EP1872360A1

EP1872360A1 - Method for reproducing a secondary path in an active noise reduction system

Info

Publication number: EP1872360A1
Application number: EP06721921A
Authority: EP
Inventors: Harry Bachmann
Original assignee: Anocsys AG
Current assignee: Anocsys AG
Priority date: 2005-04-22
Filing date: 2006-04-21
Publication date: 2008-01-02
Also published as: KR20080003914A; WO2006111039A1; US20080317256A1; JP2008538420A; CN101203905A

Abstract

The invention relates to a method for reproducing a secondary path in an active noise reduction system comprising a transmission path (S, 9', 10, 11), an adaptively adjustable filter (13), and an addition unit (14), the adaptively adjustable filter (13) being adjusted according to an output signal of the addition unit (14). The inventive method includes the following step: a delay time (T) of a signal along the transmission path (8, 9, 10, 11) is eliminated in the transmission function of the adaptively adjustable filter (13) in order to generate the reproduction of the secondary path.

Description

A method for simulating a secondary path in an active Geräuschreduktionsssysteπi

The present invention relates to a method for reproducing a secondary path in an active noise reduction system, comprising a transmission path, an adaptively adjustable filter and an addition unit, wherein the adaptively adjustable filter is adjusted in response to an output signal of the addition unit, and a method for operating a active noise reduction system.

Noise sources are increasingly perceived as an environmental impact and are considered to reduce the quality of life. 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 are those of one or more sensors -? -

supplied information. 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, information about the remaining residual signal is 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. These converters are subject Restrictions, both in terms of resolution and dynamics.

When using Active Noise Canceling, hereinafter ANC, the stability of the algorithm used is a crucial factor. Currently, a number of specific algorithms are used, such as LMS (Least Mean Square) or the related Fx-LMS algorithm. In particular, the Fx algorithms have a good stability and can therefore be used well in ANC system. The prefix "Fx" indicates the reproduction of the so-called "secondary path", which shows 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 Muhammad Tahir Akthar, Masahide Abe and Masayuki Kawamat on the occasion of International Workshop on Acoustic Echo and Noise Control (IWAENC2003) "was published in Kyoto in September 2003.

This document describes offline modeling of the secondary path (component influence). The known method for determining the secondary path is referred to as "offline modeling", since the properties of the secondary path are determined in advance, that is, 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.

This method of determining the secondary path (component influence) has in common the property that the time delay which occurs between the actuator and the sensor is taken into account for the calculation of the component influence (secondary path) independently of the frequency response. However, since this time delay is an essential property of the secondary path, neglecting this time delay in the modeling of secondary influence affects the efficiency and stability of the whole system. When changing the environmental parameters, such as the air pressure or the temperature, the duration of the signal changes. If the running time of the signal becomes smaller, the algorithm is characterized by that in the model of the secondary path given delay too slow to deliver a satisfactory result. As a result, poorer damping properties and, in extreme cases, an unstable system can arise.

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 and to determine the properties of the secondary path (component influence) due to the change of this signal. The additional signal is filtered out again before the "Antinoise signal" is output via the actuator, in this case a loudspeaker. The disadvantage of this method is that this signal is always present.

If ANC uses a model of the secondary path (component influence), its properties are automatically included in the calculation of the "anti-noise". If the model of the secondary path contains a time delay, as is the case in the usual models, the system is limited by the fact that a change in the delay of the signal can no longer be compensated. This is especially the case when the duration of the signal decreases. 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 specified in claim 1. Advantageous embodiments and a method for operating an active

Noise reduction system are specified in further claims.

The invention firstly relates to a method for simulating a secondary path in an active noise reduction system comprising a transmission path, an adaptively adjustable filter and an addition unit, wherein the adaptively adjustable filter is set in dependence on an output signal of the addition unit. The method according to the invention comprises the following steps:

a known signal is supplied to the transmission link and to the adaptively adjustable filter having an adjustable transfer function;

the adaptive filter or its transfer function is set in such a way that the output signal of the addition unit is minimal;

- A delay time of a signal through the transmission path is or is eliminated in the transfer function of the adaptively adjustable filter to produce the replica of the secondary path. Thus, for the first time, a method is created with which the influence of the signal propagation delay on the model of the secondary path is no longer present, whereby a substantial improvement in the system stability of the active noise reduction system is achieved.

In one embodiment variant of the method according to the invention, the delay time is determined, for which purpose a method based on the peak search method is used in particular. As a result, an extremely accurate determination of the delay time is possible, which leads to an extremely good system behavior during later operation.

In a further embodiment of the method according to the invention, the adaptively adjustable filter operates in the frequency domain. j

In yet another embodiment of the method according to the invention, white noise is supplied as a known signal to the transmission path and to the adaptively adjustable filter.

In a further embodiment of the method according to the invention, it is provided that the known signal is transformed by means of a transformation from a time domain to a frequency domain before the known signal is fed to the adaptively adjustable filter, and that an output signal of the transmission path is transformed by means of a transformation from the time domain. in the frequency domain is transformed before the output signal of the transmission path of the addition unit is supplied.

In yet another embodiment of the method according to the invention, it is provided that only the amplitude spectrum is used further in the transformation from the time domain to the frequency domain. This will further simplify the modeling of the secondary path and thus increase efficiency.

In yet another embodiment of the method according to the invention, it is provided that the adaptively adjustable filter is supplied with a known signal having a constant amplitude spectrum, and that an output signal of the transmission path is transformed by means of a transformation from the time domain to the frequency domain before the output signal of the Transmission line of the addition unit is supplied.

In a further embodiment of the method according to the invention, the phase spectrum of the known signal is no longer used. This achieves a further simplification.

Finally, a method for operating an active noise reduction system is specified, comprising a transmission path, an adaptively adjustable filter and an addition unit, wherein the adaptively adjustable filter is adjusted in response to an output signal of the addition unit and wherein a simulated secondary-path acts on the adaptive filter such that secondary-path influences are taken into account, wherein the replica of the secondary-path has been carried out according to the method described above.

The present invention is based on

Embodiments with reference to drawings further explained below. Show it:

1 is a simplified block diagram of a known method for determining the secondary path according to the "Offline Modeling" method,

Fig. 2 is a simplified block diagram of a

Embodiment of a method according to the invention in a schematic representation,

3 is another simplified block diagram of a known method for determining the properties of the secondary path;

4 is a simplified representation for explaining a method according to the invention,

Fig. 5 is a further simplified representation of

Explanation of a method according to the invention,

6 shows a further simplified block diagram of a method according to the invention, _. , _ _{. ,} _ _{"_}

- 10 -

Fig. 7 is a block diagram of another

Embodiment of a method according to the invention,

Fig. 8 shows an example of a waveform and

Fig. 9 shows another example of a waveform.

Fig. 1 consists of a noise generator unit 1, the transmission path 2 with the transfer function H (z) whose properties are to be replicated, and a filter 3 in which a model H (z) of the actual transfer function H (z) is included and is controlled by an adaptive unit 4, in which an adaptive algorithm is processed. The model H (z) is thus the replica of the transfer function H (z) in the transmission path 2.

The transmission path 2, the filter 3 and the adaptive unit 4 are supplied with a randomly generated by the noise generator unit 1 signal (Random Noise Generator). From the signals d (n), y (n) resulting at the output of the transmission path 2 and the filter 3, a sum is formed in an addition unit 5, wherein the output signal y (n) of the filter 3 is inverted before the addition.

The resulting residual signal e (n) 6 is supplied to the adaptive unit 4. The one in the adaptive unit 4 processed algorithm sets the filter 3 so that the residual signal e (n) is minimized. An optimal adjustment of the overall system is achieved when the residual signal e (n) 6 is equal to zero. In this case, the transfer function H (z) coincides with the model H (z).

3 shows a known method for determining the properties of the secondary path (component influence). A transmission path is formed of an amplification unit 8, an actuator 9 (for example a loudspeaker), a sensor 10 (for example a microphone) and a sensor amplifier 11. A noise generator unit 7 supplies this transmission path, the filter 13 and the white noise adaptive unit 15. The filter 13 is adjusted by the adaptive algorithm, which is executed in the adaptive unit 15, so that the result of the addition unit 14 is minimized, wherein one of the two summands must be inverted. In this method, time delays attributed to the secondary path (component influence) flow into the calculation of the filter 13. The secondary path (component influence) here consists of the specific influence of the amplifiers 8, 11, the actuator 9, the sensor 10 and the transmission medium between the actuator 9 and the sensor 10. This is only one of the possibilities, such as a secondary path can be. Instead of a speaker and a microphone, other actuators and sensors may be used. The microphone amplifier 11 may also include a filter under certain circumstances. The present invention consists in the fact that the influence of the signal propagation times occurring in the secondary path is canceled out by the fact that the signals are transformed from the time domain into the frequency domain. This will be illustrated with reference to the embodiment of the invention shown in FIG.

2 shows the basic structure of a system according to the invention for determining the properties of the secondary path (component influence) which comprises the various components, such as amplifier 8, actuator or loudspeaker 9, sensor or microphone 10, sensor or microphone amplifier 11 and consists of the transmission medium between the actuator 9 and sensor 10. The noise generator unit 7 supplies the secondary path with white noise. At the same time, the noise is supplied to a transformation unit 12, which carries out a transformation from the time domain to the frequency domain. Another transformation unit 16 transforms the signal into the frequency domain at the end of the secondary path. The adaptive algorithm used in the unit 15 adjusts the filter 13 so that the sum formed in the adding unit 14 is minimized, inverting the signal resulting from the filter 13 before the summation.

Due to the transformation carried out in the transformation units 12 and 16 from the time domain to the frequency domain, the occurring in the secondary path The temporal change of the maturity is excluded by a majority. It has been found that certain signal components, which are offset by a multiple of 2 * π, can not be excluded. The filter 13 thus only represents the properties of the secondary path (component influence) in the frequency domain.

The difference from the method shown in FIG. 3 is the transformations carried out in the transformation units 12 and 16 from the time domain to the frequency domain.

A further embodiment variant of the method according to the invention is explained with reference to FIG. 8, 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. 8, wherein a signal 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. 9 is obtained. The advantage of this method of excluding the time delay is that the delay T can be determined very accurately. Fig. 4 shows the frequency spectrum of white noise. On the abscissa the frequency is 20, on the ordinate the amplitude 19 is plotted. The spectrum shows a constant course of the amplitude 17.

Fig. 5 shows the frequency spectrum after the white noise according to Fig. 4 has passed the secondary path. The abscissa again has the frequency 20, and the ordinate the amplitude 19. The spectrum no longer shows a constant amplitude spectrum but a frequency spectrum that varies with the frequency. This amplitude spectrum shows a possible output signal in the frequency range of a secondary path, after it has been excited with the course of FIG.

In Fig. 2, white noise is generated by the noise generator unit 1, which means that the amplitude 17 of each individual frequency is the same. This is shown in Fig. 4.

After the white noise has now passed the secondary path, the amplitude 18 is no longer the same size at each frequency, as shown in FIG. 5 can be seen.

Fig. 6 shows a block diagram, with the two noise generators 21 and 22, in which white noise is generated. To calculate the secondary path, a constant value is used at the input of the filter 13 and at the adaptive unit 15. The use of a number - in this case a constant value instead of a complex one Signals - is another simplification in the modeling of the secondary path.

In Fig. 7, a simple ANC system is shown. The operation of an ANC system whose secondary path has been determined in the frequency domain is explained below.

The signal x (n) to be minimized is 28, the remaining residual signal e (n) is 29, the transmission link with the transfer function H is 23, and the filter H, with which the transmission link H is replicated, is referenced 24 , Blocks 25 and 26 deserve special attention. Thus, at 25, the secondary path (component influence) is identified, and at 26 an estimate of the secondary path (component influence) is indicated. In block 26, therefore, the parameters are stored which were previously determined by means of the methods described in FIGS. 2 and 3.

If the known method described in FIG. 3 is used, the restrictions which have already been described above come into play, namely the temporal change in the propagation time of the signals in the secondary path is not taken into account. If the influence of the signal propagation time in block 26 is large, this can no longer be corrected by the filter 24.

On the other hand, if the parameters were determined by the method according to the invention described in FIG Influence of the signal propagation time on the model of the secondary path no longer exists. However, the parameters determined in the filter 13 (FIG. 2 or 6) must first be transformed back into the time domain from the frequency domain by means of an inverse transformation, before they are stored in the model of the secondary path (block 26). The block 26 thus describes the frequency characteristics of the secondary path 25. In the addition unit 14, in turn, a sum is formed after the signal x (n) to be minimized has been influenced on the one hand by the transmission path 23, and on the other hand by the filter 24 and the secondary Path 25 was edited. It should be noted that one of the two summands must be inverted to form the difference with the addition unit 14, as can be seen from the drawing. The adaptive unit 27, in which an adaptive algorithm is executed, controls the filter 24 such that the residual signal e (n) 29 is as small as possible, ie minimal.

Claims

claims

A method for simulating a secondary path in an active noise reduction system, comprising a transmission link (8, 9, 10, 11), an adaptively adjustable filter (13) and an addition unit (14), wherein the adaptively adjustable filter (13) in response to an output signal of the addition unit

(14), the method comprising the steps of:

a known signal is supplied to the transmission link (8, 9, 10, 11) and to the adaptively adjustable filter (13) having an adjustable transfer function;

- The adaptive filter (13) or its

Transfer function is set so that the output of the addition unit (14) is minimum;

- A delay time (T) of a signal through the transmission path (8, 9, 10, 11) is or is eliminated in the transfer function of the adaptively adjustable filter (13) for generating the replica of the secondary path.

2. The method according to claim 1, characterized in that the delay time (T) is determined, for which purpose in particular a method based on the peak search method is used.

3. The method according to claim 1, characterized in that the adaptively adjustable filter (13) operates in the frequency domain.

4. The method according to claim 1 or 3, characterized in that a white noise as a known signal of the transmission path (8, 9, 10, 11) and the adaptively adjustable filter (13) is supplied.

5. The method according to any one of claims 1, 3 or 4, characterized in that the known signal is transformed by means of a transformation of a time in a frequency range before the known signal is supplied to the adaptively adjustable filter (13), and that a Output signal of the transmission path (8, 9, 10, 11) is transformed by means of a transformation from the time to the frequency range, before the output signal of the transmission path (8, 9, 10, 11) of the addition unit (14) is supplied.

6. The method according to claim 5, characterized in that in the transformation from the time to the frequency range, only the amplitude spectrum is further used.

7. The method according to any one of claims 1, 3 or 4, characterized in that the adaptively adjustable filter (13), a known signal is supplied, which has a constant amplitude spectrum, and in that an output signal of the transmission path (8, 9, 10, 11 ) by means of a transformation from the time domain to the frequency domain is transformed before the output signal of the transmission path (8, 9, 10, 11) of the addition unit (14) is supplied.

8. The method according to claim 7, characterized in that a phase spectrum of the known signal is not used further.

9. Method for operating an active

Noise reduction system comprising a transmission path (8, 9, 10, 11), an adaptively adjustable filter (13) and an addition unit (14), wherein the adaptively adjustable filter (13) is adjusted in response to an output signal of the addition unit (14) and wherein a modeled secondary path acts on the adaptively settable filter (13) in such a way that secondary-path influences are taken into account, characterized in that the replication of the secondary path takes place according to one of claims 1 to 8.