WO2008006404A2

WO2008006404A2 - Method for operating an active noise canceling system

Info

Publication number: WO2008006404A2
Application number: PCT/EP2006/064173
Authority: WO
Inventors: Harry Bachmann; Sigmund Eggenberger
Original assignee: Anocsys Ag
Priority date: 2006-07-13
Filing date: 2006-07-13
Publication date: 2008-01-17
Also published as: WO2008006404A3

Abstract

The invention relates to a method for operating an active noise canceling system which comprises a transmission path that consists of at least one actuator unit (3), a room (11) and at least one sensor unit (4). The transmission path is reproduced by eliminating or at least reducing noises that reach the room (11). The invention is characterized in that a time delay (T) caused by the transmission properties of the room (11), the actuator unit (3) and the sensor unit (4) is determined and the determined time delay (T) is taken into consideration for reproduction. As it is considerably less time-consuming to determine the delay time (T) than all transmission properties during operation, computation time is reduced, thereby also reducing the costs for the required hardware.

Description

Method for operating an active noise reduction system

The present invention relates to a method for

Operation of an active noise reduction system according to the preamble of claim 1.

Noise sources are increasingly perceived as environmental pollution and are regarded as a reduction of

Life quality. However, as noise sources are often unavoidable, noise reduction methods based on the principle of wave cancellation have already been proposed.

The principle of Active Noise Canceling (ANC) is based on the cancellation of sound waves due to interference. These interferences are generated by one or more electro-acoustic transducers, such as loudspeakers. The signal radiated by the electro-acoustic transducers is calculated by means of a suitable algorithm and continuously corrected. The basis for the calculation of the signal to be radiated by the electro-acoustic transducers is the information supplied by one or more sensors. These are on the one hand information about the nature of the signal to be minimized. For this purpose, for example, a microphone can be used which detects the noise to be minimized. On the other hand, but also information about the remaining residual signal needed. Again, microphones can be used.

The basic principle applied with active noise reduction was Paul Lueg in a patent document from the year 1935 and the disclosure number AT-141 998 B described. By this document is disclosed how noise in a tube can be extinguished by generating a signal with opposite phase.

An active noise reduction algorithm requires information from at least one sensor (for example, a microphone) that determines the residual error. Depending on the application and algorithm used, another sensor is provided that provides information about the nature of the signal to be minimized. Further, an adaptive noise reduction system requires one or more actuators (for example in the form of loudspeakers) to output the correction signal. The information from the sensors must be converted by an analog / digital converter into a suitable format. After processing by the algorithm, the signal is reconverted from a digital to analog converter and transmitted to the actuators.

One problem associated with active noise reduction through adaptive processes is the fact that the adaptive process takes some time to process the signals; All signals to be processed are thus subject to one through the adaptive process caused time delay. This time delay is a limitation. Thus, it is decisive whether a signal can even be processed by the adaptive process at all.

With active noise reduction or "Active Noise Canceling", hereinafter called ANC, the stability of the algorithm used is a decisive factor. At present, a number of specific algorithms are used, such as LMS (Least Mean Square) or the related Fx-LMS algorithm (Filtered-x-Algorithm). In particular, the Fx algorithms have good stability and can therefore be used well in ANC systems. The prefix "Fx" indicates the reproduction of the so-called "secondary path", which includes the characteristics of the actuators, sensors, amplifiers, analog / digital converters, digital / analog converters and the transmission path as well as all other influences includes the signal to be transmitted. The "secondary path" is also referred to below as "component influence".

Below are some common methods for determining the secondary path (component influence) are described and their weaknesses shown.

A complete ANC system with integrated secondary path is described, inter alia, in the document "A New Structure For Feed Forward Active Noise Control Systems With Online Secondary Path Modeling", by the authors - A -

Muhammad Tahir Akthar, Masahide Abe and Masayuki Kawamat were published in Kyoto in September 2003 at the International Workshop on Acoustic Echo and Noise Control (IWAENC2003).

This document describes offline modeling of the secondary path (component influence). The known method for determining the secondary path is therefore referred to as "offline modeling", since the properties of the secondary path are determined beforehand, ie while the system is not in operation.

Once the component influence (secondary-path properties) is determined using white noise, a filter that mimics these properties is included in the calculation by the LMS algorithm.

Another method for determining the secondary path during operation is described by Sen M. Kuo in US Pat. No. 5,940,519. The idea with this method is to add a signal in addition to the noise that is to be canceled, which signal is then correlated with the microphone signal. Due to the change of this signal, the properties of the secondary path (component influence) can be determined. The disadvantage of this method is that this additional signal is always present. The known methods for determining the secondary path (component influence) have the property that the change in the time delay which occurs between the actuator and the sensor, for the calculation of the component influence (secondary-path) is ignored. However, since the change of the time delay is an essential property of the secondary path, neglecting a change of the time delay in the modeling of the secondary influence affects the efficiency and the stability of the whole system. In the case of a change in the environmental parameters, such as the temperature, in fact, the duration of the signal and thus the time delay change. As the signal's runtime becomes smaller, the algorithm's delay in the model of the secondary path makes the algorithm too slow to deliver a satisfactory result. As a result, poorer damping properties and, in extreme cases, an unstable system can arise.

The present invention is therefore based on the object to provide a method which does not have the disadvantages mentioned above.

This object is achieved by the features stated in the characterizing part of claim 1. Advantageous embodiments are specified in further claims.

The invention relates to a method for operating an active noise reduction system comprising a transmission path consisting of at least one Actuator, a space and at least one sensor unit, wherein a replica of the transmission path is made such that in the room reaching noise is eliminated or at least reduced. The invention is characterized in that a time delay resulting from transmission characteristics of the space and / or actuator unit and / or sensor unit is determined and that the determined time delay is taken into account in the simulation of the transmission path. This ensures that with changing characteristics of the transmission line, the active noise reduction system does not become unstable. Of the changing transmission characteristics in particular the time delay is of paramount importance, as it is strongly dependent on temperature changes and at the same time exerts a great influence on the stability.

A further embodiment of the method according to the invention is characterized in that the time delay is determined from a residual signal which is generated from an output signal of the transmission path and an output signal of the simulated transmission path and either an input signal of the transmission path or a signal derived therefrom.

A more specific embodiment is that the determination of the time delay in the time domain. Another embodiment is that the time delay is determined from a convolution of the residual signal with an input signal of the transmission path or a signal derived therefrom.

A still further embodiment variant is that spectral transmission properties of space, actuator unit and sensor unit are determined in the frequency domain and that the spectral transmission characteristics are taken into account in the simulation of the transmission path.

A still further embodiment variant is that an amplitude spectrum is determined from the spectral transmission characteristics and that the amplitude spectrum is taken into account in the simulation of the transmission path.

A still further embodiment variant is that at least one of the following properties is or are determined continuously during operation of the active noise reduction system:

- Time Delay;

- amplitude spectrum; - Spectral transmission properties.

A further embodiment variant is that at least one of the following properties is determined when a change in these properties must be expected: Time delay, ^•

- amplitude spectrum;

- Spectral transmission properties.

A further embodiment variant is that the temperature in the room is determined and that a redetermination of the transmission properties, namely time delay, amplitude spectrum and / or spectral transmission properties, is triggered when the temperature changes above a predetermined value.

Finally, there is another embodiment of the invention is that newly determined

Transmission properties lie within a predetermined range around existing or prior to the start of the active noise reduction system determined transmission characteristics.

The present invention will be further explained by means of exemplary embodiments with reference to drawings below. Show it:

Fig. 1 is a simplified block diagram of an active

Noise reduction system using a known "off-line modeling" method to determine the

Secondary path is used

2 is a simplified block diagram of an active one

Noise reduction system in operation, the known "offline" method was used to determine the secondary path,

3 is a simplified block diagram of an active noise reduction system employing a secondary path determination method of the invention during operation.

4 shows a further simplified block diagram of an active noise reduction system according to the invention, in which a method according to the invention for determining the secondary path during operation is used,

Fig. 5 is a perspective view of a room in which an active noise reduction system is used, and

6 and 7 show two courses of an impulse response, with reference to which a further embodiment for determining the time delay is explained.

Fig. 1 shows, in a simplified representation, a block diagram of a noise reduction system, with which the so-called Secondary-Path (component influence) is determined. The noise reduction system consists of a signal source 1, which generates a reference signal x (n) 2. The reference signal x (n) is emitted by an actuator unit 3 into a room 11 and received there by a sensor unit 4. The received signal is influenced in the example shown by the properties of the actuator unit 3, the sensor unit 4 and the space 11, which is referred to as secondary-path or component influence. The signal emitted by the actuator unit 3 moves in the space 11 at a predetermined speed, which depends in particular on the air humidity and the temperature of the air. The transit time (or the time delay) which the reference signal x (n) sent by the actuator unit 3 to the sensor 4 requires is designated T.

The reference signal x (n) is further supplied to a filter unit 9 with an adjustable transfer function, which is adjusted by means of an adaptive unit 10 such that a residual signal e (n), which is determined from the output signals of the sensor unit 4 and the filter unit 9, is minimal. This comes in the

Filter unit 9 is an adaptive algorithm used. To form the residual signal e (n), an addition unit 7 is provided after the sensor unit 4, in which the difference between the output signals of the sensor unit 4 and the filter unit 9 is formed

Output of the filter unit 9 is inverted before the addition unit 7 is applied. The filter unit 9 thus describes in an optimum case the properties of the actuator unit 3, the sensor unit 4 and the space 11. The influence of room 11 should be as small as possible in the calculations. It is therefore advisable to place the actuator unit 3 as close as possible to the sensor unit 4. However, the influences of the space 11 and the time delay T required by the reference signal x (n) can not be completely ruled out. The secondary path determined with such a method is therefore subject to an error. Accordingly, the quality of the active noise reduction system based on such secondary-path adjustment is reduced.

The method used is also referred to as "offline modeling". The parameters of the secondary path are thereby preferably determined in a time period during which the system does not make an active noise reduction, i. the noise reduction system, only the known reference signal x (n) of the signal source 1 is supplied. As a reference signal x (n), for example, a white noise is used.

As soon as the residual signal e (n) at the output of the noise reduction system is minimal, the transfer function of the filter unit 9 completely describes the secondary path, ie the influence of the components, for example in the form of a frequency-dependent function. In one embodiment, however, it is not the entire frequency-dependent function that is of primary interest, but primarily the time delay T, which experiences a signal from the actuator unit 3 to the sensor unit 4. by virtue of the "off-line" determination of the time delay T is an indication of the range in which the time delay T is most likely to be during the operation of the noise reduction system. Thus, the conditions are created to continuously determine the changing component influence during normal operation of the noise reduction system and make appropriate corrections. The method used in this regard will be explained with reference to FIG. 2.

Fig. 2 shows, again in a simplified representation, a block diagram of a known active noise reduction system in which the parameters of the secondary path, in particular the time delay T, in advance, i. "off-line" or before the normal operation of the active noise reduction system, have been determined. The determination of the parameters of the secondary path, i. the transfer function of the secondary path, in particular the time delay T, is carried out, for example, according to the explanations in connection with FIG.

A sensor unit detects the unknown signal x (n) to be minimized, which is applied to the transmission path 6 with the transfer function H (n). Part of the

Secondary path, namely in particular the properties of the actuator units and sensor units which are known in advance or known by the manufacturer, and the acoustic path between the actuator unit and the sensor unit is denoted by 13 and has a transfer function S (n) on. S (n) is an estimate of the characteristics of the secondary path determined according to the "off-line" method, wherein the parameters of the actuator unit and the sensor unit, in addition to the room characteristics and the temperature at the time of determining the secondary path also included in the esteemed secondary path S (n).

The adaptive algorithm processed in the adaptive unit 10 provides the transfer function of

Filter unit 9 so that the output of the addition unit 7 -. the resulting residual signal e (n) - is minimized.

The time delay T between the actuator unit and

Sensor unit also influences the stability of the overall system during the processing of the input signal x (n).

Since the instantaneous time delay T is not continually adjusted to the current conditions, the whole system is less stable when, for example, the temperature in the room changes. Accordingly, the performance deteriorates or the active noise reduction system becomes unstable.

3 shows, in a simplified representation, a block diagram of an inventive active noise reduction system, in which a method for determining the properties of the secondary path under Inclusion of the occurring and time-varying time delay T is used. In the illustrated and described embodiment of the present invention, a delay time difference ΔT is determined which corresponds to the deviation from the delay time T determined by the "off-line" method. Variants in which each of the actual delay time is determined directly, are also conceivable.

Block 14 includes an estimate of the secondary-path characteristics in the frequency domain, where signal delays - such as the delay time - are not included in this estimate because only the amplitude spectrum is considered. The delay time T of the estimated secondary path S (n) is set in the variable time delay element 18, wherein the adjustment of the time delay T is ongoing, i. during operation of the active noise reduction system. The interaction of the blocks 14 and 18 thus gives a complete estimate of the transfer function of the secondary path, including the occurring delay time of the signal.

In one embodiment variant, it is provided that the block 14 representing the estimated secondary path is not provided, which means that the amplitude spectrum of the secondary path is not taken into account, but only the current time delay. It has been shown that the current time delay by far the has the greatest influence on a possible instability of the active noise reduction system. With the consideration of the constantly changing delay time during the operation of the active noise reduction system, ie with the determination and

Using the current delay time, there is the possibility that the active noise reduction system remains extremely stable. The following explanations, which in turn relate to the block diagram with the block 14, are analogous to the

Execution variant without block 14 applicable and valid.

Block 13 is the secondary path S (n) including the delay time T. The transmission path with the transfer function H (n) is again denoted by 6.

The input signal x (n) is supplied to the transmission link 6, the estimated secondary path (block 14), a switching unit 17 and the filter unit 9. The filter unit 9 is set by the adaptive unit 10, in which an adaptive algorithm is executed, wherein the filter unit 9, which may be in particular of the type transverse FIR filter, recursive IIR filter or lattice filter, set such that the residual signal e (n + a) designated output signal of the addition unit 7 is minimal.

Either the output signal y of the filter unit 9 or the input signal x (n) is selected by means of the switching unit 17 and used as the signal f (n), which is supplied to the correlation unit 16. The switching unit 17 can be set automatically by a user, for example, manually or by means of a control unit (not shown in FIG. 3). This is a choice to choose the way of determining the current delay time. So there are two variants that can be selected on the switch position in the switching unit 17. However, it is also conceivable that one or the other embodiment variant is implemented, so that no switching unit is necessary and the determination of the

Delay time is limited to the realized embodiment. Of course, such a realization does not allow switching to another variant.

In the correlation unit 16, which is also supplied with the residual signal e (n + a), the cross-correlation between the signal f (n) and the residual signal e (n + a) is calculated, giving a measure of the time delay difference ΔT between these signals - or possibly the instantaneous time delay directly - is determined. As a result, the adjustment of the current time delay in the time delay element 18 can be adjusted.

The signal f (n) may be either the input signal x (n) or the output signal y of the filter unit 9 in accordance with the switching state of the switching unit 17. Both signals are suitable, since in both signals information about the waveform is included before an injection into the to calm the room. Thus, a cross-correlation of the signal f (n) with the residual signal e (n + a) as calculated in the correlation unit 16 gives results with respect to the time delay ΔT that a signal experiences through the room to be quieted. The time delay ΔT calculated in the correlation unit 16 is given to

Time delay element 18 is supplied for adaptation, whereby the controlled with this adaptive algorithm, which is processed in the adaptive unit 10, much more stable.

Due to the relatively slow changes in the time delay, it is not necessary that the

Correlation unit 16 is constantly in operation. It is conceivable in a further embodiment of the invention that a temperature measuring unit (not shown in Fig. 3) is provided, which is arranged in the room to be calm and sends a start signal to the correlation unit 16 as soon as the temperature changes by a predetermined value. Only by an active start signal to the correlation unit 16 do a recalculation of a correlation and a forwarding of the new value for the delay time to the delay element 18 begin.

In the calculation of the correlation in the correlation unit 16, essentially a convolution between the input signals, namely either between the input signal x (n) and the residual signal e (n + a) or the Output signal y of the filter unit 9 and the residual signal e (n + a) performed. Since the greatest correlation between the two input signals is not always clearly recognizable, in a further embodiment of the present invention the result of the convolution of a

Subject to plausibility check. For this purpose, an area in which the greatest correlation is to be expected is predefined, this range being derived either from a previous value for the time delay or from the time delay determined with the "off-line" method, for example the time delay value, a tolerance range is superimposed. The determination of the new time delay is then limited to the determination of the largest correlation in this tolerance range. This largest correlation then gives the new time delay.

4 shows, again in a simplified representation, a block diagram of a further active noise reduction system according to the invention, in which a further method for determining the properties of the secondary path, taking into account the occurring time delays, is used.

For the sake of simplicity, only the differences from the block circuit diagram according to FIG. 3 are shown below:

Block 23 is an additional time delay. Likewise - in the same signal path - another block with the estimated secondary path 25 included. Blocks 14 and 25 are the same estimate S (n) of the secondary path, even though they are not physically the same filter, ie, blocks 14 and 25 have the same transfer function, but become independent of each other operated. The transfer functions may be the

Amplitude spectrum of the estimated secondary path act, in which case the delay time is not taken into account. The delay time is then contained in the delay element 18 and in the further delay element 23.

As is apparent from Fig. 4, both time delay elements 23 and 18 are set at the same value ΔT. Different values selected in the

Time delays 23 and 18 can lead to instabilities of the system.

Another difference from FIG. 3 is the additional adaptive algorithm in block 24 with the associated one

Filter 22, which in turn can be of the type transverse FIR filter, recursive IIR filter or lattice filter. The output signal from the filter 22 is subtracted in a second addition unit 21 from the output signal of the filter unit 9, which may also be of the type transverse FIR filter, recursive IIR filter or lattice filter. With the residual signal e (n + a), including the additional time delay element 23, the additional adaptive algorithm block 24, the estimated secondary path block 22, and the filter 22, a signal is inputted Feedback path formed. Here again, a minimization of the residual signal e (n + a) is attempted.

As in the case of the embodiment according to FIG. 3, a variant of the embodiment according to FIG. 4 is that the blocks 14 and 25, which contain the estimated transfer function of the secondary path without delay time, are omitted since the delay time is the most important one Information for the stabilization of the entire system applies.

Furthermore, the switching unit 17 again points to the two alternative embodiments, as they have already been explained with reference to FIG. 3.

Fig. 5 shows a space 26 in which an actuator unit 27 (for example a loudspeaker) emits a reference signal which is received by a sensor unit 28 (for example a microphone). The time t _amb 29 is the time that the signal output by the actuator unit 27 is for

Overcoming the distance to the sensor unit 28 required. The time t _amb is thus dependent on the distance between the actuator unit 27 and the sensor unit 28.

It should be noted that the time t _{amb is} a part of the total

Time is, which has been referred to in Fig. 1 as a time delay T.

With reference to FIG. 6, a further embodiment of the inventive method will be explained, with which the Time delay T can be determined. In this case, a possible impulse response H (t) of the transmission path is shown in Fig. 6, wherein a pulse at the time t = 0 is given in the transmission path. From the impulse response H (t), the time delay T required for the suppression is determined. For this purpose, the component contained in the impulse response H (t), which occurs before a first maximum 31 of the impulse response H (t), is erased, for example by means of a known peak search method, by a certain number in the information contained in the impulse response Samples are looked back. Thus, after applying the peak search method, a curve as shown in FIG. 7 is obtained. The advantage of this method is that the delay time T can be determined very accurately.

Claims

claims

A method for operating an active noise reduction system, comprising a transmission path, comprising at least one actuator unit (3), a space (11) and at least one sensor unit (4), wherein a replica of the transmission path is made such that in the room (11) resulting noise is eliminated or at least reduced, characterized in that due to transmission characteristics of space (11) and / or actuator unit (3) and / or sensor unit (4) resulting time delay (T, .DELTA.T) is determined and that the determined time delay (T, .DELTA.T) is taken into account in the simulation of the transmission path.

2. The method according to claim 1, characterized in that the time delay (T, .DELTA.T) is determined from a residual signal which is generated from an output signal of the transmission path and an output signal of the simulated transmission path, and either an input signal of the transmission path or a signal derived therefrom becomes.

3. The method according to claim 1 or 2, characterized in that the determination of the time delay (T, ΔT) takes place in the time domain.

4. The method according to any one of claims 2 to 3, characterized in that the time delay (T, .DELTA.T) is determined from a convolution of the residual signal with an input signal of the transmission path or a signal derived therefrom.

5. The method according to any one of claims 1 to 4, characterized in that spectral transmission characteristics of space (11), actuator unit (3) and sensor unit (4) are determined in the frequency domain and that the spectral transmission characteristics are taken into account in the replica of the transmission path.

6. The method according to any one of claims 1 to 4, characterized in that an amplitude spectrum is determined from the spectral transmission characteristics and that the amplitude spectrum is taken into account in the replica of the transmission path.

7. The method according to any one of claims 1 to 6, characterized in that at least one of the following properties is determined continuously and during operation of the active noise reduction system or are:

Time delay (T, ΔT); - amplitude spectrum;

- Spectral transmission properties.

8. The method according to any one of claims 1 to 6, characterized in that at least one of the following Properties is then determined when a change in these properties must be expected:

Time delay (T, ΔT);

- amplitude spectrum; - Spectral transmission properties.

9. The method according to claim 8, characterized in that the temperature in the space (11) is determined and that a redetermination of the transmission properties, namely time delay, amplitude spectrum and / or spectral

Transmission characteristics, is triggered when the temperature changes above a predetermined value.

10. The method according to any one of claims 1 to 9, characterized in that newly determined