Method for directional signal processing of acoustic systems

Abstract

The invention relates to a method for directional signal processing of an acoustic system (1), wherein a first input signal (E1) is generated from ambient sound (4) by a first input transducer (M1) of the acoustic system, wherein a second input signal (E2) is generated from ambient sound (4) by a second input transducer (M2) of the acoustic system, wherein a first intermediate signal (Z1) and a second intermediate signal (Z2) are generated from the first input signal and the second input signal, respectively, wherein a temporary superposition parameter (a0) is obtained for a first superposition (U1) of the first intermediate signal and the second intermediate signal, so that for the first superposition, the attenuation in a first target direction (42) has a maximum value, wherein a superposition parameter (a) is formed according to the temporary superposition parameter (a0) so that for a second superposition (U2) of the first intermediate signal and the second intermediate signal formed using the superposition parameter (a), the gain in the first target direction (42) has a predetermined first value (g1) greater than zero, and wherein an output signal (10) of the acoustic system is formed according to the second superposition.

Description

Technical Field

[0001] The present invention relates to a method for directional signal processing of an acoustic system, wherein a first input signal is generated from ambient sound by a first input transducer of the acoustic system, wherein a second input signal is generated from ambient sound by a second input transducer of the acoustic system, wherein a first intermediate signal and a second intermediate signal are generated based on the first input signal and the second input signal, respectively, wherein superposition parameters for superimposing the first intermediate signal and the second intermediate signal are obtained so that for the superposition, the attenuation in a first target direction has a maximum value, and wherein an output signal of the acoustic system is formed based on the superposition. Background Art

[0002] For hearing devices and communication devices, directional signal processing in a specific acoustic environment (i.e., a useful signal source with specific spectral characteristics is specifically arranged and a specific interference signal source is simultaneously present) can significantly improve the signal-to-noise ratio (SNR). Usually, interference signals are adaptively shielded by specifically orienting the directional characteristics of the output signal formed from the intermediate signal toward the main useful signal source, mostly by minimizing the signal energy of the output signal under the constraint that one of the intermediate signals is fixedly oriented toward the useful signal source.

[0003] An important application example here is an adaptive directional microphone, in which the intermediate signal, for example as a cardioid or anti-cardioid signal, respectively largely completely cancels the interfering signal in a specific interfering signal direction, so that the weighted superposition of the intermediate signals mentioned can be optimized with respect to the signal energy by means of corresponding superposition parameters. In this case, the best solution usually provides an output signal in which strongly directional interfering signal sources are largely completely suppressed.

[0004] However, complete suppression of interfering signal sources is usually undesirable for reasons of spatial hearing and also for safety reasons, if the interfering signal source is provided, for example, by another road user in street traffic and should therefore generally be perceived and preferably also remain localized. Summary of the invention

[0005] The object of the present invention is therefore to provide a method for directional signal processing of an acoustic system, by means of which strongly directional interfering signal sources are not completely suppressed, but rather remain audible, in particular in the output signal of the acoustic system.

[0006] According to the invention, the above technical problem is solved by a method for directional signal processing of an acoustic system, wherein a first input signal is generated from ambient sound by a first input transducer of the acoustic system, wherein a second input signal is generated from ambient sound by a second input transducer of the acoustic system, wherein a first intermediate signal and a second intermediate signal are generated from the first input signal and the second input signal, respectively, wherein a (especially real-valued) temporary superposition parameter for a first superposition of the first intermediate signal and the second intermediate signal is obtained, so that for the first superposition, the attenuation in a first target direction has a maximum value, wherein a superposition parameter is formed according to the temporary superposition parameter, so that for a second superposition of the first intermediate signal and the second intermediate signal formed using the superposition parameter, the gain in the first target direction has a predetermined first value greater than zero, and wherein an output signal of the acoustic system is formed according to the second superposition. Advantageous and partially inventive embodiments are the subject of the invention and the following description.

[0007] Here, an acoustic system includes any arrangement of multiple input converters for generating and further processing corresponding input signals, i.e. in particular a hearing device, a hearing device coupled to a smartphone and / or a smartwatch via Bluetooth or the like, or a communication facility for recording speech signals, for example within the scope of a conference or the like.

[0008] In this case, the input converter comprises in particular an electroacoustic converter which is designed to generate a corresponding electrical signal from a sound signal. In particular, when the first or second input signal is generated by the corresponding input converter, preprocessing, for example in the form of linear preamplification and / or A / D conversion, can also take place. In this case, the correspondingly generated input signal is provided in particular via an electrical signal, the current and / or voltage fluctuations of which substantially represent the sound pressure fluctuations in the air.

[0009] Preferably, the first intermediate signal and the second intermediate signal are generated as directional signals with mutually different directional characteristics, wherein the directional characteristics are symmetrical to each other in particular with respect to a preferred plane of the acoustic system. In this case, such a preferred plane can be defined in particular by the arrangement of the first and second input transducers relative to each other and / or by conventional applications, for example, in the case of a hearing device as an acoustic system, it is a frontal plane of the user. In this case, the first intermediate signal can be generated in particular by a time-delayed superposition of the first input signal and the second input signal and / or by a superposition of input signals filtered with different filters. In this case, the directional characteristic of the first intermediate signal has at least one minimum direction in which the attenuation has a global minimum (in all directions) and, therefore, the sensitivity of the first intermediate signal in the minimum direction is minimum, and the directional characteristic of the first intermediate signal has at least one maximum direction in which the sensitivity of the first intermediate signal (in all directions) is maximum. A similar situation applies to the second intermediate signal.

[0010] In particular, a temporary superposition parameter is now obtained by superimposing a first intermediate signal with a second intermediate signal, the second intermediate signal being weighted with a free weighting factor, and the free weighting factor being varied for the first superposition mentioned so that the first superposition in a first target direction has a maximum, in particular as complete as possible, attenuation. The first target direction is preferably provided by the direction of the first interfering signal source. In the mentioned case, the maximum attenuation can be achieved, for example, by minimizing the signal energy of the signal resulting from the superposition, in particular when the second intermediate signal has the maximum possible attenuation in the direction of the main useful signal source (towards which the first intermediate signal is preferably oriented) (so that in particular the maximum direction of the first intermediate signal coincides with the minimum direction of the second intermediate signal).

[0011] The value of the weighting factor for which the desired maximum attenuation is formed is used as a temporary superposition parameter for the first superposition. However, in order to determine the temporary superposition parameter, an explicit signal-technical generation of the first superposition is not absolutely necessary, and with knowledge of the directional characteristics of the first and second intermediate signals, for example, a tabulation can also be performed depending on the input level of the first and second intermediate signals, wherein the temporary superposition parameter is tabulated so that the correspondingly generated first superposition will have the desired attenuation.

[0012] The superposition parameters for the second superposition of the two intermediate signals are now formed based on the temporary superposition parameters, so that in a first target direction, in which the attenuation of the first superposition has an especially global maximum and is preferably as large as possible, i.e. in particular can be referred to as almost complete suppression, the gain has a predetermined first value greater than zero. This means in particular that now in the first target direction, the attenuation is no longer complete or approximately complete, but rather limited, whereby the sound signal of the first interfering signal source now remains audible in the output signal.

[0013] It is now possible in particular to form a superposition parameter a for a second superposition based on a temporary superposition parameter a0 (which is obtained from the first superposition U1=Z1-a0·Z2), and the second superposition can in particular have the formula U2=Z1-a·Z2 (wherein the first or second intermediate signal is Z1 or Z2), so that the superposition parameter a is complex-valued, i.e. a=aRe+i·aIm, where the real part is aRe and the imaginary part is aIm. This results in an additional degree of freedom for the second superposition U2. While the real-valued temporary superposition parameter a0 is determined entirely by the condition that a maximum attenuation should exist in the first target direction, the first value g1 of the gain in the first target direction can also be determined by means of an additional degree of freedom for the second superposition, which is provided by the usually complex-valued superposition parameter a.

[0014] Depending on which constraints are additionally imposed on the second superposition, the real part aRe and the imaginary part aIm of the superposition parameter can therefore be expressed in particular as a function of the provisional superposition parameter a0 and the first value g1 of the gain in the first target direction.

[0015] Preferably, a cardioid signal is generated as the first intermediate signal and an anti-cardioid signal is generated as the second intermediate signal. These signals have the advantage of being symmetrical to one another, i.e. the minimum direction of the anti-cardioid signal (i.e. the direction of minimum sensitivity and therefore maximum attenuation) coincides with the maximum direction of the cardioid signal (i.e. the direction of maximum sensitivity), and vice versa. Furthermore, both signals have, in an ideal case, complete attenuation in their respective minimum directions. Furthermore, these signals can be easily generated from the superposition of the time delays of the two input signals by using the acoustic run time between the two input converters as the delay in each case. In the ideal case, the generated cardioid and anti-cardioid signals are rotationally symmetrical here in each case about the connecting line through the two input converters.

[0016] The superposition parameter a is advantageously formed such that for the second superposition U2=Z1-a·Z2, the gain in the first target direction has a global minimum with a predetermined first value g1. If the two input signals E1, E2 are shown in the frequency domain in dependence on the ambient sound X, then the intermediate signals Z1, Z2 can be generated in the frequency domain in dependence on the direction of incidence of the ambient sound X. Expressed as (And a similar situation applies to Z2, where In the case of a time-delayed superposition, the corresponding phase position is also taken into account). From this, the transfer function can be determined for the second superposition U2(ω) with respect to the ambient sound X. Due to the directional dependence of the intermediate signal, the transfer function itself is direction-dependent.

[0017] It has proven to be advantageous here if the real part aRe of the superposition parameter a=aRe+i·aIm is formed by a linear function of the temporary superposition parameter a0, which is dependent on a first value g1 of the gain in the first target direction and is transformed into the temporary superposition parameter a0 when the first value g1 approaches zero, and / or if the imaginary part aIm of the superposition parameter a is linearly dependent on the mentioned real part aRe and approaches zero when the first value g1 approaches zero.

[0018] The linear function of the temporary superposition parameter a0 for the superposition parameter a allows a uniform treatment of all spatial directions, which are represented by the respective temporary superposition parameter a0 corresponding to the first target direction of the associated first superposition and its maximum, preferably complete attenuation. As g1 approaches zero, the superposition parameter changes to the temporary superposition parameter a0 set for the first target direction, taking into account the fact that complete attenuation should preferably occur in the first target direction for it (i.e. g1=0).

[0019] The linear function of the real part aRe of the superposition parameter a can be, in particular, the following formula:

[0020]

[0021] Here, a0 specifies a temporary superposition parameter and ε denotes a (preferably continuous) function of g1, wherein if g1=0, then ε=0. Particularly preferably, ε=g1 2 / (1-g1 2 ). For the imaginary part of the superposition parameter aIm, it is particularly applicable that

[0022] The superposition parameters are preferably formed such that for the second superposition, the gain in the first target direction has a predetermined first value and the gain in the second target direction has a predetermined second value, wherein in particular the second value is less than the first value and / or equal to zero.

[0023] By means of the usually complex-valued superposition parameter, an additional degree of freedom is introduced into the generation of the second superposition (compared to a purely real-valued superposition parameter). As a result, on the one hand, a first value g1 of the gain can be determined for the first target direction. Since the selection of the first target direction is a priori free, a first condition for the complex-valued superposition parameter is proposed by means of such a determined relationship between the first target direction and the gain to be applied there, whereby an additional degree of freedom is also retained due to the imaginary part of the superposition parameter. This degree of freedom can now be used to predetermine a second value g2 for the gain in the second target direction. This second value can here be determined on the one hand as g2=0, or also as 0<g2<g1, so that a global minimum of the gain is determined by the second value, but no complete attenuation takes place in any direction.

[0024] Preferably, the real part aRe and the imaginary part aIm of the superposition parameter a=aRe+i·aIm can be described according to a circle in the complex plane, which has the temporary superposition parameter a0 as the origin and the radius ρ, the square of the radius ρ 2 Depends on the square of the first value g1 of the gain and the square of the temporary superposition parameter a0. The dependence of this formula can be in particular

[0025] (aRe-a0) 2 +aIm 2 =g1 2 (a0+1) 2

[0026] where ρ 2 =g1 2 (a0+1) 2 .

[0027] The superposition parameter a is preferably formed in this case so that the signal generated by the second superposition has a maximum directivity index (DI). In this case, the DI can be determined for the signal generated by the second superposition (in the maximum direction φ0) with respect to the incident sound signal according to the maximum square of the absolute value of the transfer function G(ω, φ), the DI being normalized over the integral of the square of the absolute value of the transfer function in all spatial directions. In this case, the DI is usually defined by the decimal logarithm of the mentioned variable:

[0028]

[0029] Therein, the integration in the denominator is performed over the normalized unit sphere, yielding DI=0 for an omnidirectional signal.

[0030] In this case, in particular, the superposition parameter a=aRe is real-valued, and the formula a=aRe=a0±g1·(a0+1) is particularly preferred. Preferably, a plus sign is selected for a0<0.5 and a minus sign is selected for a0>0.

[0031] In this case, preferably, in a preset environment of the temporary superposition parameter a0, regularization of the superposition parameter is carried out around a critical value, in particular 0.5, so that the value of the superposition parameter a to be applied is preset for the critical value of the temporary superposition parameter a0, and the temporary superposition parameter a0 is continuously mapped to the superposition parameter a for values ​​[a0-d1, a0+d2] from the preset environment around the critical value (in particular where d1=d2). In particular, the real part aRe of the superposition parameter a can be preset for the critical value of the temporary environment parameter a0, so that at the critical value of a0, DI is maximized by the assigned real part aRe. This can be achieved as follows, in that, depending on aRe and where ρ 2 =g1 2 (a0+1) 2 The above DI is calculated, and with respect to aRe, DI is maximized for a critical value of a0 (particularly a0=0.5).

[0032] It has been found to be further advantageous to form the superposition parameters such that for the second superposition, the gain in the first target direction has a predetermined first value and the gain in the second target direction has a predetermined second value, wherein the gain in the second target direction has a global minimum with the predetermined second value. The second value g2 for the global minimum of the gain can in particular be greater than zero in all directions, so that no complete attenuation is performed in any spatial direction. It can thereby be achieved that, on the one hand, in a first target direction, which can be provided, for example, by the direction of a main source of interfering signals, a defined gain is ensured (by means of the first value g1), so that the corresponding interfering signal remains audible in the output signal in any case within the range of the first value g1, and the complete attenuation of possible further interfering signals is additionally suppressed.

[0033] The invention also relates to an acoustic system, which comprises at least one first input converter for generating a first input signal from ambient sound and a second input converter for generating a second input signal from ambient sound, and also comprises a control unit, which is designed to implement the above-mentioned method. The method according to the invention shares the advantages of the acoustic system according to the invention. Here, the advantages described for the method and its extensions can be transferred to the acoustic system accordingly.

[0034] The acoustic system preferably comprises a hearing device, in which a first input converter and a second input converter are arranged. In particular, the control unit is also arranged in the hearing device. Here, the hearing device is preferably designed as a local device worn by the user on his ear. However, the control unit can also be implemented at least partially on a device associated with the hearing device, such as a smartphone connected to the hearing device via Bluetooth or the like.

[0035] However, the hearing device can also be designed as a binaural hearing device with two local devices, wherein the user wears one of the two local devices on each ear for operation of the hearing device. Preferably, the first input converter and the second input converter are each arranged in one of the two local units, so that a cardioid signal or an anti-cardioid signal can be generated as the first or second intermediate signal from the associated first and second input signals. In the case of a binaural hearing device, the control unit for implementing the method can also be distributed to the two local devices and implemented by their corresponding signal processing devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The embodiments of the present invention are described in detail below based on the accompanying drawings. Herein, schematically:

[0037] Figure 1 A hearing device is shown in a top view in an environment with main wanted signal sources and interfering signal sources;

[0038] Figure 2 The top view shows the Figure 1 Hearing devices use adaptive directional microphones to suppress interfering signal sources;

[0039] Figure 3 The block diagram shows a method for Figure 1 Method for directional noise suppression of hearing devices;

[0040] Figure 4 The top view shows the constraints of limited and global minimum gain in the preset direction, according to Figure 3 Directional characteristics of directional noise suppression;

[0041] Figure 5 The directional characteristics of directional noise suppression in a preset direction under the constraint of the maximum directivity index are shown in a top view;

[0042] Figure 6 The directional characteristic of a directional noise suppression with a predetermined noise suppression in a predetermined direction and at a predetermined minimum gain is shown in a top view.

[0043] Parts and variables that correspond to one another are provided with the same reference symbols in all the figures. DETAILED DESCRIPTION

[0044] Figure 1An acoustic system 1 is schematically shown in a top view, which is in the present case designed as a hearing device 2. The hearing device 2 has a first input converter M1 and a second input converter M2, which are in the present case provided respectively by a microphone, and are designed to generate a first input signal E1 or a second input signal E2 from an ambient sound 4. The mentioned input signal E1 or E2 is respectively supplied to a control unit 6 for implementing a method for directional signal processing which will be described further. The control unit 6 is in the present case implemented on a signal processing device 8 of the hearing device 2. In a manner which will be described further, an output signal 10 is generated by the signal processing device 8 as a function of the two input signals E1, E2, which is converted into an output sound signal (not shown) by an output converter 12 of the hearing device 2. In the present case, the output converter 12 is provided by a loudspeaker.

[0045] Now, a first intermediate signal Z1 (dashed line) is generated from the first input signal E1 and the second input signal E2 by means of a time-delayed superposition. The first intermediate signal Z1 is generated here as a cardioid signal 16, the directional characteristic 18 of which is rotationally symmetrical about the connecting line 20 through the first input converter M1 and the second input converter M2 in an ideal case (at Figure 1 In the image plane of FIG. 2 , only the axis symmetry about the connecting line 20 can be seen. Likewise, the second intermediate signal Z2 (dotted line) is generated from the first input signal E1 and the second input signal E2 by superposition of additional time delays. The second intermediate signal Z2 is generated here as an anti-cardioid signal 22, whose directional characteristic 24 is also rotationally symmetrical about the connecting line 20. Furthermore, the first intermediate signal Z1 and the second intermediate signal Z2 are ideally mirror-symmetrical to each other about the symmetry plane 26 of the first input converter M1 and the second input converter M2 ( Figure 1 A section through the symmetry plane 26 and the image plane are shown).

[0046] According to its directional characteristic 18, the first intermediate signal Z1 has a maximum sensitivity in a maximum direction 28 and a minimum sensitivity in a minimum direction 30 opposite to the maximum direction 28. In the minimum direction 30, there is an ideally complete attenuation in the first intermediate signal Z1. The maximum direction 28 and the minimum direction 30 extend along the connecting line 20. According to its directional characteristic 24, the second intermediate signal Z2 has a maximum sensitivity in a maximum direction 32 and a minimum sensitivity in a minimum direction 34. The maximum direction 32 of the second intermediate signal Z2 coincides with the minimum direction 30 of the first intermediate signal Z1, and the minimum direction 34 of the second intermediate signal Z2 coincides with the maximum direction 28 of the first intermediate signal Z1.

[0047] The hearing device 2 is designed so that when the user normally wears it, the connecting line 20 is oriented in the front direction of the user. When using the hearing device 2, it is usually the case that the user and a conversation partner are in a conversation. Accordingly, the user orients his line of sight and therefore his front direction toward the conversation partner, whereby, due to the spatial relationship just described, the maximum direction 28 of the first intermediate signal Z1 is also oriented toward the conversation partner as the main useful signal source 36 (schematically shown here by a loudspeaker symbol). If an interference signal 38 of an interference signal source 40 now appears in the ambient sound, the mentioned interference signal 38 is suppressed with the aid of an adaptive directional microphone. In this case, usually, a first superposition U1 is formed from the first intermediate signal Z1 and the second intermediate signal Z2 by minimizing the signal energy of the first superposition U1 using a first superposition parameter a1, and the formula is:

[0048] U1=Z1–a1·Z2

[0049] Under the assumption Figure 1 In the case where the maximum direction 28 of the first intermediate signal Z1 is oriented toward the conversation partner as useful signal source 36 and the second intermediate signal Z2 has an ideally complete attenuation in its minimum direction 34 (which is also oriented toward the conversation partner), in the mentioned minimization of the signal energy, the contribution of the conversation partner is not influenced by the second intermediate signal Z2 due to its assumed shielding. The minimization of the signal energy therefore affects only the interference signal 38 of the interference signal source 40.

[0050] for Figure 1 The situation shown is that the interference signal source 40 is relative to the front direction (and therefore the direction of the useful signal source 36 or according to Figure 1 The maximum direction 28 of the first intermediate signal Z1 is arranged at right angles, which is for the first superposition U1 (as shown in FIG. Figure 2 ) leads to complete attenuation in a first target direction 42 oriented toward the interference signal source 40. However, there are situations in which a complete attenuation of the interference signal 38 of the interference signal source 40 in the output signal generated by the adaptive directional microphone is undesirable, for example for pedestrians in street traffic, where timely hearing of other road users is important for safety, but in conversations with multiple conversation partners, where it may also be advantageous to at least perceive the interruptions of another conversation partner to whom the user is currently not paying attention (and therefore not focusing his or her gaze), so that, if necessary, he or she can turn to this other conversation partner in order to listen attentively.

[0051] To achieve this, according to Figure 1 A method for directional signal processing is implemented in a hearing device 2 according to Figure 3 This is explained with the help of a corresponding block diagram. Figure 1 As shown in FIG. 1 , the first input converter M1 and the second input converter M2 of the hearing device 2 generate the first input signal E1 or the second input signal E2 from the ambient sound 4. A first intermediate signal Z1 and a second intermediate signal Z2 are generated from the first input signal E1 and the second input signal E2 by superposition 44 of time delays, which are only schematically shown here. The first intermediate signal Z1 is generated here as a cardioid signal 16, and the second intermediate signal Z2 is generated as a cardioid signal according to FIG. Figure 1 An anti-cardioid signal 22 is generated.

[0052] Now in order to Figure 2 To obtain information about the first target direction 42 of the interference signal source 40 (and thus to implicitly determine the first target direction 42), according to Figure 2 The first superposition U1 is formed from the first intermediate signal Z1 and the second intermediate signal Z2 by means of the adaptive directional microphone 46. In this case, the first superposition U1=Z1-a0·Z2 provides the method with a temporary superposition parameter a0 (which corresponds to Figure 2 The first superposition parameter a1 of the first superposition U1) is, according to Figure 2 , through the relationship with the first intermediate signal Z1 and the second intermediate signal Z2, the first target direction 42 of the interference signal source 40 is also implicitly determined by the temporary superposition parameter a0.

[0053] For the first target direction 42 mentioned, a first value g1>0 of the gain is now predefined, which the signal to be formed from the two intermediate signals Z1, Z2 should have. Figure 2 As shown, the gain is assumed to be zero (complete attenuation). With the aid of the adaptive directional microphone 48, a second superposition U2=Z1-a·Z2 is now formed from the first intermediate signal Z1 and the second intermediate signal Z2 using the superposition parameter a, which is usually complex. More precisely, under the mentioned constraint that the gain in the first target direction 42 is g1, the first target direction is determined by the temporary superposition parameter a0 of the first superposition U1. If necessary, with the aid of further signal processing steps, such as band-dependent amplification and / or compression, an output signal 10 is generated from the second superposition U2, which is based on Figure 1 The output signal is converted into an output sound signal by the output converter 12.

[0054] From the cardioid signal 16 and according to Figure 1 The anti-cardioid signal 22 provides the knowledge that the first and second intermediate signals Z1 and Z2 can be determined (depending on the running time difference T between the two input converters M1 and M2) relative to the second superposition U2 from the angle Transfer function of the incident sound signal (relative to the front direction) The transfer function mentioned can be expressed as

[0055]

[0056] In the case of the approximation ωT<<1, this approximation is particularly advantageous for low frequencies and run-time differences T (for hearing instruments, T is between 10 -5 seconds, so the approximation is good for a large part of the audible spectrum), the above formula can therefore be approximated by

[0057] (i)G(ω,φ)=|ωT(1+cosφ)-a·ωT(1-cosφ)|

[0058] Therefore, in the front direction (i.e. for ), (the distance d between the two input converters M1 and M2 and the wave number k) are independent of a. It can now be shown that for real-valued superposition parameters The transfer function provided in equation (i) is at the angle approaches zero, the applicable ones are:

[0059] (ii) or

[0060] For a generally complex-valued superposition parameter a=aRe+i·aIm, the transfer function (i) can be used to achieve the following in the first target direction by making the transfer function equal to g1. The gain on the first target direction 42 is g1. Due to the additional degree of freedom provided by the imaginary part aIm, the superposition parameter a is not yet completely determined by the first value g1 of the gain in the first target direction provided by a0. It can be shown here that in the complex plane of aRe, aIm, for the provided first value g1 and the provided first target direction 42 (and thus the provided or a0) allows the real and imaginary parts to form a circle around a0 with radius g1(a0+1):

[0061] (iii) (aRe-a0) 2 +aIm 2 =g1 2 (a0+1) 2

[0062] In this case, a plurality of special cases may now occur, or they may be realized accordingly for the second superposition U2 .

[0063] As a constraint, it can be required that the gain in the first target direction 42 should also form a global minimum, which, however, is now (with respect to Figure 2In the case shown, however, the value 0 is no longer used, but a first value g1>0 is used. Figure 4 This is shown in .

[0064] The general angle at which the transfer function is minimum can be determined by minimizing the absolute value square of the transfer function according to equation (i) (with the complex number a) Here, Substituting the value thus determined into the square equation (i) first provides the following conditions

[0065]

[0066] And equating the transfer function to the first value g1 required for the (global minimum) gain in the first target direction 42 provides the following dependency

[0067]

[0068] where ε = g1 2 / (1-g1 2 ). Substituting the intermediate result provided in equation (v) into equation (iv) and expressing the minimum angle according to equation (iii) by the corresponding real-valued superposition parameter a0 (For this superposition parameter, The gain will disappear when and corresponding substitutions from equation (iii)) provide

[0069]

[0070] The superposition parameter a=aRe+i·aIm thus results from the relations mentioned in equations (v) and (vi), Figure 4 The second overlay shown is based on this overlay parameter.

[0071] according to Figure 5 A further possibility is shown: In this case it is not necessary that the gain in the second target direction should be minimum, but rather that the resulting second superposition U2 = Z1 −a·Z2 should have a maximum directivity index (DI).

[0072] Here, the transfer function can be calculated in the maximum direction (i.e., in the direction according to Figure 1 DI is determined by the square of the absolute value of the first intermediate signal Z1 in the maximum direction 28), and DI is normalized on the integral of the square of the absolute value of the transfer function in all spatial directions. DI is usually defined here by the decimal logarithm of the variables mentioned:

[0073]

[0074] wherein the integral in the denominator is carried out over a normalized unit sphere, so that for an omnidirectional signal DI = 0 results. It can be shown that the DI according to equation (vi) can be expressed in terms of the superposition parameters a = aRe + i·aIm as

[0075] (viii) DI = –10·log 10 (aRe 2 –aRe + 1 + aIm 2 ) + 10·log 10 (3).

[0076] The base of the logarithm remains constant for aRe, aIm, which describe a circle around the point (0.5, 0) in the complex plane. For the point aRe = 0.5, aIm = 0, DI has its maximum value. In this case, the associated second superposition U2 forms a directional characteristic in the form of a supercardioid. Thus, it can be seen from equation (viii) and equation (iii) that for a0 ≠ 0 and the superposition parameter a according to equation (iii), the DI according to equation (viii) is maximized by the real-valued superposition parameter a = aRe, which can be determined according to equation (iii), and thus

[0077] (ix) aRe = a0 ± g1·(a0 + 1)

[0078] This results in a desired gain with a first value g1 in the first target direction 42, while complete attenuation (i.e., a gain with a second value g2 = 0) typically occurs in the second target direction 50.

[0079] The plus sign in equation (ix) applies here for a0 < 0.5, and the minus sign applies for a0 > 0.5. In order to avoid a discontinuity in aRe in the case of a movable interference signal source 40 and thus a variable a0 near a0 = 0.5, regularization can be carried out in a manner yet to be described (not shown in detail), which first sets the value for aRe according to equation (ix) for the ranges a0 ≤ 0.5 - d1 and a0 ≥ 0.5 + d2, where d1, d2 > 0, preferably d1 = d2, and particularly preferably d1, d2 << 1. In the ranges 0.5 - d1 < a0 < 0.5 and 0.5 < a0 < 0.5 + d2, a non-zero imaginary part aIm ≠ 0 can be set here such that the real part aRe resulting from equation (iii) extends along the maximum gradient of the DI according to equation (ix). It can be shown that this is the case if for a0 = 0.5, the value aRe = (1 - 3·g1 2 ) / 2.

[0080] Figure 6Another possibility is shown. There, the superimposition parameter a for the second superimposition U2 is determined such that a gain with the value g1 occurs in the first target direction 42. Furthermore, a second value g2 < g1 is preset as the global minimum of the gain, which should not be undershot in any direction. In particular, there is a second target direction 50 in which the gain, i.e., the gain factor, has exactly the second value g2. In this case, the additional degree of freedom in the imaginary part alm of the superimposition parameter a is used to preset, in addition to the preset first value g1 of the gain in the first direction, the second value g2, below which the gain cannot fall in any direction.

[0081] Here, to determine the superimposition parameter a, the relationship between the real and imaginary parts of the gain at the minimum finite gain provided in equation (v) is substituted into the general equation (iii) to preset the first value g1, but here the parameter ε = g2 2 / (1 - g2 2 ) is to be substituted in order to take into account the second value g2 of the gain here, which should now form the global minimum (in equation (v), the global minimum is provided by the first value g1, which now determines the gain in the first target direction).

[0082] This results in a quadratic equation for the real part aRe of the superimposition parameter a, the positive or negative solution of which is selected according to the value a0 (a0 > 0.5 or a0 < 0.5) and thus according to the first target direction 42, in which the first value g1 of the gain is determined. Directly in the vicinity of a0 = 0.5, regularization of the type already described can be implemented to avoid discontinuities. Now the imaginary part aIm can be determined according to equation (v) (where ε = g2 Figure 5 / (1 - g2 2 )) 2 ).

[0083] Although the invention has been illustrated and described in detail by the preferred embodiments, the invention is not limited to the disclosed examples, and other variants can be derived therefrom by those skilled in the art without departing from the scope of protection of the invention.

[0084] List of reference numerals

[0085] 1 Acoustic system

[0086] 2 Hearing device

[0087] 4 Ambient sound

[0088] 6 Control unit

[0089] 8 Signal processing device

[0090] 10 Output signal

[0091] 12 Output Converter

[0092] 16 Cardioid signal

[0093] 18 Directional characteristic (of the first intermediate signal)

[0094] 20 Connection wires

[0095] 22 Anti-cardioid signal

[0096] 24 Directional characteristic (of the second intermediate signal)

[0097] 26 Symmetry plane

[0098] 28 Maximum direction (of the first intermediate signal)

[0099] 30 Minimum direction (of the first intermediate signal)

[0100] 32 Maximum direction (of the second intermediate signal)

[0101] 34 Minimum direction (of the second intermediate signal)

[0102] 36 Useful signal sources

[0103] 38 Interference signal

[0104] 40 Interference signal source

[0105] 42 First target direction

[0106] 44 Superposition of time delays

[0107] 46 Adaptive directional microphones

[0108] 48 Adaptive directional microphones

[0109] 50 Second target direction

[0110] A. Overlay parameters

[0111] a0 Temporary overlay parameter

[0112] E1, E2 First or second input signal

[0113] g1, g2 first or second value (of gain)

[0114] M1, M2 First or second input converter

[0115] U1, U2 first or second superposition

[0116] Z1, Z2 First or second intermediate signal

Claims

1. A method for directional signal processing of an acoustic system (1), in, A first input signal (E1) is generated from the ambient sound (4) by a first input transducer (M1) of the acoustic system (1), wherein a second input signal (E2) is generated from the ambient sound (4) by a second input transducer (M2) of the acoustic system (1), wherein a first intermediate signal (Z1) and a second intermediate signal (Z2) are generated according to a first input signal (E1) and a second input signal (E2), respectively; wherein a temporary superposition parameter (a0) for a first superposition (U1) of the first intermediate signal (Z1) and the second intermediate signal (Z2) is obtained such that for the first superposition (U1) the attenuation in the first target direction (42) has a maximum value, wherein the superposition parameter (a) is formed according to the temporary superposition parameter (a0) so that for a second superposition (U2) of the first intermediate signal (Z1) and the second intermediate signal (Z2) formed using the superposition parameter (a), the gain in the first target direction (42) has a predetermined first value (g1) greater than zero, and Therein, an output signal (10) of the acoustic system (1) is formed according to the second superposition (U2), and - wherein the superposition parameter (a) is formed so that for the second superposition (U2) the gain in the first target direction (42) has a global minimum, the global minimum having a predetermined first value (g1), and wherein the real part of the superposition parameter (a) is formed by a linear function of the temporary superposition parameter (a0), the linear function being related to the first value (g1) of the gain in the first target direction (42), the real part transforming into the temporary superposition parameter (a0) when the first value (g1) approaches zero, and / or the imaginary part of the superposition parameter (a) is linearly related to the real part and approaches zero when the first value (g1) approaches zero, and / or wherein the superposition parameters (a) are formed such that for the second superposition (U2) the gain in the first target direction (42) has a predetermined first value (g1) and the gain in the second target direction (50) has a predetermined second value (g2).

2. The method according to claim 1, wherein: A cardioid signal (16) is generated as a first intermediate signal (Z1), and an anti-cardioid signal (22) is generated as a second intermediate signal (Z2).

3. The method according to claim 1, wherein: The second value (g2) is smaller than the first value (g1), and / or the second value (g2) is equal to zero.

4. The method according to claim 3, wherein: The real and imaginary parts of the superposition parameter (a) can be described according to a circle in the complex plane having the temporary superposition parameter (a0) as origin and a radius, the square of which depends on the square of the first value of the gain (g1) and the square of the temporary superposition parameter (a0).

5. The method according to claim 1 or 2, wherein: The superposition parameter (a) is formed so that the signal generated by the second superposition (U2) has a maximum directivity index.

6. The method according to claim 5, wherein: In the context of the preset temporary superposition parameter (a0), the regularization of the superposition parameter (a) is performed around the critical value, so that - presetting the value of the superposition parameter (a) to be applied for the critical value of the temporary superposition parameter (a0), and - For values ​​from a preset environment around the critical value, the temporary superposition parameter (a0) is continuously mapped to the superposition parameter (a).

7. The method according to claim 1 or 2, in, forming the superposition parameters (a) so that for the second superposition (U2) the gain in the first target direction (42) has a predetermined first value (g1) and the gain in the second target direction (50) has a predetermined second value (g2), The gain in the second target direction (50) has a global minimum value, and the global minimum value has a predetermined second value (g2).

8. An acoustic system (1) comprising at least one first input converter (M1) for generating a first input signal (E1) from ambient sound (4) and a second input converter (M2) for generating a second input signal (E2) from ambient sound (4), and further comprising a control unit (6) designed to implement a method according to any of the preceding claims.

9. The acoustic system (1) according to claim 8, comprising a hearing device (2), in which the first input transducer (M1) and the second input transducer (M2) are arranged.

Images