help to appreciate the advances that have been made with the present electronic control circuitry. The linear summation of two equal and opposite Signals has long been viewed as a solution to electronically or acoustically generate the value zero. Different of the acoustic cancelers that have been developed to date make use of the same and use of opposite values, whereby a cancellation wave generated by a loudspeaker is limited in one Space, such as a duct, is introduced to accommodate the pressure variations produced by sound waves. which propagate through the channel from a given source · In the simplest model, this can be done through the The signal generated by the source can be viewed as a pure tone represented by a single rotating vector. The cancellation signal must follow the source signal with some maximum allowable error in order to achieve the desired attenuation to achieve. A significant problem with this solution is that

was recognized and solved by the present invention, can be seen in the fact that the permissible amplitude and phase errors must keep within tolerances only available in the very best acoustic facilities. IN THE For example, to achieve a cancellation level of 20 dB, the errors in the microphones, speakers and electronic circuits of such systems must be less than 1 dB in amplitude and must not be phase shifted of more than 6 °. As explained in more detail below, the present invention achieves address this problem using a tolerance relaxation technique based on feedback principles. Instead of Demanding accuracy for all components, the feedback system concentrates the performance requirements in a few easily controllable facilities.

One of the most recent so-called active damping systems can do the U.S. Patent No. 4,122,303 which allegedly uses an adaptive process to generate a cancellation wave capable of forming the vert zero when combined with the sound waves of a source. Examination of this system, however, reveals that the electronic circuitry is not based on a true adaptive process but contains an approximation and an error solution, whereby a series of successive assumptions or Estimates of the error signal can be made, which is determined by the difference between the desired and actual Damping is given. Under certain circumstances, the assumptions of the error signal become closer and closer to the one sought Error signal approximated according to preset values. The approximation and error method is not only disproportionate slowly (in the range between 5 and 30 minutes) but the system must finish the procedure of Can be switched off manually to prevent system oscillation or Oscillation between better and worse values

I BAD ORIGINAL

avoid. If the approach and error process in the Once the known system has started, every change takes place of the source sound in the 5 to 30 minutes required to complete the process, in addition to a system oscillation.

A basic aspect of the electronic circuit presented here is that it is an inevitable system, the real adaptive filter in contrast to the approximation and error solving of the aforementioned known device is used. This means that the electronic Circuit according to the invention automatically adjusts its own parameters and its performance according to specific Tried to optimize criteria. By using the aforementioned feedback principles, this is present here electronic circuit also not so dependent on the tolerance requirements of the acoustic equipment, as used in many known systems. "If the permissible amplitude and phase errors of the acoustic Devices in known systems are on the order of 1 dB in amplitude and 6 ° in phase, so the present electronic circuit can in fact amplitude and phase errors in the range of at least Tolerate 10 dB and 45 °. By relaxing the tolerance requirements, the installation and maintenance of the system be carried out by technicians usually trained, whereby the active acoustic damping device 11 Commercially viable in a variety of use cases.

The electronic circuit of the active damping device 11 uses a modified form of the adaptive

i ^Λ

BATH ORIGINAL

Widrow-Roff UIS algorithm in a true adaptive acoustic Override configuration. The LIiS algorithm was designed for Designed for use in signal enhancement systems, improving the signal-to-noise ratio i.e. the noise reduction achieved in the electronic circuit alone. This algorithm got sizeable modified in the present circuit to allow its operation even when the vibration or vibration Noise reduction and / or cancellation in a physical system is to be achieved with delays such as for example in an acoustic field. The modifications of the Widrow-Roff LMS algorithm keep the Advantages of the signal processing in the original algorithm and allow these advantages to be applied active acoustic cancellation problems. Other adaptive algorithms are available for solving the quadratic error function before and many of these solutions could be modified to work satisfactorily in an acoustic canceller to work. The purpose of the adaptive algorithm, which is explained in detail below, is to find an optimal or to find a near optimal solution for the cancellation filter. Other algorithms that can be used in accordance with the present teaching Modified can also realize this function and are therefore intended to be included within the scope of the present Invention to be viewed filing.

Before discussing the modification of the LMS algorithm in the adaptive filters of the present invention, let us know adaptive filter, the operation of which is controlled by an LMS standard algorithm. According to Fig. 2 is an adaptive filter is shown, which the Widrow-Roff

ORIGINAL "

LMS standard algorithm used. The basic element of a adaptive filter is known as transversal filter 53, which is indicated by dashed lines in FIG. That Transversal filter 53 can be viewed as a series of delay elements with the filter output from one weighted summation of the delayed output signals is derived. In Fig. 2 a group of η measurements S (t) sampled to form η sample measurements S (3), where j denotes Represents sampling time index. Each of the points 37 in Fig. 2 can be viewed as a source of sampled input values, where the factor Z represents a delay. Each sample 37 is assigned a corresponding weight coefficient ¥ (d) is multiplied in the multiplier 39 and the weighted measured values are input into a summer 41, an output signal y. " at the output of the transversal filter 53 to build. This output signal y. » is required with a Answer d .. compared in a summer 43 to an error signal e. to build.

The goal of the LMS algorithm that controls the operation of the transversal filter 53 is to forcibly obtain the weight coefficient in such a manner that the error signal e ,, is reduced to a minimum and the weighted Sum of the input signals is found that best corresponds to the requested response. Changes in the weight vector to achieve this goal are made along the Direction of the estimated gradient vector is made, these changes being based on the steepest descent method based on the square error area. A detailed treatment of this subject can be found in the article by Prof. Bernard Widrow with the title "Adaptive Filters"

Aspects of Network and System Theory edited by Rudolph E. Kalman and Nicholas-DeClaris. A block diagram representation of a digital implementation of the IMS algorithm is found in the right part of Fig. 2, which is included in the dashed lines and generally designated by the reference point 55. Although a digital form is shown here, it should be noted that an analog implementation of the LMS algorithm and the transversal filter can also be used.

For purposes of illustration, only two sample input values and their corresponding weighted functions are shown in the drawing. The input S (3) is sent to a multiplier 45. The error signal e. together with a scaling factor u, which controls the speed of convergence and stability of the algorithm, is input to a multiplier 45. The scaled error signal is then multiplied by the signal S (J) in the multiplier 45 and this product is introduced into a summer 49 and a unit delay 51 or into a memory C RAM (coefficient random access memory). The weight _{setting W 1} (3 + 1) is sent back to the adjustable weight 39 in accordance with the input signal S (3 + 1), the product of which then forms part of the output signal of the transversal filter 53 . The same operation is carried out for the input signal S (3-1) corresponding to the weight W ₂ (3). As previously mentioned, the LMS algorithm has proven effective in many adaptive signal processing applications where the weight setting determined by the error signal can be fed back directly to the adjustable weight corresponding to the input signal for

which it was determined essentially without delay.

In the form shown in Figure 2 are the IJiS algorithm and the transversal filter is not suitable for use in an acoustic compensation problem. The repeal an acoustic wave propagating in a channel requires an equal and opposite signal to interact with this wave to reduce the maximum pressure variations generated in the acoustic mixer 15 will. The interaction of two waves requires a finite length of the hiking route and a corresponding one Amount of time. With reference to Figs. 3 and 4 it can be observed that in the physical system the active Damping device 11 required a finite period of time to proceed to microphone assembly 33 at which the input signal is detected. In addition, there is a Time delay for the cancellation sound coming through the Loudspeaker 17 is generated and propagates through wave guide 19 to acoustic mixer 15. These delays must be estimated based on the distances in meters in the channel 13, the acoustic mixer 15 and the wave guide 19, the distance by 330 m / sec is to be divided according to the speed of sound. The expected delay for most systems is moving between a few milliseconds and a few tenths of a second. Using a scanning speed of more than 2000 samples per second results - expressed in sampling intervals - the delay by several A hundred sampling intervals.

BAD ORIGfNAL

Expressed in sampling intervals, the delay results as follows:

Delay = K (1 / scanning speed) Here, K = an integer constant.

The ¥ idrow-Roff UiS algorithm was modified to accommodate the the physical system of the active damping device 11 inherent delay to take into account, so that the Weight coefficients, which are determined in the algorithm, to the corresponding signal inputs in the transversal filter, for which the weight coefficients have been determined.

For a new signal sample S (j) the corresponding Weight coefficient can be specified as follows:

w ₁ (3-1) + J7 7 [

Here ρ means «scaling factor.

The next value of the weight can be given as follows:

Herein means: i «= tap identification

Input sampling K intervals thereafter.

'lDRlQiNAL

COPY %

As each new sample of the input signal to the transversal filter 53, it produces the product of the first weight coefficient and the last input sample which product to the product of the second weight coefficient and the penultimate input sample, etc. up to is added to the last weight coefficient times the oldest sample. The total collection of these terms makes up the output signal y (j) of the transversal filter 53. The weight coefficients are updated by the appropriate input sample and the error signal to the adherent Delays in the physical system must be taken into account. Fig. 13 shows a digital implementation of the modified one LMS algorithm according to the present invention (with K = 2), taking into account the delay explained above so that corresponding weight coefficients, input samples and error signals in the adaptive filter can be combined to produce the output signal y (j).

With reference to FIGS. 1 and 4, the electronic circuit of the active damping device 11 is now inclusive which explains adaptive filters using a modified Fidrow-Roff I2tfS algorithm. In its simplest form according to Fig. 1 can be the basic operation of the electronic implementation of the active damping device 11 can be described by the following sequence of steps:

1. Sampling the source sound wave propagating in the channel 13 (signal).

2. Delay, filtering and scaling of the signal.

BATH

3. Control of the loudspeaker 17 with the output signal from step 2 in order to reverse the signal appropriately to inject.

4. Detection of the acoustic output signal of the acoustic mixer 15 (error).

5. Setting (2) using the modified LMS algorithm.

6. Return to (1).

The adaptive filter 23 of FIG. 1 is a three-port arrangement, with two inputs and one output having. It receives an input signal from the microphone arrangement 33, which is an electrical signal accordingly the source sound is in the channel 13, and it generates an output signal for the control of the loudspeaker and the generation of a mirror-image counterpart of the source sound, this counterpart being in the acoustic Mixer 15 is introduced via the wave guide 19. The detected by the microphone 35 output signal of the acoustic Mixer 15 is introduced into adaptive filter 23 as an error signal resulting from the summation of the source sound and the mirror image counterpart in the acoustic mixer 15 results

The acoustic propagation delay associated with the physical System adhered to the active attenuator 11 must be considered a delay over the full range of interest Frequency range occur. The phase tolerance of this delay is approximately + _ 45 ° at each frequency. In the block diagram scheme of the electronic circuit in Fig. 1, a phase correction filter 29 of the second order is included

C ° ^PY jl

BATH ORIGINAL

to compensate for the acoustic resonances in the acoustic mixer 15. The characteristics of the filter 29 must be done by hand can be determined using appropriate instrumentation and based on the characteristics of the channel in a special application. As explained further below, this puncture can be realized with an adaptive filter manual calibration of the filter 29 is not necessary.

A DC loop is also included in the block diagram of the active attenuator 11 in FIG. 1 and generally provided with the reference number 31. The microphone 35 is incapable of picking up very low frequency components at the output of the acoustic mixer 15, these components in addition to a direct current component for the stable operation of the UiS algorithm of the adaptive Filters 23 are required. The DC loop 31 is included to provide the same component and the IMS algorithm has been modified to accommodate to bring about such an input signal.

4 to 6 is an advanced embodiment of the electronic component of the active damping device 11 shown. Three discrete adaptive filters are in the electronic component of the active attenuator 11, each filter performing a separate function. Before discussing the adaptive filtering processes the function of the remaining circuit elements should be mentioned. A digital realization of the adaptive algorithm is used in the embodiment of the electronic circuit of the active damping device 11. The analog /

Digital converters (A / D) shown generally in Figures 4 "through 6 are provided with the reference numeral 63, provide the sample values of the selected input signals in digital representation. The analog / digital converters 63 are 12-bit converters with successive approximation.

Most of the signal processing within the electronic The circuit is digital and the sampling speed of the analog-to-digital converter 63 is a upper limit value with regard to the permissible bandwidth of the signal and the error inputs. The limited scanning speed requires that the maximum input frequency be less than half the scan speed got to. The low-pass filters 65 of the system inputs are required to reduce the maximum input frequency to less than that Half of the analog / digital sampling speed. The output signal of the digital / analog converter 67 also has a frequency assignment problem. The output of the adaptive filters can stimulate a resonance in the system, this being any multiple of the output signal of the digital-to-analog converter 67. Hence the low pass filter 65 at the output of the digital / analog converter 67 is required to get the maximum output frequency to the point of interest Limit frequency band.

The multipliers 39, 45, 47 and 69 and the accumulators and adders 41, 43, 49 and 71 in Figures 2-6, respectively, perform the calculations required for the adaptive process by. It has been found that the multiplier / accumulator for 12 parallel bits from TRW with the Type designation TDC1OO3J or a suitable equivalent

Facility that has high demands on computing speed Fulfills.

The first of the three adaptive filters used in the electronic circuit component of the active attenuator 11 may be referred to as adaptive cancellation filter 22 which is shown in Figs. The adaptive cancellation filter 23 comprises a modified LMS algorithm, to control the operation of its transversal filter in the manner previously explained. The basic filter functions of the adaptive cancellation filter 23, which include the phase response, the amplitude and the delay realized by the transversal filter which is a non-recursive finite impulse response filter. The filter realization is digital in nature, with both input waveform and tap weighting in a random access memory (RAM) are stored. As mentioned earlier, the operation of the transversal filter can be described by generating the product of the first weighting obtained from the LMS algorithm and the last input sample plus the second weight from the algorithm times the penultimate input sample, and so on, to the last weighting times the earliest sample summed up is. In equation form, this can be stated as follows:

Output «= / S ₁ W ₁ (3)

COPY

Herein mean:

S. = sampled input signal corresponding to the

Tap i
W. «= weighting of the tap ii = tap identification

I = number of taps

The adaptive cancellation filter 23 can be used as a transversal filter can be viewed when the weighting settings are retained. The operation of the adaptive cancellation filter 23 determines the required values for the weightings of the filter sought (i.e. the optimal filter conditions) adaptive devices and it then realizes the filter function sought by using the weightings determined, in the transversal filter. The purpose of the modified LMS algorithm is to control the operation of the transversal filter control by adapting the filter function of the adaptive cancellation filter 23 to the desired filter function.

The error signal detected by the microphone 35, which to the adaptive cancellation filter 23 is sent as the summation of the source pressure wave detected by the microphone 33, delayed by the acoustic running time from this point to the microphone 35, and that generated by the loudspeaker 17 Cancellation wave, delayed by the running time over the Waveguide 19 and acoustic mixer 15 to microphone 35 can be seen. When it comes to physical positioning the microphone arrangement 33 "of the loudspeaker 17" of the wave guide 19 and the microphone 35 with respect to each other in the channel 13 and in the acoustic mixer 15, the total delay Operators from the microphone assembly 33 to the

Microphone 35 greater than the total delay or length from speaker 17 to microphone 35 plus the delay through the associated low-pass filter 65. In equation form, this relationship can be expressed as follows (see Fig. 1):

Herein mean:

^. min = shortest wavelength of interest

L = distance between the microphone arrangement 33 and the microphone 35

d .. = distance from the wave guide 19 to the microphone 35

d ₂ = distance from acoustic mixer 15 (ie in alignment with microphone 35) to loudspeaker 17

D _f = delay associated with the low-pass filters 65

c = speed of sound.

The delay D _{f of} the low-pass filter 65 can be determined by the cut-off frequency F_. _Q __ be expressed. A fourth order low pass filter 65 has a delay of 45 for each pole of the cutoff frequency and a good filter design approximates a constant delay filter. Therefore, the delay for a fourth order filter _{can be approximated to 1 / 2Fg ren2} and D ^ can be considered to be approximately equal to 1 / FSS _Renz be considered for two groups of fourth order filters in series.

The second adaptive filter in the electrical circuit component of the active attenuator 11 can be referred to as an adaptive decoupling filter 75, which is shown in FIG and 5 is shown. An important problem associated with most active attenuators is generation of standing waves or mechanical vibrations of the channel due to the cancellation caused by the loudspeaker is introduced and propagates in the direction of the microphone or the microphone arrangement, the source sound recorded. This coupling tends to improve the assessment of the signal detected by the microphone of the downstream affecting propagating source noise and reducing the effectiveness of the cancellation system. Although Unidirectional microphones have been proposed to avoid this problem, they are alone not effective enough to avoid this limitation inherent in the system.

The present adaptive decoupling filter 75 solves this problem as follows. 5 controls a broadband noise source 76 the cancellation loudspeaker 17 and the adaptive decoupling filter 75. Before the start of the System, the noise source 76 is automatically adjusted to operate the channel 13 at a sound level that is higher than the sound level of the source and is required to cancel the expected source noise. The adaptive Process reduces the output signal of the error summer and adapts the transfer function of the transversal filter the adaptive decoupling filter 75 to the acoustic coupling present in the channel 13, at which point in time the Noise source 76 is complete. When the system starts, the adaptive decoupling filter 75 controls the loudspeaker

at. The components of the loudspeaker control which appear in the output signal of the microphone arrangement 33 are removed by subtracting an equal and opposite signal in the summer 71. The subtraction process is implemented in the digital domain of the adaptive decoupling filter 75 and in the summer 71. Although this method does not completely eliminate the influence of the output signal from the loudspeaker 17 on the microphone arrangement 33, a sufficient reduction can be achieved in most applications.

4 and 6, the third adaptive filter is the Electronic circuit component shown in the active damping device 11, which as an adaptive compensation filter 79 are designated lenn. As previously related mentioned with the explanation of the basic electronic component according to FIG. 1, the compensation device provided in series between the output signal of the adaptive cancellation filter 23 and the input of the error signal to ensure stable operation of the modified LMS algorithm. During this compensation by a phase correction filter 29 of the second order in FIG. 1

it was noted that the manual calibration of the filter 29 can be accomplished by an adaptive filter. ^: The adaptive compensation filter 79 performs this function.

According to FIG. 6, the adaptive compensation _{filter 79 f of} the loudspeakers 17, the wave guide 19 #, the acoustic mixer 15 and the microphone 35 are connected in series with one another and in parallel with the broadband delay circuit consisting of the delay or the memory 78. Before starting the

COP * ■ ÖAD ORIGINAL

- .. «,. ¹ V ■■ '■

System and after completing the previously described decoupling process the noise source 76 is activated in order to control the channel 13 and the acoustic mixer 15. the Summing up the parallel signal paths gives the error input signal to the adaptive compensation filter 79. The adaptive Process matches the overall transfer function of the series circuit to the real delay of the noise source 76 by delay 78. In this way, the weightings for the filter you are looking for are generated, and that by the adaptive cancellation filter 23 received error signal is within the correct phase tolerance to ensure stable operation to ensure.

The delay circuit 78 in FIG. 6 delays the signal from the noise source 76 by an integral multiple the input sampling intervals (K). The modification of the LMS algorithm in adaptive filter 23 in Fig. 4 provides a delay of K sampling intervals in the calculation for the next value of W ^ the value of K will be greater as the acoustic delay from the loudspeaker 17 to the signal input of the adaptive compensation filter 79 in FIG Fig. 6. The value of K can be set large so that one value can be used in most applications. Adjusting the delay from the output to the error input of the adaptive cancellation filter 23 in FIG. 4 to the shift of the sampled signal that is used in the calculation of the value is used for the next weighting, ensures the stability of the modified LMS algorithm.

When using the active acoustic damping device In an open space environment, the general term is the reduction of sound levels generated by sound waves by emitting sound waves from another source with a phase relationship to the original source, that the sound pressure in the area of interest is substantially reduced is not new. For example were Attempts to reduce sound levels in the ears of a person working in a noisy environment shortly after made during World War II. Measured against the technical level of electronics at the time, the successes were relative low and the equipment bulky. As pointed out above, many overtures have been made to systems to develop that allow the reduction of sound propagation in ducts. A general discussion of the The status of this acoustic damping area, including the free space situation, can be "Active Attenuation of Noise - The State of the Art "by Glenn E. Varnaka, Noise Control Engineering (May-June 1982).

Before describing the use of active acoustic attenuators in a free space environment, some theoretical aspects of source interference in free space should be considered. For a simple monopole sound source in an unlimited space, the intensity I _x can be expressed in the radial direction by:

I ₁ . »S ² £ c / 2 / k / 4 TrJ ² (5)

In this expression, r is the distance from the source, k is the volume number (equal to jP-) ₉ $ is the density of the medium, c is the speed of sound and S is the source strength (the mean amplitude of the surface vibration speed multiplied by the source surface).

For two sources S ^, S ₂ , each of which has the same S with respect to the amplitude but with an opposite sign, the sound intensity can be approximately described by:

I _r * S ² $ c / 2 / k / 2 TrJ ² sin ² (kd cos 0) (6)

As shown in Fig. 7, both sound sources are through separated by a distance of 2d and the angle 0 is measured from the line connecting the two sources. In the derived expression for the sound intensity for two monopoles, r is much larger than d (i.e., the expression is for the far field). It can be seen from the two-monopole equation that the cancellation of the sound is most effective along the axis of symmetry between the two sources, where β equals 90 °.

It can be shown that zeros (cancellations) are generated where

cos 0 = η Ä / 2d η «0, 1, 2 ... (7)

In this expression } L is the wavelength of each frequency of the sound and cos δ is of course less than or equal to one. Because cos θ is less than or equal to one, are

Values of η for the n ^ / 2d less than or equal to one are permissible. Therefore, for capital d versus JV. (X / d <^ C1) the maximum permissible value for η must be high, which means that the radiation diagram consists of many loops (lobes). The zero at θ «0 is relatively close and the area with reduced sound pressure level is small. Just when the radiation into the side loops has been reduced by absorption, this source arrangement is quite ineffective and impractical. The closer the sources are, the further the zero (cancellation) near Q "If / 2. When d // l or kd becomes small, the two monopoles can be treated as dipoles. For a dipole, the approximate one is Sound intensity:

I ₁ . ~ fs ² y c / 2J [2ak ^Z / hTr] ² cos ² θ (8)

As can be seen from formula (8), in order to achieve effective sound reduction by an additional source which emits in opposite phase, the position of this source must be as close as possible to the original source. The distance 2d between the sources must be such that the condition for a dipole, k (2d) ** 2ψ (2d // I) ^ d, must be fulfilled. The higher the frequency, the smaller the distance must be.

Because of the physical size of the sources, it is more difficult to meet these conditions at high frequencies. In this higher frequency region, however, any envelopes that create the free space environment offer tim envelopes, usually good insertion loss. On the other

On the other hand, the low frequency range is usually more difficult to attenuate passively, so that active sound control systems can be used here.

The total emitted sound power can be determined from the double integral of I _r · dS over a spherical surface that surrounds the source or sources. In relation to the sound power resulting from a single monopoly, the sound power for a double monopoly is twice as great if they are far apart. In the case of dipoles, however, the total radiated power is less than that for two monopoles. The exact amount of total radiated power depends on the space between the sound sources and the frequency of the sound generated.

There are methods for calculating the far-field radiation characteristics by computers using near field techniques. This can be the article, for example by Robert D. Marciniak, "Near Field, Underwater Measurement System" in J. Acoust. Soc. Am., 66 (4), October 1979 ». This article describes an acoustic near-field measurement technique for precise calculation the far-field radiation characteristics of sound transducers by computer. The analysis was based on a development of a form of the Heimholtz integral which is a Green function used, which disappears over the integrated surface, with only the knowledge of the acoustic Near-field pressure is required to calculate the far-field radiation using a computer.

From the theoretical considerations of far-field calculations and taking into account the source field, for example From Figure 7, it can be seen that a first sound source and a second sound source, which is a mirror image Can be canceling sound source, generate a far-field intensity model, which in the main can be calculated As discussed earlier, it occurs when the two sound sources get closer and closer to each other be moved to a more and more complete suspension. Using other techniques such as those highlighted in Marciniak's article, the far-field effects of more complex near-field fields can as well be calculated. The active acoustic damping device described earlier is partially corresponding in FIG to use such near-field fields in order to achieve good sound cancellation and / or to generate reduction in some or all areas of the far field. Correct near-field positioning for a variety of cancellation sources can either be empirical or by use of the above-mentioned computation techniques can be achieved.

As shown, for example, in FIG. 8, an active acoustic damping device 11 has a first canceling sound generator such as a loudspeaker S ^ and a second canceling sound generator S _c2 . Both canceling _{sound generators} S q1 and S _c2 are operated by the common output of the circuit of the active acoustic damping device. The damping device 11 also has first scanning means, such as a microphone Aj, for detecting the oscillations (vibrations) from a source S generating the cancellation sound. A two-

The scanner (sensor), such as a microphone M ₂ , detects the voice of the vibrations from the source S and the cancellation _{sound generators S ^ and S c2} to generate an error input to the circuit of the attenuator 11. The single cancellation loop of the attenuator can, for example, adapt the average acoustic delay from the generators to the fault sensor. In the arrangement shown, the microphone is substantially coincident with the source S of the sound to be canceled and the three sound sources and the microphone M ₂ are arranged in a straight line.

The electrical circuit portion of that shown in FIG active acoustic damping device 11 can essentially correspond to that shown in FIG. It was however, stated "that certain changes in the circuit, as shown in Fig. 9, result in improved performance of the damper. In 9, the adaptive compensation filter 79 * has been removed from the cancellation loop and placed in series in the error signal path from the error microphone to the adaptive cancellation filter 23 '. The adaptive compensation filter 79 * is initialized as shown in Fig. 6 and then in this location in the circuit during operation of the attenuator circuit. The elimination of the adaptive compensation filter out of the cancel loop eliminates some of the delays that exist between the recording microphone 33 and the Override speaker 17 are introduced. To do this, the override speaker can be closer to the recording microphone and thus arranged to the source sound.

Figures 10-14 represent a series of far field radiation curves Simulations of dual cancellation sources acting on a single point source. In In each case, the frequencies are measured in Hertz and all distances in meters (or feet). The curves represent a point source that generates the sound to be canceled and is located in the center of the grid system. The strenght of the source is arbitrarily + 30 dB. All 10 dB are through concentric around the noise source in the center of the grille arranged dotted circles marked. The +30 dB circle is the outermost circle of the three and is insignificant shown stronger to indicate the strength of the sound. The orientation of the sources of cancellation and detectors at Each simulation essentially corresponds to the orientations shown in FIG. Two coordinated sources of repeal are symmetrical on each side of the sound source with a distance on the one that is also shared with the error sensor arranged horizontal axis. The source sensor is mainly used in the place of the original sound source arranged assumed. As disclosed, the system is using a single adaptive acoustic cancellation loop introduced for the acoustic damping device 11.

The full curve is the combination of the far-field radiation model the source and two override generators. The curve diagram (plot) only shows the far-field noise reduction in a 360 ° sweep with the center around the noise source. The diagram has no information about to take the distance. Where the solid model crosses the heavily dotted circle is the combined one Field strength greater than that of the original source or, with

In other words, the cancellation arrangement does not reduce the size of the emitted sound, but increases it. The advantage by greatly reducing the emitted sound level in a large "cone" overcomes any in many cases Problems caused by the sound rising in other directions.

A comparison of Figures 10 and 12 and also a comparison of Figures 11 and 13 shows the improved performance at lower frequencies for the same source relative to the cancellation generator spacing. One will remember that the narrower the general proposal was Space between the sound sources or the longer the wavelength (the lower the frequency), the better the performance in a two-source arrangement. The same seems to be true for three linearly arranged sources of the simulated radiation curves. Vie continue It can be seen from FIG. 14 if the distance between the source generating the sound to be canceled and the cancellation sound source becomes significantly larger, the far-field radiation curve becomes more complex. Again, as in 14, just in the illustrated arrangement, a substantial "cone" of raised or substantial generated attenuated sound.

In the radiation curve diagrams shown, the error microphone is far from the source of the one to be canceled Sound removed in the direction of the desired cancellation. In the curve diagrams this is to the right of the source the horizontal axis. Although the radiation curve diagrams are shown for individual frequencies, the active acoustic damping device of the present

Invention a broadband reduction of acoustic vibrations or noise. That through the damping device The far-field radiation model generated would be based on the effective attenuation spectrum, depending on the spectral components of noise, vary.

An active acoustic damping device for damping a relatively wide band of low sound frequency from a given source by introducing cancellation sound with the mirror image amplitude and phase characteristics of the source sound. the active acoustic damping device comprises one or more loudspeakers at a distant location of the source generating the sound to be canceled is (are) arranged. Every speaker will operated by an electronic circuit that has an adaptive filter using a modified LMS algorithm includes that dominates the operation of its transversal filter, which the given delays associated with the propagation of acoustic waves and other delays in the system.

Claims (8)

Or. FfiS ^ Pj Pf. P.O. Box 7D0i-! 5
iU?; v; Jx ^ hoUrafie 27 ; c..vjn iOSil) 6170T9 November 24, 1983 FuKk / Ra. Lord Corporation, Erie, PA 16512 / U.S.A. Active acoustic damping device Claims .j Device for canceling the vibrations of a Source, characterized by: a first sensor device for detecting the vibrations, the first sensor device first generating electrical signals representative of the amplitude and phase characteristics of the vibrations; a cancellation means for generating canceling vibrations to be combined with the vibrations from the source having a first cancellation vibration generator spaced from the source in a first direction of propagation of the vibration, and a second. Suspension vibration generator that with distance located from the source in a second direction of propagation of the vibration; a second sensor device at a point at a distance from the first sensor device in the direction of the Propagation of the vibrations, wherein the second sensor means the summation of the vibrations from the Source and the canceller detected and second generates electrical signals that determine the amplitude and Represent the phase characteristic of the summation; and an adaptive filter for controlling the cancellation device to detect the cancellation vibrations, the adaptive filter being a transversal filter, using an IÄS algorithm to generate a Adaptation to the delay in the propagation of the vibrations from the first sensor device to the second sensor means and delaying the propagation of the cancellation vibrations from the first and the second canceling vibration generator to the combining point of the canceling vibrations with the vibrations the source, wherein the adaptive filter successively the first electrical signals of the first sensor device delays, filters and scales to provide appropriate output signals for triggering the override device to generate and then inevitably based on the delay, filtering and scaling to adjust to the first electrical signals of the second sensor device and revised accordingly To provide output signals for the control of the cancellation device and cancellation vibrations so that their amplitude and phase characteristics are a reflection of the vibrations of the source within given limits. 2. Device according to claim 1, characterized in that the first cancellation vibration generator, the second cancellation vibration generator and the second sensor means are aligned in a linear field. 3. Device according to claim 2, characterized in that the first sensor device is substantially at the Vibration source is arranged. 4. Device according to claim 1 or 3, characterized in that that the second sensor device at a location which is at a distance from the first sensor device is arranged in the vibration propagation direction in which the cancellation is desired. 5. Device for the cancellation of vibrations from a source, characterized by: a first sensor device for detecting the vibrations, the first sensor device first generating electrical signals representative of the amplitude and phase characteristics of the vibrations; a cancellation device for generating cancellation vibrations to combine with the vibrations of the source; a second sensor device at a location spaced from the first sensor device in one direction the propagation of the vibrations, wherein the second sensor device the summation of the vibrations the source and the cancellation device detects and generates second electrical signals that the amplitude and represent phase characteristics of the summation; and an adaptive filter for driving the cancellation device to generate the cancellation vibrations, the adaptive filter having a transversal filter using an IMS algorithm to adapt to the delay in the propagation of the vibra- functions from the first sensor device to the second sensor device and the delay in the propagation of the canceling vibrations from the canceling means to the combining point of the canceling vibrations with the vibration of the source, the adaptive filter successively the first electrical Signals from the first sensor device are delayed, filtered and scaled to produce corresponding output signals to generate the control of the cancellation device and then inevitably the delay, filtering and adjust scaling based on the second electrical signals of the second sensor device and to provide corresponding revised output signals for actuating the override device and generate cancellation vibrations so that their amplitude and phase characteristics are mirror images the vibration of the source is within predetermined limits, the adaptive filter still being an adaptive one Compensation filter, the devices for the inevitable adjustment of the phase characteristics of the second electrical signal generated by the second sensor device for introduction into the adaptive Compensation filter, to ensure stable operation, has and in series between the second Sensor device and an input of the adaptive cancellation filter is connected, has. 6. Control for a device for canceling vibrations from a source with an adaptive cancellation filter, which is a transversal filter, which is a modified LMS algorithm used to adjust acoustic delays includes, the adaptive Cancellation filter has an input that is connected to a first sensor device for detecting vibrations from a vibration generating source to be canceled, an output connected to a cancellation means to generate cancellation vibrations to combine with the vibrations from the source, is connected, and has an error input, and an adaptive compensation filter that is between a second sensor device, which is at a spatially separate location and in a direction of propagation the vibrations to be canceled and the error input of the adaptive cancellation filter 1st switched, the adaptive compensation filter being capable, the phase characteristics inevitably of the output of the second sensor means to ensure stable operation of the adaptive cancellation filter to ensure. 7. Device for the cancellation of the vibration in a physical system, characterized by: a first sensor device for detecting the vibration, the first sensor device first generating electrical signals representative of the amplitude and phase characteristics of the vibration; a cancellation means for generating a cancellation vibration to be combined with the vibration the physical system; a second sensor device at a point at a distance from the first sensor device in the direction of the Propagation of vibration, with the physical system and the dissipation vibration second electrical Generates signals that determine the amplitude and phase Show the characteristic of the summation; and an adaptive filter using a modified LMS algorithm and a transversal filter, the adaptive filter successively the first electrical Signals from the first sensor device are delayed, filtered and scaled to produce corresponding output signals to generate for the control of the cancellation device and then inevitably the delay, filtering and scaling based on the second set electrical signals from the second sensor device and corresponding revised output signals to deliver for the control of the Aufhebimgseinrichtung and a cancellation vibration so to produce that their amplitude and phase characteristics are a reflection of the vibration of the physical System within given limits. 8. Device according to claim 7 »characterized in that the modified LMS algorithm means for adaptation including delays in propagating the vibration from the first sensor means to the second sensor means and the propagation of the cancellation vibration from the cancellation means to the combining point of the cancellation vibration with the vibration of the physical system cling.