EP1653768A2 - Method for reducing interference power in a directional microphone and corresponding acoustical system - Google Patents

Abstract

The method involves adjusting the directivity by changing adaptation parameters such that a sum of interfering power consisting of microphone noise and radiated power of undesired signal sources is minimized as the output signal power of a directional microphone is minimized. The microphone is aligned in the given direction and a signal source is considered as undesired, when the source lies outside an angle around the given direction. An independent claim is also included for an acoustic system with a directional microphone.

Description

The present invention relates to a method for reducing interference power in a directional microphone by providing at least two microphone signals and adaptive filtering of the at least two microphone signals to obtain a directivity, wherein at least one adaptation parameter can be optimized. Moreover, the present invention relates to a corresponding acoustic system with directional microphone.

Basically, there is a need in acoustic systems and in particular for hearing aids to combine several microphone signals spatially and spectrally in such a way that the output signal contains the lowest possible noise components. Disturbances are defined on the one hand as the signals coming from undesired directions, in this case outside a certain angular range around the 0 ° direction, e.g. +/- 60 °, and on the other hand as microphone noise, which can be amplified when forming the directivity, especially in low-frequency ranges. In particular, there is the problem that the microphone noise increases when the directivity of a directional microphone is increased.

Approaches for the suppression of interference signals are available for directional microphones of the first order. However, these only deal with the aspect of suppressing external interferers by aligning so-called notches (areas of high attenuation).

According to WO 01/01731 A1, a method is also available for generating an omni-characteristic with low microphone noise by parameter selection. A suppression of interference is not provided here.

A disadvantage of the known solutions is thus that no uniform approach for the simultaneous optimization of the total power of microphone noise and signal sources that come from undesirable directions is available. In particular, no solution exists for directional microphones of second order.

From the patent DE 103 27 889 B3 a hearing aid with a microphone system is known in which different directional characteristics are adjustable. However, a higher level of directivity also increases the microphone noise caused by the microphone system. It is therefore always necessary to find a compromise between the strength of the directivity and the maximum microphone noise accepted. However, the hearing aid wearer has to find the compromise himself, at least he has to take part in it.

The object of the present invention is thus to propose a method for the reduction of interference power in a directional microphone, in which both the microphone noise and the signal powers of interference sources are taken into account. In addition, a corresponding acoustic system should be specified.

According to the invention, this object is achieved according to claim 1 by a method for reducing interference power in a directional microphone by providing at least two microphone signals and adaptive filtering of the at least two microphone signals to achieve a directional effect, wherein at least one adaptation parameter can be optimized, and adjusting the directivity by changing the at least one adaptation parameter such that the sum of interference power is reduced. The interference power, as mentioned, is composed of microphone noise and the power of unwanted signal sources. Thus, noise and noise sources can be considered equally. The directional microphone is in a given direction, in particular in the 0 ° direction and signal sources are considered undesirable if they are outside a specified angle around the given direction. By adjusting the adaptive filters, angular ranges can thus be defined in which sound sources are treated as sources of interference. In this case, the at least one adaptation parameter of the filter device is modified such that the total output signal power is minimized, but the signal from the predetermined or 0 ° direction is not changed. With the reduction of the output signal power, the noise power is automatically reduced as a whole, and yet the gain in the main incident direction remains unaffected.

In addition, the invention provides a corresponding acoustic system according to claim 4.

Advantageously, the invention guarantees an uninfluenced signal from the 0 ° direction in combination with an adaptation method which causes a minimization of the sum of the interfering signals.

The filters of the filter device are preferably first-order or second-order adaptive finite impulse response (FIR) filters. As a result, an adaptive directional microphone of high quality can be achieved.

In a particularly preferred embodiment, the reduction of interference power takes place in several subbands. As a result, the reduction of interference power in different frequency ranges can take place selectively.

FIG 1

ein Prinzipschaltbild eines Differenzialmikrophons erster Ordnung gemäß dem Stand der Technik;

FIG 2

ein äquivalentes Schaltbild zu FIG 1 mit zwei FIR-Filtern;

FIG 3

ein Richtdiagramm für das Differentialmikrophon von FIG 1;

FIG 4

die Abhängigkeit des Mikrophonrauschens von der Frequenz und der Richtwirkung;

FIG 5

ein Diagramm zur Optimierung des Adaptionsparameters;

FIG 6

ein Prinzipschaltbild eines Differentialmikrophons zweiter Ordnung gemäß dem Stand der Technik;

FIG 7

ein äquivalentes Schaltbild zu FIG 6 mit drei-FIR-Filtern;

FIG 8

ein Richtdiagramm des Differentialmikrophons von FIG 6; und

FIG 9

die Abhängigkeit des Mikrophonrauschens von der Frequenz und den Adaptionsparametern für das Differentialmikrophon zweiter Ordnung.

The present invention will now be explained in more detail with reference to the accompanying drawings, in which:

FIG. 1

a schematic diagram of a differential microphone of the first order according to the prior art;

FIG. 2

an equivalent circuit diagram to FIG 1 with two FIR filters;

FIG. 3

a directional diagram for the differential microphone of FIG 1;

FIG. 4

dependence of microphone noise on frequency and directivity;

FIG. 5

a diagram for optimizing the adaptation parameter;

FIG. 6

a schematic diagram of a differential microphone second order according to the prior art;

FIG. 7

an equivalent circuit diagram to FIG 6 with three FIR filters;

FIG. 8

a directional diagram of the differential microphone of FIG 6; and

FIG. 9

the dependence of the microphone noise on the frequency and the adaptation parameters for the differential microphone of the second order.

The embodiments described in more detail below represent preferred embodiments of the present invention.

zugeführt. Die Entzerrung liefert ein Monoausgangssignal y(k).For a better understanding of the present invention, however, first a differential microphone according to the prior art will be explained with reference to FIG 1. Two microphones M1 and M2 receive a time-dependent sound signal s (t). In each case, a microphone noise signal n1 (t) or n2 (t) mixes with the ideal microphone signals. The respective sum signals are digitized with an analog-to-digital converter so that the microphone signals x ₁ (k) and x ₂ (k) result. A first-order differential microphone subtracts the two microphone signals x ₁ (k) and x ₂ (k) crosswise, as is known for directional microphones. The signals are delayed in the corresponding paths with timers T and a difference signal is multiplied by an adaptation parameter a. The resulting signals are added and for equalization in the useful signal direction an equalizer EQO with the transfer function $\mathrm{H}\left(\mathrm{z}\right)=\frac{1}{1-z^{-2}}$

fed. The equalization provides a mono output signal y (k).

The differential microphone DM1 can be realized by two FIR filters FIR1 and FIR2 with the transfer functions 1 + az ^-1 and -az ^-1 . This is shown schematically in FIG. The filter coefficients can not be chosen freely, but depend on the parameter a. This dependence, which results from the conversion of the filters from the differential microphone DM1, ensures that the output signal after directional microphone processing contains the signal from the 0 ° direction (useful signal direction) unchanged, regardless of the choice of the parameter a. To optimize parameter a, it must be adapted to the respective acoustic situation.

3 shows the effect of the parameter a in a directional diagram. At a = 0 the sound is attenuated from the direction 180 °. With increasing a, the notches (directions of strongest damping) move forward.

As can be seen from the diagram of FIG. 4, the microphone noise also increases with increasing a.

The aim now is to keep the total interference power of a directional microphone as low as possible. Therefore, on the one hand to adjust the directivity of the directional microphone so that the sound of a source of interference is suppressed as much as possible and on the other hand, the microphone noise is as low as possible. In FIG. 5, the power of the interference signal ST and the microphone noise are qualitatively plotted over the parameter a. A sum signal SUM from the two signals ST and MR represents the total interference power for the directional microphone. The goal is to find the minimum of this curve and the corresponding Use parameter value a _min for the adaptive filter.

The adaptation of the directional microphone to a specific interference source or the optimization of the parameter a can be carried out, for example, by a gradient method comparable to the LMS method (least mean squares). But there are also other embodiments conceivable. With the gradient method, the adaptation condition is very simple. It can simply be noted as a minimization of the average output power of the directional microphone.

For the adaptation of the directional microphone minimizing the average output signal power is only possible because it is ensured by the specific choice of the filter coefficients in dependence of the parameter a, that the useful signal from the 0 ° direction is not changed. The minimization of the total power (useful signal + disturbance) is therefore equivalent to minimizing the power of the disturbance. The interference consists of two components: microphone noise and interference from signal sources that are incident from undesired directions. An attenuation of direction-dependent signal sources can be achieved by selecting the parameter a> 0. By limiting to a maximum value, e.g. 2, you set the range in the 0 ° direction - in this case +/- 60 ° - in which incident signal sources are not or only slightly attenuated. If the adaptive method is also allowed to select the parameters smaller than 0, the directivity is reduced, but also the power of the microphone noise is reduced. By adapting the parameter in individual frequency bands, the method that measures the sum of the noise powers, i. of microphone noise and signal sources from undesired directions, in each frequency band is minimized.

The parameter a can be found by finding the minimum of the mean square error. This means that the expected value of the output signal should be minimal, ie $$e\left\{{\left|y\left(k\right)\right|}^2\right\}\stackrel{!}{=}\min$$

Thus, a simple and robust method for adaptive directional microphony of first order is given.

besitzt.A directional microphone of the second order has according to FIG. 6 in addition to the microphones M1 and M2 via a third microphone M3. Also, the output signal is disturbed by noise microphone n3 (t), and digitally converted the corresponding sum signal to a microphone output signal x ₃ (k). The differential microphone DM2 generates after the usual equalization EQO from the three microphone signals x ₁ (k), x ₂ (k) and x ₃ (k) an output signal y (k). In this case, in a first stage, the microphone signals are subtracted crosswise after a corresponding time delay T and a signal weighting with the factor a takes place in two sub-branches, so that the two intermediate signals z ₁ (k) and z ₂ (k) result. Analogously to the differential microphone of the first order according to FIG. 1, in a second step, the intermediate signals z ₁ (k) and z ₂ (k) for the equalizer in the 0 ° direction, here the transfer function $\mathrm{H}\left(\mathrm{z}\right)=\frac{1}{1-2z^{-2}+z^{-4}}$

has.

The output signal of the equalizer EQO is again denoted by y (k).

The differential microphone of the second order can be described analogously to the differential microphone of the first order by three FIR filters, as shown in FIG. 7 (see also FIG. Here, the first FIR filter FIR1 has the transfer function $$1+\left(a+b\right)z^{-1}-ab{z}^{-2}$$

und der dritte Filter FIR3 die Übertragungsfunktion $$ab+\left(a+b\right)z^{-1}+z^{-2}$$

The second filter FIR2 has the transfer function $$-\left(a+b\right)-2\left(ab+1\right)z^{-1}-\left(a+b\right)z^{-2}$$

and the third filter FIR3, the transfer function $$ab+\left(a+b\right)z^{-1}+z^{-2}$$

For the determination of the two parameters a and b, in turn, the minimum of the mean square error of the output signal is to be calculated, ie $$e\left\{{\left|y\left(k\right)\right|}^2\right\}\stackrel{!}{=}\min$$

For a specific interference situation, this results in specific parameters a and b, in which the total interference power is minimized. The effects of the two parameters a and b can be seen in the directional diagram of FIG. The values a = -1 b = -1 lead almost to an omnicharacteristic of the directional microphone, as indicated by dotted line in FIG. A pronounced directional characteristic, on the other hand, results for the values a = 0.1 b = 0.9 in the 0 ° direction. Even with the directional microphone of the second order, the microphone noise increases with increasing directivity, as shown in FIG. 9 for the same parameter combinations that were used in FIG.

The parameters a and b determined above also lead to the desired compromise between directivity and microphone noise according to FIG. 5, in which the total signal interference power is minimal.

In addition, the method has an increased robustness to mismatch of the microphones or mismatch by head influences, for example, of a hearing aid wearer or headset wearer. In this case, the adaptive method will select the parameters to reduce the overall disturbance again. In extreme cases, the choice of parameters which can achieve spatial attenuation without mismatching is then automatically prevented in favor of reducing microphone noise. The reason is, that the spatial attenuation due to the mismatch anyway can not be formed. In contrast, with a fixed, non-adaptive directional microphone attempting to achieve maximum directivity, mismatching could result in spatial attenuation (attenuation in one or more spatial directions), but in addition to this microphone noise is amplified.

Claims (6)

Method for reducing interference power in a directional microphone by Providing at least two microphone signals (x ₁ (k), x ₂ (k) and adaptive filtering (DM1) of the at least two microphone signals to obtain a directivity, wherein at least one adaptation parameter (a, b) can be optimized,
marked by - automatically adjusting the directivity by changing the at least one adaptation parameter (a, b) such that the sum (SUM) of current noise power consisting of microphone noise (MR) and power from unwanted signal sources (ST) is minimized by the total output signal power (SUM) of the directional microphone is minimized, but the signal from a given direction is not changed, the directional microphone is aligned in the predetermined direction and a signal source is considered undesirable when it is outside a predetermined angle to the predetermined direction. The method of claim 1, wherein adaptive filtering (DM1) uses first or second order FIR filters. Method according to one of the preceding claims, wherein the reduction of interference power is performed separately in a plurality of subbands. Comprising acoustic system with directional microphone - At least two microphones (M1, M2) for the delivery of at least two microphone signals (x ₁ (k), x ₂ (k)) and a filter device (DM1, DM2) for adaptively filtering the at least two microphone signals to obtain a directivity, wherein at least one adaptation parameter (a, b) can be optimized,
marked by - A computing device with which the at least one adaptation parameter (a, b) is automatically changed such that the sum (SUM) of current noise power consisting of microphone noise (MR) and benefits of unwanted signal sources (ST) is minimized by the at least one adaptation parameter (a, b), the entire output signal power of the directional microphone is minimized, but the signal from a given direction is not changed, wherein the directional microphone is aligned in the predetermined direction and a signal source is considered undesirable if they are outside a predetermined Angle is around the given direction. Acoustic system according to claim 4, wherein the filters of the filter means (DM1, DM2) are first or second order adaptive FIR filters. Acoustic system according to claim 4 or 5, wherein the filter means (DM1, DM2) for a plurality of subbands having separate filters, so that the reduction of noise power in a plurality of subbands is separately feasible.

Images