SENSOR ARRANGEMENT WITH OPTIMIZED GROUP RUNTIME AND SIGNAL PROCESSING METHOD

Abstract

Circuit arrangement (200), comprising: a sensor, set up to provide an analog signal (204); an analog-to-digital converter (202), set up to receive the analog signal (204) and provide an initial signal (206); a first filter (208) configured to receive a signal based on the first signal (206) and to provide a second signal (210); where the first filter (208) is set up such that: • that the second signal (210) is passed through unamplified in a frequency range from 20 Hz to 10 kHz, and • that the second signal (210) has a gain greater than 0 dB at least from a specified frequency greater than 20 kHz; and a second filter (214) connected to the analog-to-digital converter (202) and configured to provide a third signal (216) based on the first signal, wherein the second filter (214) is configured as a frequency-selective filter to process the signal received by the second filter, that the amplitude of the signal received by the second filter (214) is reduced in a specified frequency range, wherein the specified frequency range has a frequency greater than 20 kHz.

Description

AREA

[0001] Embodiments relate to a sensor arrangement with optimized group delay and a method for signal processing. BACKGROUND

[0002] Sensor arrangements, such as microphones, are used to record ambient noise or sound. Examples are disclosed in US 2014 / 0 177 874 A1, DE 101 03 812 A1, and US 2011 / 0 099 213 A1. To provide good quality recorded sound or to meet customer requirements, high linearity, high signal-to-noise ratios (SNR), or conformity to a predefined spectral mask for the microphone's response function may be necessary.

[0003] A conventional 900 microphone, as shown in the block diagram in Fig. Figure 9 shows a microphone diaphragm 902, for example a microelectromechanical membrane (MEMS). The diaphragm 902 is deflected from a rest position by means of sound-induced pressure fluctuations, thereby generating an analog electrical signal that is amplified by an amplifier 904 or a readout circuit, for example a source follower. The signal from the amplifier 904 is detected at a sampling frequency Fs (706) by a sensor circuit 908. The sensor circuit 908 includes an analog-to-digital converter 910, which converts the signal from the amplifier 904 into digital signals. The sensor circuit also includes a digital filter 912, which blocks high-frequency digital signals from the analog-to-digital converter 910 (low-pass filter). The filtered digital signals are converted into a customer-specific 1-bit output signal 916 by a modulator 914 connected to the digital filter 912.

[0004] The low-pass filter is necessary to stabilize the setup and attenuate any peaks in the MEMS frequency response. This setup introduces a group delay.

[0005] In some applications, such as headsets, acoustic suppression of ambient noise using a loudspeaker driven out of phase is necessary. A microphone measures the disruptive ambient noise, which is then digitally processed and sent to the loudspeaker. The group delay of the microphone array plays a crucial role in the suppression. This group delay is the sum of the group delays of the microphone, the group delay of the digital signal processing in the sensor circuit, and the group delay of the loudspeaker. SUMMARY

[0006] There is a need to provide a sensor array with optimized group runtime.

[0007] Such a need can be met by the subject matter of one of the claims. BRIEF DESCRIPTION OF THE FIGURES

[0008] Some exemplary embodiments of devices and / or methods are described below only as examples and with reference to the accompanying figures. These show: Fig. 1 a diagram of the group delay of a sensor arrangement according to various embodiments; Fig. 2 a block diagram of an exemplary embodiment of a sensor arrangement; Fig. 3A-D block diagrams of different embodiments of a sensor arrangement; Fig. 4A-C Block diagrams of various embodiments of a sensor arrangement; Fig. 5A and Fig. 5B Representations of the frequency response of a filter of exemplary embodiments of a sensor arrangement; Fig. 6A, B Diagrams of the group delay of sensor arrangements according to various embodiments; Fig. 7 a diagram of the group delay of a sensor arrangement according to various embodiments; Fig. 8 a diagram of a method for signal processing according to various embodiments; and Fig. 9 a block diagram of a conventional microphone. DETAILED DESCRIPTION

[0009] Fig. Figure 1 shows a diagram 100 of the group delay 120 in units of µs of a sensor array, for example, a microphone array, as a function of the signal frequency 110 in units of Hz for different attenuations or suppressions of ambient noise (-10 dB, -15 dB, -20 dB, -30 dB), where -10 dB is the lowest attenuation and -30 dB the highest. The diagram 100 shows that the required group delay increases with increasing attenuation. In other words, to achieve greater attenuation of ambient noise, it is necessary to reduce the group delay of the array over a wide frequency range. Attenuating the ambient noise also improves the signal-to-noise ratio. Good attenuation can be achieved by minimizing the group delays of the components of the sensor array.

[0010] Fig. Figure 2 illustrates a block diagram of a section 200 of an embodiment of a sensor arrangement according to various embodiments.

[0011] The circuit arrangement includes a sensor (not illustrated), an analog-to-digital converter 202, a filter arrangement 212 and an interface 222.

[0012] In various embodiments, the sensor arrangement can further include a modulator 218 and / or a decimation filter, which is described in more detail below.

[0013] In various embodiments, the filter arrangement 212 includes a first filter 208 which is configured to reduce the group delay of a passing signal, as described in more detail below.

[0014] In further embodiments, the filter arrangement 212 can have a second filter 214, which can be intuitively configured as a low-pass filter or a band-pass filter.

[0015] To illustrate, the first filter 208, the second filter 214 and the modulator 218 are connected to the analog-to-digital converter 202, so that the signal 206 provided by the analog-to-digital converter 202 is processed or converted into a signal 220 provided by the modulator 218.

[0016] The sensor is configured to provide an analog signal.

[0017] The analog-to-digital converter 202 is configured to receive the analog signal and provide an initial signal 206. Optionally, the analog signal from the sensor can be amplified by an amplifier, for example a source follower, before it is received by the analog-to-digital converter 202.

[0018] The first filter 208 is configured to receive a signal and provide a second signal 210. The signal received by the first filter has a first group delay. The first filter 208 is configured to process the signal received by the first filter 208 such that the second signal 210 output by the first filter 208 has a second group delay, for example, by means of suitable filter coefficients, as described in more detail below. The second group delay is smaller than the first group delay. For example, in a frequency range of approximately 20 Hz to approximately 4 kHz, the second group delay is smaller than the first group delay by a value in the range of approximately -0.01 µs to approximately -25 µs, for example, by a value in the range of approximately -0.01 µs to approximately -15 µs, or by a value in the range of approximately -5 µs to approximately -11 µs, for example, by -10 µs.

[0019] The second filter 214 is set up to receive a signal and is set up to provide a third signal 216.

[0020] The second filter 214 is a frequency-selective filter designed to reduce the amplitude of the signal received by the second filter 214 in a specified frequency range, wherein the specified frequency range has a frequency greater than 20 kHz.

[0021] The second filter 214 is, for example, a low-pass filter, a notch filter, a band-pass filter, or a band-stop filter. The second filter 214 is designed to reduce the amplitude of a high-frequency component of the signal output by the first filter. This prevents overloading of the downstream modulator.

[0022] For example, the second filter 214 is a low-pass filter and is configured to reduce the amplitude of the received signal above a certain frequency, where the frequency is, for example, 22 kHz, 40 kHz, 50 kHz, 60 kHz, 70 kHz, or 80 kHz. This frequency can also be referred to as the cutoff frequency of the second filter. Above the cutoff frequency, the amplitude can be negligible or essentially negligible, thus preventing overloading of the modulator. Different cutoff frequencies are associated with different group delays for the processed signal. The higher the cutoff frequency, the lower the group delay of the processed signal. The higher the cutoff frequency, the greater the amplitude of the unfiltered high-frequency component of the signal received by the modulator. The type of second filter and / or its cutoff frequency can be selected, i.e., optimized, for specific applications.Application-specific optimization can be performed, for example, depending on the type of modulator and the sampling frequency. For instance, a higher cutoff frequency can be selected if the modulator cannot process high-frequency signals, such as if it has an integrated bandpass or bandstop filter.

[0023] Modulator 218 is configured to receive a signal and to provide a fourth signal 220. The signal received by modulator 218 has the second group delay.

[0024] The fourth signal 220 provided by the modulator has a third group delay. The analog signal has a fourth group delay. The third group delay of the analog signal is smaller than the fourth group delay of the fourth signal 220 provided by the modulator, at least in a frequency range from approximately 20 Hz to approximately 4 kHz.

[0025] In the Fig. In the illustrated example, the first filter 208 is set up to receive the first signal, the second filter 214 to receive the second signal, and the modulator 218 to receive the third signal. In other words, the sensor provides the analog signal. The analog-to-digital converter 202 receives the analog signal and provides a first signal 206. The first filter 208 receives the first signal 206 and provides the second signal 210. The second filter 214 receives the second signal 210 and provides the third signal 216. The modulator 218 receives the third signal 216 and provides the fourth signal 220.

[0026] Furthermore, interface 222 can be configured to receive the fourth signal 220, and to process the fourth signal 220 and output a fifth signal 224. For example, interface 222 can be configured to split the output signal across multiple channels or pins.

[0027] Interface 222 can provide the fifth signal 224 to an external environment of a sensor assembly, or be configured to do so, for example, by having a socket. An acoustic and / or optical output device, such as a loudspeaker or a display, can be connected to the socket. The output device may include further filters and / or signal processing components that further process and modify the signal provided at the interface.

[0028] The signals recorded and provided by the filters can each be a digital signal.

[0029] The first filter 208 reduces the group delay of the signal received by the first filter. This allows the fourth signal 220, provided by the modulator, to have a group delay that is smaller than the group delay of the analog signal provided by the sensor. This makes it possible to attenuate or suppress the noise of the fourth signal more effectively. As a result, the signal-to-noise ratio (SNR) of the sensor array can be improved. The modulator can be optional if an unmodulated multi-bit (also referred to as m-bit or multi-bit) output signal 224 is desired for a specific application.

[0030] The second filter 214 can be used to reduce the amplitude of a high-frequency signal or signal component of the signal received by the modulator 218 and / or provided by the interface 222. This prevents the modulator 218 from being unnecessarily overloaded with data and signals. In a microphone application, the second filter 214 can be a low-pass filter with a cutoff frequency above the upper limit of human hearing, i.e., above approximately 22 kHz. This allows high sound quality to be maintained while simultaneously reducing the signal processing requirements of the modulator 218.

[0031] The reduction of group delay is achieved intuitively by using a digital filter with negative group delay. Negative group delay means that the signal output by the first filter has a lower group delay than the signal received by the first filter.

[0032] The sensor array is a linear, time-invariant system. Due to the linear behavior of the sensor array, the signal output by the first filter has the same frequency as the signal received by the first filter. However, the two signals differ in amplitude and phase. The ratio of the amplitudes of the received signal (input signal) and the output signal as a function of frequency is the amplitude response. The difference in phase between the input signal and the output signal as a function of frequency is the phase response.

[0033] The first filter is configured, for example by appropriately selecting the filter coefficients, such that the phase response and the amplitude response are positive, meaning they exhibit a positive slope with increasing frequency, for example in the high-frequency range. In other words, the first filter amplifies the recorded signal in the high-frequency range. The high-frequency range begins, for example, at approximately 20 kHz.

[0034] The group delay is defined as -1 times the derivative of the phase response with respect to frequency. Therefore, the positive phase response of the first filter results in a negative group delay.

[0035] This can be illustrated by using a first filter, for example, an FIR filter. The FIR filter is configured such that a low-frequency signal has an amplitude gain of approximately 0 dB. For example, a signal is considered low-frequency in a frequency range of approximately 20 Hz to approximately 4 kHz. The same FIR filter is further configured such that a high-frequency signal has an amplitude gain greater than approximately 0 dB. For example, a signal with a frequency greater than approximately 4 kHz, such as from approximately 20 kHz and above, is considered high-frequency. The FIR filter can also be configured to have a sub-harmonic amplitude gain in a certain frequency range (see Fig. 4B). In the subharmonic range, the FIR filter can exhibit negative gain, i.e., attenuation. The negative gain can, for example, be in the region of the resonant frequency of an oscillating sensor in the sensor array.

[0036] Such an FIR filter can be implemented at a given sampling frequency, for example, by appropriately choosing the filter coefficients.

[0037] In one embodiment, the circuit arrangement is configured as, or includes, a pressure sensor arrangement. Alternatively or additionally, the circuit arrangement is configured as, or includes, a microphone arrangement.

[0038] The sensor can, for example, consist of a membrane. A deflection of the membrane from a rest position can generate the analog signal. The membrane is, for example, a microelectromechanical structure (MEMS) or incorporates one. Alternatively, or in other words, the sensor can be or incorporate a microelectromechanical structure.

[0039] In various embodiments, the microphone is used to record ambient sound, speech, music, or the like, and to provide a microphone signal. Recording or providing a microphone signal can be understood as providing an electrical signal that depends on the ambient sound, or in other words, on the sound pressure acting on the microphone. Various microphone types can be used, such as electret microphones or other condenser microphones. A specific example is a silicon microphone implemented as a microelectromechanical system. This means that the diaphragm and other components that make up the microphone can be manufactured using processing steps and techniques commonly employed in microprocessor manufacturing.

[0040] Some of the microphone's characteristics that relate the resulting microphone signal to the acting sound pressure can be adjusted by hardware characteristics of the microphone itself, such as the rear volume or the stiffness of a microphone diaphragm.

[0041] According to the exemplary embodiment from Fig. 2. The analog-to-digital converter 202 is a multi-bit converter, so the first signal is a multi-bit representation.

[0042] In one example, the first filter 208 is a finite impulse response filter (FIR filter) or has one, for example, a second-order FIR filter. Intuitively, the first filter is configured such that it modifies, i.e., reduces, the group delay of the captured signal. The high-frequency component of the captured signal, i.e., the wave packet enclosed by the envelope, is not altered by the first filter. The envelope is the amplitude profile of the wave packet and propagates with the group delay.

[0043] If the sensor's response exhibits a resonant characteristic or a resonance peak within the investigated spectrum, an FIR filter may be able to model the inverse of the microphone 102's frequency response. According to some exemplary embodiments, the coefficients of the FIR filter are programmable or variable. This can serve to maintain the desired filter characteristics when the tone control device is operated at different sampling frequencies.

[0044] Alternatively, the first filter 208 is a recursive filter or has one.

[0045] The first filter 208 has a corner frequency in a range of approximately 10 kHz to approximately 40 kHz, for example in a range of approximately 16 kHz to approximately 30 kHz, for example in a range of approximately 16 kHz to approximately 22 kHz, for example in a range of approximately 25 kHz to approximately 35 kHz.

[0046] If the sensor arrangement is a microphone arrangement, the first filter 208 can, for example, have a corner frequency in the range of approximately 16 kHz to approximately 22 kHz. If the sensor arrangement is a pressure sensor arrangement, the first filter 208 can, for example, have a corner frequency in the range of approximately 25 kHz to approximately 30 kHz. One reason for the different corner frequencies could be the application-specific design of the sensor and its resonant frequency.

[0047] The first filter 208 is set up such that the first signal in a frequency range of approximately 20 Hz to approximately 22 kHz has a group delay of less than or equal to 0 seconds.

[0048] The first filter 208 is configured such that the first signal in a frequency range of approximately 20 Hz to approximately 22 kHz has a group delay in a range of 0 seconds to approximately -1 · 10 -5 seconds.

[0049] The first filter 208 is configured such that the first signal exhibits a group delay minimum within a specific frequency range. The circuit arrangement may exhibit a resonant frequency within this frequency range. For example, the group delay minimum is in a frequency range from approximately 10 kHz to approximately 30 kHz.

[0050] A weaker attenuation of the input signal can result in a lower group delay. Conversely, a stronger attenuation can lead to a relatively higher group delay, which, however, can be less than 0 µs. Thus, it is possible to select, depending on the application, whether stronger attenuation of a resonant signal from the sensor array is desired in a given frequency range, or alternatively, a lower group delay in the same or a different frequency range. This can be achieved at a given sampling frequency by changing the attenuation of the first filter.

[0051] The first filter 208 can be set up such that the first signal in a frequency range of approximately 20 Hz to approximately 20 kHz has a gain that is less than or equal to 0 dB.

[0052] The second filter 214 is a low-pass filter or incorporates one. A low-pass filter can be implemented as a digital filter. For example, the second filter 214 is a third-order low-pass filter or incorporates one.

[0053] The second filter 214, for example, has a corner frequency of, for example, 22 kHz, 40 kHz, 50 kHz, 60 kHz, 70 kHz or 80 kHz.

[0054] In one embodiment, the corner frequency of the second filter 214, i.e., the frequency from which the second filter 214 blocks transmission, corresponds to the corner frequency of the first filter 208.

[0055] The corner frequency of the second filter 214 or the corner frequency of the first filter 208 can also be referred to as the cutoff frequency, pass frequency or threshold.

[0056] The signal received by modulator 218 has a first word width. Modulator 218 is configured to process the received signal such that the fourth signal 218 provided by modulator 218 has a second word width. The second word width is smaller than the first word width; for example, the first word width is greater than 4 bits, greater than 8 bits, greater than 20 bits; and the second word width is smaller than 8 bits, less than 4 bits, 1 bit.

[0057] The fourth signal can be provided in any number of different representations. For example, a single-bit protocol can be used, so that the fourth signal is provided as a bitstream. Other implementations can provide the fourth signal as a sequence of bytes, for example in hexadecimal or decimal. Still other embodiments can provide a fourth signal as an analog signal.

[0058] Some exemplary embodiments provide a fourth signal in a single-bit representation and can be provided by means of the modulator 218s to provide the single-bit representation from a multi-bit representation that can be used in previous processing steps within the sensor arrangement.

[0059] A sensor arrangement according to some exemplary embodiments further comprises one or more terminals to provide the possibility of connecting all components within the sensor arrangement in a single assembly step by means of the terminal(s) to other circuit arrangements, printed circuit boards or the like.

[0060] Some exemplary embodiments of a sensor arrangement comprise a common housing arrangement that at least partially encloses the sensor and the other components, for example, the amplifier, source follower, ADC converter, filters 208, 214, and modulator 218, wherein the common housing arrangement has feed connectors for the electrical connection of all components to other circuit arrangements. A sensor arrangement according to some exemplary embodiments can be understood as a single unit that can be treated as a discrete, independent device, such that the components within the sensor arrangement can be connected to other devices or circuit arrangements by electrically connecting the sensor arrangement as a whole to the other circuit arrangements.This can make it possible to reduce the number of connections used within an application, for example by using a single supply voltage connection for the sensor and the other components within the housing.

[0061] According to one embodiment, the modulator 218 of the sensor arrangement converts a multi-bit representation into a single-bit representation.

[0062] According to the embodiment from Fig. 2. The sampling frequency Fs of the analog-to-digital converter 202 is variable, so that the sensor arrangement can support multiple sampling frequencies. According to some exemplary embodiments of sensor arrangements, a property of the sensor arrangement 104 is variable, which makes it possible to achieve similar modification properties of the sensor arrangement for different sampling frequencies of the analog-to-digital converter 202. The sampling frequency, for example, has a value in the range of approximately 1 MHz to approximately 4 MHz.

[0063] In one embodiment, the first filter 208 is configured to receive the third signal, the second filter 214 to receive the first signal 206, and the modulator 218 to receive the second signal 210. In other words, the sensor provides the analog signal. The analog-to-digital converter 202 receives the analog signal and provides a first signal 206. The second filter 214 receives the first signal 206 and provides the third signal 216. The first filter 208 receives the third signal 216 and provides the second signal 210. The modulator 218 receives the second signal 210 and provides the fourth signal 220.

[0064] Fig. Figures 3A to 3D show block diagrams of various embodiments of a sensor arrangement 200, as described above.

[0065] In the Fig. In the embodiment illustrated in Figure 3A, no low-pass filter, modulator, or decimation filter is provided in the sensor arrangement 200. However, such components may be provided downstream of the interface 222. For example, in the case of a microphone arrangement, a loudspeaker connected to the interface 222 may include a low-pass filter, a modulator, and / or a decimation filter.

[0066] In other words, the sensor provides the analog signal. The analog-to-digital converter 202 receives the analog signal and provides a first signal 206. The first filter 208 is configured to receive a signal based on the first signal 206, for example, the first signal 206, and provides the second signal 210, which may substantially correspond to the signal 220 provided at interface 220. The first signal, the second signal, and / or the signal provided by the sensor array 200 at interface 222 may each be a multi-bit signal, for example, with the same or substantially the same word width.

[0067] In the Fig. The illustrated embodiment in 3B is in the sensor arrangement 200 made of Fig. 3A further provides a second filter 214. The sensor arrangement 200 is, in various embodiments, a linear, time-invariant system. The second filter 214 can thus be arranged between the analog-to-digital converter 202 and the first filter 208 and / or between the first filter 208 and the interface 222 with respect to the signal flow. The signal provided at the interface 222 is identical in both arrangements. The second filter 214 can, for example, be a multi-stage filter, such as a multi-stage low-pass filter.

[0068] This allows, for example, the reduction of the amplitude of the high-frequency signal component of the signal provided at interface 222. In a microphone setup, this can improve sound quality. In other words, the second filter 214 can limit the high-frequency signal components that carry no or essentially no information, also known as "out-of-band" energy. In some applications, the maximum amplitude of the high-frequency signal components is limited or a limitation is required; for example, the amplitude response should be below a predefined upper limit above a frequency of 40 kHz. In this case, the group delay is adjusted according to... Fig. 6A, which features a low-pass filter as a second filter, is used to meet the "Out Of Band" specification. The second filter increases the group duration.

[0069] In this embodiment, the signal provided at interface 222 can be a multi-bit signal. For example, the multi-bit signal can have the same or substantially the same word width as the first signal 206.

[0070] In the Fig. 3C and Fig. The 3D illustrated embodiments are shown in the sensor arrangement 200. Fig. 3A also provides a decimation filter 300. The decimation filter is configured to provide a signal 302 with a reduced sampling rate, as described in more detail below. Alternatively or additionally to the decimation filter 300, a filter can be provided that increases the sampling rate, for example, an interpolation filter.

[0071] In various embodiments, this can be Fig. 3B illustrated the second filter 214, or its function could additionally be implemented in the first filter 208 or in the decimation filter 300.

[0072] At the in Fig. In the embodiment illustrated in Figure 3C, the decimation filter 300 is arranged with respect to the signal flow between the analog-to-digital converter 202 and the first filter 208. In the embodiment shown in Figure 3C, the decimation filter 300 is arranged between the analog-to-digital converter 202 and the first filter 208. Fig. In the 3D illustrated embodiment, the decimation filter 300 is arranged with respect to the signal flow between the first filter 208 and the interface 222. The sensor arrangement 200 with decimation filter 300 in various embodiments is no longer a linear, time-invariant system, as is the case in the embodiments of the Fig. 3A, B is the case. In this respect, the signal provided at interface 222 differs from that of the sensor arrangement 200. Fig. 3C from the signal provided at interface 222 Fig. 3D, if the filters used in the sensor arrangements have the same filter coefficients.

[0073] In these embodiments, the signal provided at interface 222 can be a multi-bit signal.

[0074] It clearly shows Fig. 3C is a sensor array with group delay optimization when the signal from the sensor array is to be provided at a lower sampling rate for specific applications.

[0075] The decimation filter 300 can be used for sample rate reduction. The decimation filter can perform both low-pass filtering and sample rate reduction.

[0076] The exemplary embodiment from Fig. 3D also shows that the first filter 208 can also be placed at the higher sampling rate of the first signal.

[0077] The decimation filter makes the modulator and / or the second filter optional; for example, they can be omitted. This is described in more detail below. Fig. 6B and Fig. Figure 7 shows a comparison of the group runtime, from which it can be seen that the application examples with multi-bit interface result in an additional improvement in the group runtime.

[0078] The analog-to-digital converter is, for example, a third-order sigma-delta analog-to-digital converter. The decimation filter could, for example, be a third-order comb filter.

[0079] Fig. 4A, Fig. Figure 4B further details a first filter 208 in the form of a finite impulse response filter, for example, a second-order finite impulse response filter. The first filter is part of a described sensor arrangement according to various embodiments. The finite impulse response filter operates in the discrete-time digital domain and provides an output signal at each processing step, which depends on the current input signal multiplied by a first scaling parameter (c0), for example, c0 = 1227. The input signal can be the signal received by the first filter 208.

[0080] The second signal 210 output by the first filter 208 further depends on the preceding input signal or sample value multiplied by an associated second scaling parameter c1, for example, c1 = -2411, and on the penultimate input signal multiplied by a third scaling parameter c2, for example, c2 = 1185. The second signal 210 is the sum of a scaled input sample value, a scaled preceding input sample value, and a scaled penultimate input sample value.

[0081] For example, the first filter is a second-order FIR filter with a corner frequency of approximately 20 kHz.

[0082] The in Fig. 4A illustrates a second filter that stabilizes the sensor arrangement and is, for example, a third-order wave digital filter, such as a Chebyshev filter, for example with a corner frequency of 70kHz.

[0083] In one embodiment, the second filter can be configured as an FIR filter.

[0084] Fig. Figure 4C shows in further detail a decimation filter 300 of a sensor arrangement 200 according to various embodiments, which essentially corresponds to one described embodiment.

[0085] The decimation filter can be set up as a single-stage or multi-stage filter.

[0086] Fig. Figure 4C, for example, shows a two-stage third-order decimation filter 300 with a first stage 404 that provides a decimated signal 402, and a second stage 406 that provides the signal 302 from the decimation filter 300. The second stage 406 can further reduce the sampling rate of the decimated signal 402. This allows, for example, arbitrary decimations, such as by a factor in the range of 2 to 48.

[0087] A decimation filter, or a stage of a decimation filter, can be implemented, for example, as a comb filter. The order of the decimation filter can correspond to the order of the analog-to-digital converter.

[0088] Alternatively or additionally to the decimation filter, an interpolation filter can be incorporated into the sensor arrangement to increase the sampling rate. In combination with a second filter in the form of a notch and / or pass filter, this allows the frequency-dependent amplitude response to be smoothed or attenuated within a predefined frequency range, or an application-specific sampling rate to be set.

[0089] In various embodiments, the sensor arrangement can have a single filter that fulfills the described function of the first filter, the second filter and / or the decimation filter (or an interpolation filter), for example, the one described in Fig. The amplitude and phase response described in sections 5A and B. In other words, the first filter and the second filter, i.e., their functionality, can be implemented in a single filter. This single filter can also be called the combined filter. In other words, the combined filter includes the first filter, the second filter, and / or the decimation / interpolation filter. Alternatively, the combined filter includes the function of the first filter, the function of the second filter, and / or the function of the decimation / interpolation filter.

[0090] The first and / or second filter can be zero-order, first-order, second-order, third-order, or even higher-order filters in various embodiments, provided that the filter fulfills the described function. For example, the first filter exhibits a phase response that increases positively with increasing frequency. The second filter, for example, is configured to reduce the amplitude of a high-frequency signal.

[0091] Fig. 5A and Fig. Figures 500 and 510 of 5B show the frequency response of a first filter from exemplary embodiments of a sensor arrangement described above. The first filter is implemented as a second-order FIR filter. The Fig. Examples 502, 504, 506, and 508 of the first filter, illustrated in Figures 5A and B, exhibit different damping values ​​of the oscillating system. The (normalized) damping can have a value between 0 and 1, where a value of 0 indicates a resonator and a value of 1 suppresses all oscillation. The first example, 502, has the highest damping among the four examples (502, 504, 506, and 508), and the fourth example, 508, has the lowest damping. The second example, 504, and the third example, 506, represent intermediate values.

[0092] The x-axis 110 in Fig. 5A, B shows the frequency in units of Hz. Fig. Figure 5A shows the y-axis 120, the group runtime 120 in units of seconds through the first filter. Fig. Figure 5B shows the y-axis 512, which represents the gain in dB. A negative gain corresponds to an attenuation of the signal amplitude of the signal received in the first filter as it passes through the first filter.

[0093] In Fig. 5A and Fig. Figure 5B shows the group delay 120 and the gain 512 of signals through first filters with different corner frequencies.

[0094] In the depicted frequency range from approximately 100 Hz to approximately 4 kHz, the group delay 120 and the gain 512 are essentially linearly dependent or essentially constant, i.e., independent of the frequency 110 of the signal received by the first filter. Thus, there is essentially a 1:1 mapping of the signal received by the first filter in this frequency range.

[0095] The first filter, which exhibits a high but less than zero group delay in this frequency range, has a lower group delay and lower gain (or higher attenuation) in the frequency range between approximately 10 kHz and 40 kHz than a first filter with a lower group delay in the frequency range from approximately 100 Hz to 3 kHz. The group delay of the first filter depends on its cutoff frequency. The higher the cutoff frequency, the lower the group delay in the frequency range from approximately 100 Hz to 3 kHz, but the greater the attenuation (see Fig. 5B) of the signal in the frequency range between approximately 10 kHz and approximately 40 kHz.

[0096] In the frequency range between approximately 10 kHz and 30 kHz, a resonant frequency of a sensor in the sensor array may be located, to which the first filter is connected. This means that the signal provided by the first filter is based on the signal provided by the sensor. Therefore, the first filter may be suitable for damping the oscillation of the sensor in the sensor array. The strong resonance peak of the first filter at approximately 19 kHz leads to a reduction in the signal-to-noise ratio of the sensor signal, which can be an undesirable characteristic.

[0097] The in Fig. As illustrated in Figure 5B, positive signal amplification in high-frequency ranges, for example above 40 kHz, can lead to a significant increase in the amplitude of otherwise essentially negligible frequency ranges. For example, the amplitude of the sensor's analog signal in this range is small and essentially negligible. Amplifying the signal in these frequency ranges with the first filter can overload the modulator. The second filter, for example a low-pass or band-pass filter with a cutoff frequency of, say, 40 kHz or higher, can filter out the high-frequency components, thus preventing overload of the modulator.

[0098] In other words, in the passband region of the second filter, the amplitude response, i.e., the gain of the first filter, is flat, and higher frequencies are amplified by the first filter. The second filter thus stabilizes the circuit and prevents overloading of the modulator.

[0099] The second filter is associated with a positive group delay. To achieve an optimized group delay for the overall system, the first and second filters are therefore matched. The second filter has the lowest possible cutoff frequency, for example, close to the band limit of the second filter. The first filter has the highest possible cutoff frequency. This ensures the stability of the overall system and also minimizes the (positive) group delay of the entire system. This results in a sensor array with optimized group delay.

[0100] If a multi-bit output signal (also referred to as m-bit or multi-bit) is permitted or desired for a specific application, the second filter is optional or can have a very high pass frequency. The word width, i.e., the number of bits, of the signal provided by the sensor array can be set to a predefined value using a modulator or decimation / interpolation filter as described above. The predefined value can be in a range from 1 bit to, for example, 1024 bits, from 1 bit to, for example, 128 bits, to, for example, 64 bits, to, for example, 32 bits, to, for example, 16 bits, to, for example, 8 bits, to, for example, 4 bits, or to, for example, 2 bits. However, it is not necessary for each of these bits to be assigned amplitude information.For example, higher bit counts can be assigned to a high-frequency signal component, which is then blocked by the second filter. However, the higher bit count may be necessary for further signal processing in specific applications. This can eliminate the need for a subsequent modulator or decimation / interpolation filter, or make such a filter optional.

[0101] Fig. Figures 6A and 6B show diagrams 600 and 610 for the group delay of a sensor array. The frequency response of a MEMS structure 604 and a conventional microphone array 602 is shown according to [reference missing]. Fig. Figure 9 shows an embodiment 606 according to one of the aforementioned embodiments with a low-pass filter and modulator, and an embodiment 608 according to one of the aforementioned embodiments without a low-pass filter, decimation filter, or modulator, i.e., with only a first filter. The x-axis 110 shows the frequency in units of Hz, and the y-axis 120 shows the group delay 120 in units of seconds through the respective sensor arrangement.

[0102] Out of Fig. Figures 6A and B show that the sensor arrangement, according to various embodiments, has a lower group delay than a conventional sensor arrangement and a lower group delay than a MEMS used in these sensor arrangements. The lower group delay allows for greater attenuation of interference signals, as shown in Fig. This is described in section 1. This allows for a sensor arrangement with an improved signal-to-noise ratio. Without a second filter, i.e., in example 608, the group delay can be further reduced compared to example 606 with a second filter, since the increase in the group delay of the signal caused by the second filter is eliminated.

[0103] Fig. Figure 7 shows a diagram 700 for the group delay of a sensor arrangement. The diagram depicts the frequency response of a MEMS structure 604 of embodiments with decimation filters 702, 704, 708, and 710 with different decimation levels (710: factor 4, 708: factor 8, 704: factor 16, 702: factor 24) according to one of the aforementioned embodiments with decimation filters. The decimation filter can be single-stage or multi-stage. In a multi-stage decimation filter, the individual stages can have the same or different decimation factors. The decimation factor indicates the degree by which the sampling rate is reduced.

[0104] The x-axis 110 shows the frequency in units of Hz and the y-axis 120 shows the group delay 120 in units of seconds through the respective sensor arrangement.

[0105] Out of Fig. Figure 7 shows that with an increasing factor, i.e., with a decreasing sampling rate, the group delay in the illustrated signal range, for example for frequencies greater than 1 kHz, is significantly larger. Accordingly, an optimized group delay can be determined for specific applications by means of a compromise between group delay and sampling rate.

[0106] Fig. Figure 8 shows a diagram of a signal processing method according to various embodiments. Method 800 comprises providing S1 an analog signal, converting S2 the analog signal into a digital first signal, and processing S3 the first signal into a second signal. The first signal has a first group delay, and the second signal has a second group delay, the second group delay being smaller than the first group delay. The first signal has a first word width, and the second signal has a second word width, the second word width being smaller than the first word width.

[0107] Signal processing can be performed, for example, using a circuit arrangement as described above. This circuit arrangement can, for example, be configured as a pressure sensor arrangement or include one. Alternatively or additionally, the circuit arrangement can be configured as a microphone arrangement or include one.

[0108] The analog signal can be generated, for example, by deflecting a membrane of a sensor in the sensor assembly. The membrane is, for example, a microelectromechanical structure or has one.

[0109] The processing S3 of the first signal into a second signal can be done by passing the signal through a first filter, which is, for example, an FIR filter or has, for example, a second-order FIR filter.

[0110] The first filter can have a corner frequency in a range of approximately 16 kHz to approximately 30 kHz, for example, in a range of approximately 16 kHz to approximately 22 kHz. The first filter can be configured such that the first signal has a group delay minimum in a frequency range, with the circuit arrangement having a resonant frequency in this frequency range. For example, the first filter can be configured such that the first signal has a gain of less than or equal to 0 dB in a frequency range of approximately 20 Hz to approximately 4 kHz. Alternatively or additionally, the processing S3 of the first signal into a second signal can be achieved by passing the signal through a second filter, which is, for example, a low-pass filter. The second filter can have a corner frequency greater than or equal to 20 kHz.

[0111] Example 1, which refers to Fig. 1 to Fig. The circuit described in Figure 8 includes a sensor. The sensor is configured to provide an analog signal. The circuit further includes an analog-to-digital converter (ADC) configured to receive the analog signal and provide a first signal. The circuit further includes a first filter configured to provide a second signal. The circuit further includes a second filter configured to provide a third signal. The circuit further includes a modulator configured to provide a fourth signal. The first filter, the second filter, and the modulator are connected to the ADC so that the first signal is processed, converted, or transformed into the fourth signal.The signal received by the first filter has a first group delay, and the first filter is configured to process the received signal such that the second signal output by the first filter has a second group delay, where the second group delay is smaller than the first group delay. The second filter is configured as a frequency-selective filter to process the received signal such that the amplitude of the received signal is reduced within a specified frequency range. The specified frequency range has a frequency greater than 20 kHz.

[0112] This allows the group delay of a signal in the sensor array to be reduced. As a result, interfering environmental signals can be more effectively attenuated without any loss of information. This improves the signal-to-noise ratio of the sensor array and thus its sensitivity. Furthermore, it prevents modulator overload.

[0113] In Example 2, the subject of Example 1 may further include the first filter being set up to receive the first signal, the second filter being set up to receive the second signal, and the modulator being set up to receive the third signal.

[0114] In Example 3, the subject of Example 1 or 2 may further include the circuit arrangement being designed as, or having, a pressure sensor arrangement.

[0115] In Example 4, the subject of one of Examples 1 to 3 may further exhibit that the circuit arrangement is designed as a microphone arrangement or includes one.

[0116] In Example 5, the object of one of Examples 1 to 4 may further include the sensor having a membrane, wherein a deflection of the membrane from a rest position generates the analog signal.

[0117] In Example 6, the subject of Example 5 may further exhibit that the membrane is or has a microelectromechanical structure.

[0118] In Example 7, the object of one of Examples 1 to 6 may further exhibit that the first filter is or has an FIR filter.

[0119] In Example 8, the object of one of Examples 1 to 7 may further exhibit that the first filter is or has a second-order FIR filter.

[0120] In Example 9, the subject of one of Examples 1 to 8 may further exhibit that the first filter has a corner frequency in a range of approximately 16 kHz to approximately 22 kHz.

[0121] In Example 10, the subject of one of Examples 1 to 9 may further include that the first filter is set up, that the first signal has a group delay minimum in a frequency range, and that the circuit arrangement has a resonant frequency in that frequency range.

[0122] In Example 11, the object of one of Examples 1 to 10 may further include the first filter being set up so that the first signal has a gain of less than or equal to 0 dB in a frequency range of approximately 20 Hz to approximately 20 kHz.

[0123] In Example 12, the subject of one of Examples 1 to 11 may further exhibit that the second filter is or has a low-pass filter.

[0124] In Example 13, the subject of one of Examples 1 to 12 may further exhibit that the second filter has a corner frequency in a range of approximately 40 kHz to approximately 80 kHz.

[0125] In Example 14, the subject of one of Examples 1 to 13 may further include the fourth signal having a third group delay and the analog signal having a fourth group delay, wherein the third group delay is smaller than the fourth group delay in a frequency range of approximately 20 Hz to approximately 20 kHz.

[0126] In Example 15, the subject of one of Examples 1 to 13 may further comprise a composite filter, wherein the composite filter comprises the first filter and the second filter, or wherein the composite filter has the function of the first filter and the function of the second filter.

[0127] In example 16, which refers to Fig. 1 to Fig. As described in section 8, a signal processing method comprises: providing an analog signal; converting the analog signal into a digital first signal; processing the first signal into a second signal, wherein the first signal has a first group delay and the second signal has a second group delay, the second group delay being less than the first group delay; and wherein the first signal has a first word width, and the second signal has a second word width, the second word width being less than the first word width.

[0128] Example 17, which refers to Fig. 1 to Fig. The circuit arrangement described in Figure 8 includes a sensor. The sensor is configured to provide an analog signal. The sensor arrangement further includes an analog-to-digital converter configured to receive the analog signal and provide a first signal. The sensor arrangement further includes a first filter configured to receive a signal based on the first signal and provide a second signal. The first filter is configured such that the second signal is passed unamplified or substantially unamplified within a frequency range of approximately 20 Hz to approximately 10 kHz. The filter is further configured such that the second signal has a gain greater than 0 dB at least from a predetermined frequency greater than approximately 20 kHz.

[0129] This reduces the group delay of the first signal.

[0130] An unamplified or essentially unamplified pass of the signal received by the first filter can be understood as a 1:1 mapping of the received signal to the provided (second) signal in this frequency range, except for an amplification attributable to a current ripple. An unamplified or essentially unamplified pass can, for example, exhibit a gain in the range of -3 dB to +3 dB in this frequency range.

[0131] For example, the first filter may be set to amplify the recorded signal from a frequency of approximately 20 kHz upwards, i.e., exhibit an amplitude gain greater than 0 dB. However, the gain may also be greater than 0 dB from frequencies above 10 kHz upwards; see, for example, the shape of signal curve 502 in [reference missing]. Fig. 5B. In other words, the gain is not necessarily greater than 0 dB in magnitude only from a frequency of approximately 20 kHz.

[0132] The gain above a frequency of approximately 20 kHz can exhibit a linear or essentially linear response. A linear or essentially linear dependence of the gain on frequency refers to a representation of the signal gain in units of decibels (dB) on a logarithmic scale with respect to frequency; see, for example, signal gains in [reference to relevant source]. Fig. 5B for frequencies above 20 kHz. A linear relationship means that the frequency-dependent gain response of the signal can be approximated by a straight line. The signal gain at a given frequency may deviate slightly from the value of the straight line at that frequency. For example, the signal gain might be in a range of approximately ±10% or ±5% of the gain value of the straight line. With such a deviation, a substantially linear relationship can still be assumed.

[0133] In the event that the signal received by a filter is based on another, provided signal, it is to be understood that the received signal is identical to the provided signal or that the provided signal is first processed in another way, for example by another filter, before it is received by the filter.

[0134] In Example 18, the sensor arrangement of Example 17 further includes a second filter connected to the analog-to-digital converter and configured to provide a third signal based on the first signal.

[0135] In Example 19, the sensor arrangement of Example 18 further features the second filter being configured as a frequency-selective filter to process the signal received by the second filter in such a way as to reduce the amplitude of the signal received by the second filter in a specified frequency range, wherein the specified frequency range has a frequency greater than 20 kHz.

[0136] In Example 20, the sensor arrangement of Example 17 to 19 further includes a modulator connected to the analog-to-digital converter and configured to provide a fourth signal based on the first signal.

[0137] In Example 21, the sensor arrangement of Examples 17 to 20 further includes a decimation filter connected to the analog-to-digital converter and configured to provide a fifth signal based on the first signal.

[0138] In Example 22, the sensor arrangement of Example 17 to 21 further includes an interface for providing a sixth signal based on the second signal, wherein the interface for providing the sixth signal is set up to a circuit arrangement-external environment.

[0139] In Example 23, the sensor arrangement of Example 22 further shows that the sixth signal is a multi-bit signal.

[0140] In Example 24, the sensor arrangement of Examples 17 to 23 further shows that the circuit arrangement is designed as, or includes, a pressure sensor arrangement.

[0141] In Example 25, the sensor arrangement of Examples 17 to 24 further shows that the circuit arrangement is designed as, or includes, a microphone arrangement.

[0142] In Example 26, the sensor arrangement of Examples 17 to 25 further shows that the sensor has a membrane, wherein a deflection of the membrane from a rest position generates the analog signal.

[0143] In Example 27, the sensor arrangement of Example 26 further shows that the membrane is or has a microelectromechanical structure.

[0144] In Example 28, the sensor arrangement of Examples 17 to 27 further shows that the first filter is or has an FIR filter.

[0145] In Example 29, the sensor arrangement of Examples 17 to 28 further shows that the first filter is or has a second-order FIR filter.

[0146] In Example 30, the sensor arrangement of Example 17 to 29 further shows that the first filter has a corner frequency in a range of approximately 16 kHz to approximately 22 kHz.

[0147] In Example 31, the sensor arrangement of Example 17 to 30 further shows that the first filter is set up, that the first signal has a group delay minimum in a frequency range, wherein the circuit arrangement has a resonant frequency in that frequency range.

[0148] In Example 32, the sensor arrangement of Examples 18 to 31 further shows that the second filter is or has a low-pass filter.

[0149] In Example 33, the sensor arrangement of Example 18 to 32 further shows that the second filter has a corner frequency in a range of approximately 40 kHz to approximately 80 kHz.

[0150] In Example 34, the sensor arrangement of Example 17 to 33 further exhibits that the signal provided by the sensor arrangement has a first group delay and the analog signal has a second group delay, wherein the first group delay is smaller than the second group delay in a frequency range of approximately 20 Hz to approximately 20 kHz.

[0151] In Example 35, the sensor arrangement of Example 18 to 34 further comprises a total filter, wherein the total filter comprises the first filter and the second filter, or wherein the total filter has the function of the first filter and the function of the second filter.

[0152] In example 36, which refers to Fig. 1 to Fig.Section 8 describes a signal processing method. This method involves providing an analog signal, converting the analog signal into a digital first signal, and processing the first signal into a second signal. The first signal has a first group delay, and the second signal has a second group delay, the second group delay being smaller than the first group delay.

[0153] The procedure may require or be carried out with a sensor arrangement as described above.

[0154] In Example 37, the procedure of Example 36 further shows that the first signal has a first word width and the second signal has a second word width, the second word width being smaller than the first word width.

[0155] It should be noted that the functional blocks described here should not be interpreted as meaning that the corresponding functionality is necessarily implemented in a single hardware component or device. Instead, the various functionalities may be distributed across different devices or implemented in a single device. For example, in some examples, the source follower, amplifier, analog-to-digital converter, tone control device, and sensor array modulator may be implemented as a single ASIC or device, while in other examples, they may be implemented using two or more separate devices.

Claims

[1] Circuit arrangement (200), comprising: a sensor, set up to provide an analog signal (204); an analog-to-digital converter (202), set up to receive the analog signal (204) and provide an initial signal (206); a first filter (208) configured to receive a signal based on the first signal (206) and to provide a second signal (210); where the first filter (208) is set up such that: • that the second signal (210) is passed through unamplified in a frequency range from 20 Hz to 10 kHz, and • that the second signal (210) has a gain greater than 0 dB at least from a specified frequency greater than 20 kHz; and a second filter (214) connected to the analog-to-digital converter (202) and configured to provide a third signal (216) based on the first signal, wherein the second filter (214) is configured as a frequency-selective filter to process the signal received by the second filter, that the amplitude of the signal received by the second filter (214) is reduced in a specified frequency range, wherein the specified frequency range has a frequency greater than 20 kHz. [2] Circuit arrangement (200) according to claim 1, further comprising: a modulator (218) connected to the analog-to-digital converter (202) and configured to provide a fourth signal (220) based on the first signal. [3] Circuit arrangement (200) according to one of claims 1 or 2, further comprising: a decimation filter connected to the analog-to-digital converter (202) and configured to provide a fifth signal based on the first signal. [4] Circuit arrangement (200) according to one of claims 1 to 3, further comprising an interface for providing a sixth signal based on the second signal, wherein the interface for providing the sixth signal is set up to an environment external to the circuit arrangement. [5] Circuit arrangement (200) according to claim 4, wherein the sixth signal is a multi-bit signal. [6] Circuit arrangement (200) according to one of claims 1 to 5, wherein the circuit arrangement is designed or comprises a pressure sensor arrangement. [7] Circuit arrangement (200) according to any one of claims 1 to 6, wherein the circuit arrangement is designed or comprises a microphone arrangement. [8] Circuit arrangement (200) according to one of claims 1 to 7, wherein the sensor has a membrane, wherein a deflection of the membrane from a rest position generates the analog signal (204). [9] Circuit arrangement (200) according to claim 8, wherein the membrane is or has a microelectromechanical structure. [10] Circuit arrangement (200) according to any one of claims 1 to 9, wherein the first filter (208) is or comprises an FIR filter. [11] Circuit arrangement (200) according to any one of claims 1 to 10, wherein the first filter (208) is or comprises a second-order FIR filter. [12] Circuit arrangement (200) according to one of claims 1 to 11, wherein the first filter (208) has a corner frequency in a range of 16 kHz to 22 kHz. [13] Circuit arrangement (200) according to any one of claims 1 to 12, wherein the first filter (208) is configured such that the first signal (206) has a group delay minimum in a frequency range, wherein the circuit arrangement has a resonant frequency in this frequency range. [14] Circuit arrangement (200) according to one of claims 2 to 13, wherein the second filter (214) is or comprises a low-pass filter. [15] Circuit arrangement (200) according to one of claims 2 to 14, wherein the second filter (214) has a corner frequency in a range of 40 kHz to 80 kHz. [16] Circuit arrangement (200) according to any one of claims 1 to 15, wherein the signal (220) provided by the sensor arrangement has a first group delay and the analog signal (204) has a second group delay, wherein the first group delay is smaller than the second group delay in a frequency range of 20 Hz to 20 kHz. [17] Circuit arrangement (200) according to one of claims 1 to 16, comprising an overall filter, wherein the overall filter comprises the first filter (208) and the second filter (214), or where the overall filter has the function of the first filter (208) and the function of the second filter (214). [18] Method (800) for signal processing, comprising: providing (S1) an analog signal; converting (S2) the analog signal into a digital first signal (206); Processing (S3) the first signal (206) into a second signal (220), wherein the first signal (206) has a first group delay and the second signal (220) has a second group delay, the second group delay being smaller than the first group delay; and Reducing the amplitude of the second signal within a specified frequency range, where the specified frequency range has a frequency greater than 20 kHz. [19] Signal processing method according to claim 18, wherein the first signal (206) has a first word width and the second signal (220) has a second word width, wherein the second word width is less than the first word width.

Images