High-dynamic capacitive digital microphone system based on electrostatic force feedback

Abstract

Disclosed in the present invention is a high-dynamic capacitive digital microphone system based on electrostatic force feedback. The system comprises an MEMS microphone and an interface circuit chip thereof, wherein the interface circuit chip comprises a charge pump, an analog front end, an analog-to-digital conversion circuit, an amplitude detection module, an amplitude adjustment control module and a pulse density modulation module. In the present invention, the characteristic of the sensitivity being able to be adjusted by means of changing a bias voltage by a capacitive sensor is used, and a charge pump with an adjustable output voltage is used to provide electrostatic force feedback, thereby further expanding the adjustable range of a front-end gain, and also reducing requirements for the noise performance of a readout circuit. Therefore, the system in the present application has the characteristics of high dynamic range and high-precision quantization.

Description

High dynamic capacitive digital microphone system based on electrostatic force feedback Technical Field

[0001] The present invention belongs to the technical field of CMOS integrated circuits, and in particular relates to a high-dynamic capacitance digital microphone system based on electrostatic force feedback. Background Art

[0002] A digital microphone and its interface circuit chip are electronic devices that process analog audio signals input from the outside world through filtering, amplification, and other processing techniques, converting them into digital signals. This digital microphone interface circuit converts analog signals into digital signals, making them immune to interference and influence from other external signal sources. Furthermore, digital signals are easier to process efficiently in computer systems than analog signals, leading to their widespread application in fields such as speech communication, medical diagnosis, and environmental monitoring. A digital microphone and its interface circuit chip typically consists of a micro-electro-mechanical system (MEMS) for converting audio signals into electrical signals, an analog front end (AFE) for analog signal processing, and an analog-to-digital converter (ADC) for converting analog signals into digital signals.

[0003] There are some inherent problems with how audio signals are captured and digitized:

[0004] ① Quantization error: When the change of the analog signal is smaller than the quantization step of the ADC, quantization error will occur.

[0005] ②Signal saturation: When the input analog signal is too large, it will cause information loss after being amplified by the analog front end; when the input analog signal is too small, it will be submerged in the noise and cannot be measured.

[0006] ③Noise amplification: Noise will be amplified along with the signal.

[0007] By improving ADC accuracy and reducing noise in the analog front end, the effects of quantization error and noise amplification can be significantly reduced. The problem of signal saturation is closely related to the dynamic range of the digital microphone interface circuit. Dynamic range is the range between the system's noise floor and its maximum output voltage. The dynamic range of a digital microphone interface circuit is the range of signal amplitudes that the interface circuit can resolve. Audio signals vary widely. If the dynamic range is insufficient, large signals will experience clipping distortion, small signals will be difficult to discern, and the signal-to-noise ratio will be reduced. Therefore, a high dynamic range is crucial.

[0008] Capacitive digital microphones are increasingly used in battery-powered devices. Due to battery power limitations, the development trend of capacitive digital microphones is towards low voltage and low power consumption. However, the reduction in voltage severely limits the dynamic range of the signal chain. In order to achieve high dynamic range at low supply voltage, previous research literature [E.Bach et al., "9.5 A 1.8V true-differential 140dB SPL full-scale standard CMOS MEMS digital microphone exhibiting 67dB SNR," 2017 IEEE International Solid-State Circuits Conference (ISSCC), San Francisco, CA, USA, 2017, pp. 166-167] used a voltage multiplier module to increase the supply voltage of the analog front end, thereby expanding the dynamic range of the analog front end. The dual-backplate microphone was used to achieve a signal-to-noise ratio of 67dB and a noise reduction of 136dB. SPL The maximum acoustic overload point is 0.000, but the power efficiency of the voltage multiplier module is limited, resulting in high overall system power consumption. Another solution is to use an extremely low-noise and gain-adjustable analog front end to expand the dynamic range. (L.Sant et al., "A 130dB SPL 72dB SNR MEMS Microphone Using a Sealed-Dual Membrane Transducer and a Power-Scaling Read-Out ASIC," in IEEE Sensors Journal, vol. 22, no. 8, pp. 7825-7833, 15 April 2022) However, due to the strict trade-off between noise performance and power supply current, in order to suppress the front-end noise at a low level, the system current is limited, resulting in high overall power consumption. Summary of the Invention

[0009] In view of the above, the present invention provides a high dynamic capacitance digital microphone system based on electrostatic force feedback, which can improve the dynamic range of the system and has the characteristics of high dynamic range, high-precision quantization, and adjustable gain.

[0010] A high-dynamic capacitive digital microphone system based on electrostatic force feedback includes a MEMS microphone and an interface circuit chip thereof. The MEMS microphone is used to convert variable sound pressure input from an external source into an audio analog signal. The interface circuit chip includes:

[0011] A charge pump with adjustable output voltage to provide bias voltage for the MEMS microphone;

[0012] Low-noise gain-adjustable analog front end for filtering and amplifying audio analog signals for differential output;

[0013] The analog-to-digital conversion circuit is used to filter, sample, quantize and output digital codewords of the signals output by the analog front end;

[0014] The amplitude detection module is used to reproduce the high-speed low-bit digital codeword into a low-speed high-bit signal, obtain the difference between the maximum and minimum values ​​of the signal, and compare the difference with the set upper and lower thresholds;

[0015] An amplitude adjustment control module generates a voltage control codeword and a gain control codeword according to the comparison result output by the amplitude detection module, so as to adjust the output of the charge pump and the analog front end;

[0016] The pulse density modulation module is used to modulate the digital codeword to output a pulse density modulation wave that matches the system input amplitude;

[0017] The power module is used to provide operating voltage and current to each functional circuit module in the chip.

[0018] Furthermore, the sound pickup capacitor of the MEMS microphone is composed of a highly flexible film, a perforated rigid backplate and a cavity. The rigid backplate has dense perforations, which help sound waves to be transmitted to the film, reduce air damping and reduce noise; the film has ventilation holes, which quickly release pressure when the film vibrates, making it easier to vibrate; the cavity is formed by etching multiple different additional layers deposited on a silicon wafer to improve sensitivity; the sensitivity of a MEMS microphone refers to the microphone's sound-to-electricity conversion efficiency, which is directly related to the bias voltage provided to the MEMS microphone. The higher the bias voltage, the higher the microphone sensitivity.

[0019] Furthermore, the charge pump is used to provide electrostatic force feedback, and its feedback depth is adjusted by the voltage control code provided by the amplitude adjustment control module; the charge pump includes a boost circuit, a sequence reference voltage generation circuit and a logic control circuit, wherein the boost circuit adopts a two-stage Dickson charge pump structure, and the sequence reference voltage generation circuit selects the reference voltage through the voltage control code. The reference voltage enters the logic control circuit and is compared with the feedback voltage (a parameter related to the charge pump output voltage). The comparison result is used to control whether the boost circuit works or not, thereby realizing the function of adjustable output voltage.

[0020] Furthermore, the analog front end adjusts the degree of amplification of the audio analog signal through the gain control codeword provided by the amplitude adjustment control module. The analog front end has multiple gain gears. By adjusting the gain when the input signal size is different, the dynamic range of the overall signal chain can be expanded.

[0021] Furthermore, the analog front end is composed of three parts: a signal common-mode bias network, an amplifier circuit, and a feedback network. The signal common-mode bias network is located at the two input ends of the analog front end and is composed of two pseudo-resistors. One end of the pseudo-resistor is connected to the common-mode voltage, and the other end is connected to the input end of the analog front end. The input signal entering the bias network will be biased by the pseudo-resistor to achieve high-pass filtering, filtering out signals below the audio frequency acceptable to the human ear; the main body of the amplifier circuit is two operational amplifiers, the inverting input ends of the two operational amplifiers are connected through capacitors, and the non-inverting input ends of the two operational amplifiers correspond to the two input ends of the analog front end; the feedback network is composed of a pseudo-resistor, a capacitor array and its control switch. The capacitor array and its control switch are connected across the two ends of the operational amplifier in the form of negative feedback. The pseudo-resistor connects the feedback node and the output end of the operational amplifier to provide voltage for the feedback node. The gain is determined by the number of capacitors in the capacitor array connected to the feedback loop.

[0022] Furthermore, the analog-to-digital conversion circuit includes an anti-aliasing filter and an analog-to-digital converter, wherein the anti-aliasing filter is a low-pass filter, which is used to low-pass filter the signal output by the analog front end and suppress out-of-band signals to prevent noise from overlapping during sampling. After filtering, the signal is output to the analog-to-digital converter, which is used to sample, quantize and output digital codewords of the filtered signal.

[0023] Furthermore, the amplitude detection module includes a sampling filter for signal reproduction and a digital module for comparing signal amplitudes, wherein the sampling filter converts the high-speed, low-bit digital codeword samples output by the analog-to-digital conversion circuit into low-speed, high-bit signals. The sampling conversion process includes low-pass filtering and downsampling. The low-pass filtering uses a sinc filter to filter out out-band signals and prevent out-band signals from overlapping back into the band; downsampling is implemented using an integrator, an accumulator-dump extractor, and a differentiator. After the signal reproduction is completed through downsampling, the digital module uses the bandwidth lower limit as the detection frequency, obtains the maximum and minimum values ​​of the reproduced signal within the detection period, makes a difference, and compares the difference with the set upper and lower limit thresholds.

[0024] Furthermore, when the comparison result shows that the difference is greater than the upper threshold, the amplitude adjustment control module will output a gain control codeword to reduce the gain of the analog front end, and output a voltage control codeword to reduce the bias voltage output by the charge pump, thereby reducing the sensitivity of the MEMS microphone and expanding the range of detectable signals upward; when the comparison result shows that the difference is less than the lower threshold, the amplitude adjustment control module will output a gain control codeword to increase the gain of the analog front end, and output a voltage control codeword to increase the bias voltage output by the charge pump, thereby increasing the sensitivity of the MEMS microphone and expanding the range of detectable signals downward; when the comparison result shows that the difference is between the upper and lower thresholds, the amplitude adjustment control module will maintain the gain control codeword and voltage control codeword output in the previous cycle.

[0025] Furthermore, when the analog front-end gain has been adjusted to minimum and the comparison result still shows a difference greater than the upper threshold, the amplitude adjustment control module sends the output voltage control codeword to the charge pump for electrostatic force feedback adjustment, reducing the bias voltage provided by the charge pump to the MEMS microphone, thereby reducing the sensitivity of the MEMS microphone to detect larger sound signals. When the analog front-end gain has been adjusted to maximum and the comparison result still shows a difference less than the lower threshold, the amplitude adjustment control module sends the output voltage control codeword to the charge pump for electrostatic force feedback adjustment, increasing the bias voltage provided by the charge pump to the MEMS microphone, thereby increasing the sensitivity of the MEMS microphone to detect smaller sound signals. According to the above process, the analog front-end can be adjusted to a gain that matches the input. When a larger dynamic range is required, electrostatic force feedback is performed to adjust the charge pump to provide an appropriate output voltage, thereby expanding the dynamic range of the signal chain, ensuring a sufficient signal-to-noise ratio, and avoiding limiting distortion caused by excessive signal input.

[0026] Furthermore, the pulse density modulation module adopts a digital Delta Sigma modulator, which modulates the digital codeword output by the analog-to-digital conversion circuit according to the control codeword (voltage control codeword and gain control codeword), and generates a pulse density modulation wave that matches the system input amplitude to meet the format output required by the audio sensor.

[0027] The high-dynamic capacitive digital microphone system of the present invention provides a system-level solution for improving dynamic range and high-precision quantization. It utilizes the characteristic of capacitive sensors that can adjust sensitivity by changing the bias voltage, and adopts an output voltage-adjustable charge pump to provide electrostatic force feedback, further increasing the adjustable range of the front-end gain while relaxing the requirements for the noise performance of the readout circuit. At the same time, the present invention uses an amplitude detection module to perform signal reproduction and amplitude detection, dynamically monitoring the input signal amplitude. The amplitude adjustment control module outputs a gain control codeword, which is promptly fed back to the analog front end to adjust the gain of the analog front end. When the signal is small, the gain is increased to expand the dynamic range, and when the signal is large, the gain is reduced to avoid clipping distortion. When the gain is adjusted to the limit, the amplitude adjustment control module outputs the voltage control codeword to the charge pump. When the signal is small, the bias voltage provided to the MEMS microphone is increased to improve the sensitivity of the MEMS microphone and improve the signal-to-noise ratio of the signal input to the analog front end. When the signal is large, the bias voltage is reduced to avoid subsequent clipping distortion. Therefore, the system of the present invention has the characteristics of high dynamic range and high-precision quantization. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] FIG1 is a system structure block diagram of a high dynamic capacitance digital microphone system of the present invention, a conventional digital microphone and its interface circuit chip.

[0029] FIG2 is a schematic diagram of the circuit structure of the analog front end of the present invention.

[0030] FIG3 is a schematic diagram of the circuit structure of the charge pump of the present invention.

[0031] FIG4 is a schematic diagram showing the structure and timing of the analog-to-digital signal conversion circuit of the present invention.

[0032] FIG5 is a schematic diagram showing the signal determination principle of the amplitude detection module and the amplitude adjustment control module of the present invention.

[0033] FIG6 is a schematic diagram showing the composition and meaning of control code words in an embodiment of the present invention.

[0034] FIG7 is a diagram showing a simulated spectrum of a digital signal output by the digital microphone system of the present invention.

[0035] FIG8 is a schematic diagram showing changes in the signal-to-noise ratio and acoustic overload point of the system according to the present invention as the depth of electrostatic force feedback changes. DETAILED DESCRIPTION

[0036] In order to describe the present invention more specifically, the technical solution of the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] The high-dynamic capacitive digital microphone system based on electrostatic force feedback of the present invention includes a MEMS microphone and its interface circuit chip. The interface circuit chip includes: an output voltage adjustable charge pump, a low-noise gain adjustable analog front end, an analog-to-digital signal conversion circuit, an amplitude detection module, an amplitude adjustment control module, a pulse density modulation module and a power supply module. The MEMS microphone converts variable sound pressure input from an external source into a change in capacitance and outputs it to the analog front end. A charge pump with an adjustable output voltage provides bias voltage for the MEMS microphone. The analog front end filters, amplifies, and differentially outputs the analog signal as a voltage signal with drive capability. The analog signal output by the analog front end is input into the analog-to-digital signal conversion circuit, and then passes through an anti-aliasing filter into a high-precision analog-to-digital converter for sampling, quantization, and conversion into a digital signal. The amplitude detection module reproduces and detects the amplitude of the digital codeword output by the analog-to-digital signal conversion circuit to determine the current signal magnitude. The amplitude adjustment control module outputs a gain control codeword to the analog front end and a voltage control codeword to the charge pump. The analog front end adjusts the gain level according to the gain control codeword, and the charge pump adjusts the bias voltage according to the voltage control codeword. The digital signal output by the analog-to-digital signal conversion circuit and the control codeword output by the amplitude adjustment control module are input into the pulse density modulation module, which outputs a pulse density modulated wave that matches the system input amplitude. The power supply module provides voltage and current to the charge pump with adjustable output voltage, the analog front end, the analog-to-digital signal conversion circuit, the amplitude detection module, the amplitude adjustment control module, and the pulse density modulation module.

[0038] The bias voltage of the MEMS microphone is provided by a charge pump with adjustable output voltage. Increasing the bias voltage will increase the sensitivity of the microphone, thereby improving the signal-to-noise ratio of the microphone; decreasing the bias voltage will reduce the sensitivity of the microphone and prevent limiting distortion of large sound signals.

[0039] The charge pump with adjustable output voltage is used to provide bias voltage for the MEMS microphone. It includes a boost circuit, a sequence reference voltage generation circuit and a logic control circuit. The boost circuit is a two-stage Dickson charge pump structure. The first-stage Dickson charge pump supplies power to the second stage and can output a voltage of 6 to 13V per 1V. The sequence reference voltage generation circuit selects the reference voltage by adjusting the voltage control code output by the amplitude adjustment control module. Changing the reference voltage will change the charge pump output voltage, thereby achieving voltage adjustment. The logic control circuit controls the working state of the boost circuit by comparing the reference voltage and the feedback voltage.

[0040] The low-noise gain adjustable analog front end consists of three parts: signal common-mode bias network, amplifier circuit, and feedback network, among which:

[0041] The signal common-mode bias network is located at the two input terminals of the analog front end and consists of two pseudo resistors. One end of the pseudo resistor is connected to the common-mode voltage V cm =V dd / 2, and the other end is connected to the input end; the target audio signal frequency is between 20 and 20kHz. When the input signal enters the common-mode bias network, it will be biased by the pseudo resistor to achieve high-pass filtering, filtering out signals below the audio signal frequency acceptable to the human ear.

[0042] The main body of the amplifier circuit is two operational amplifiers, the inverting input terminals of the two operational amplifiers are connected through capacitors, and their non-inverting input terminals are the non-inverting input terminal and the inverting input terminal of the analog front end respectively.

[0043] The feedback network consists of a capacitor array, switches, and pseudo-resistors. The gain control codeword output by the amplitude control module controls the switch in series with the capacitor to determine whether the capacitor is connected to the feedback loop. The gain is determined by the size of the capacitor connected to the feedback loop:

[0044] It turns out that:

[0045] Where: R FB1 、R FB2 C1 is a pseudo resistor connected across the two ends of the amplifier (the output and the inverting input) and provides a voltage for the feedback node. C1 and C2 form a feedback capacitor array connected to the feedback loops of the two amplifiers. C is a capacitor connected between the inverting inputs of the two amplifiers. V outn 、V outp They are the inverting output and the non-inverting output of the analog front end, Vin 、V ip They are the inverting input and non-inverting input of the analog front end respectively.

[0046] It can be seen from the gain formula that the gain coefficient of the amplifier circuit of this structure can be controlled by adjusting the ratio of the two capacitors; the size of the capacitor located between the inverting input terminals of the two amplifiers is not easy to adjust and is related to the noise performance of the analog front-end. Therefore, the present invention realizes gain adjustment by designing the feedback capacitor into a capacitor array.

[0047] The analog-to-digital signal conversion circuit includes an anti-aliasing filter and a high-precision analog-to-digital converter, which is used to filter, sample, quantize and output digital codewords for the analog signal output by the analog front end.

[0048] The anti-aliasing filter is a low-pass filter that suppresses out-of-band signals and prevents noise from overlapping during sampling. The filtered signal is sent to a high-precision analog-to-digital converter.

[0049] The high-precision analog-to-digital converter includes a bootstrap switch, an integrator, a feedforward summing accumulator, a comparator, and a clock generation circuit.

[0050] The amplitude detection module is an undersampling amplitude detection module. Its work is divided into two stages: signal reproduction and amplitude detection.

[0051] The signal reproduction process is completed through the extraction filter: first, low-pass filtering is performed to filter out out-of-band signals, and then downsampling is performed; the low-pass filter uses a low-order sinc filter, which is composed of an integrator, an accumulator-dump decimator and a differentiator. The integrator suppresses shaped noise, and the accumulator-dump decimator is implemented through a counter. The output result is a downsampled signal. The differentiator is used to offset the low-frequency gain of the integrator to ensure that the amplitude of each frequency point is consistent with the input. The final result is stored in a register at regular clock cycle intervals.

[0052] After signal reproduction is completed, the amplitude detection phase begins: the lower limit of the bandwidth is used as the detection frequency (to ensure that at least one entire cycle of the signal can be obtained for low-frequency signals). The maximum and minimum values ​​of the reproduced signal are obtained within the detection period, and the difference is compared with the preset upper and lower thresholds.

[0053] The amplitude adjustment control module outputs a control codeword based on the judgment result of the amplitude detection module: when the difference is greater than the upper threshold, it means that the signal amplitude is too large, and the analog front-end gain is reduced by one level. If the analog front-end gain is already the minimum multiple at this time, the charge pump output voltage is reduced; when the difference is between the upper and lower thresholds, it means that the signal amplitude is moderate, and the current control codeword is maintained; when the difference is less than the lower threshold, it means that the signal amplitude is too small, and the analog front-end gain is increased by one level. If the analog front-end gain is already the maximum multiple at this time, the charge pump output voltage is increased.

[0054] According to the above process, the analog front end can be adjusted to a gain that matches the input signal, and the output voltage of the charge pump can be adjusted to a size that matches the input signal. While ensuring a sufficient signal-to-noise ratio, limiting distortion caused by excessive signal input is avoided.

[0055] The pulse density modulation module is used to modulate the digital codewords output by the analog-to-digital conversion circuit to meet the output format required by the audio sensor. Its input is the control codeword and the quantization result of the high-precision analog-to-digital converter, and its output is a pulse density modulation wave that matches the system input amplitude.

[0056] As shown in FIG1 , an embodiment of the high dynamic capacitance type digital microphone system of the present invention is shown. The system includes a MEMS microphone and its interface circuit chip. The interface circuit chip includes an output voltage adjustable charge pump, a low noise gain adjustable analog front end, an analog-to-digital signal conversion circuit, an amplitude detection module, an amplitude adjustment control module and a pulse density modulation module. The overall signal path of the system is as follows: the MEMS microphone powered by the charge pump will generate different electrical signals on a pair of differential capacitors as the sound pressure of the external sound signal changes. This pair of differential analog signals will be input to the digital microphone interface circuit. The power supply of the interface circuit comes from the chip power module (including a bandgap reference source and a low voltage difference linear regulator); the signal input After entering the interface circuit, it will be biased by a pseudo-resistor to achieve high-pass filtering, filtering out signals below the audio signal frequency acceptable to the human ear (signals less than 20Hz); the differential analog signal is amplified by the analog front end, and the output differential signal passes through the anti-aliasing filter and enters the analog-to-digital converter to be converted into a digital signal; on the one hand, the digital signal is downsampled and reproduced by the amplitude detection module, and the amplitude is detected. The amplitude adjustment control module outputs the gain control codeword to adjust the analog front end gain gear and the output voltage control codeword to adjust the charge pump output voltage. The control codeword participates in the pulse density modulation of the subsequent digital signal; on the other hand, the digital signal enters the pulse density modulation module, and outputs a pulse density modulation wave that matches the input amplitude of the interface circuit.

[0057] As shown in Figure 2, the low-noise gain adjustable analog front end has differential input and differential output, which includes two amplifiers OTA1 and OTA2. The inverting inputs of OTA1 and OTA2 are connected through a capacitor C, the non-inverting input of OTA1 is the non-inverting input of the analog front end, and the non-inverting input of OTA2 is the inverting input of the analog front end. The input of the analog front end is connected through a pseudo resistor R Bias1 、R Bias2 Provides common mode bias for coupled signals; capacitor array C 0dB 、C 6dB 、C 12dB 、C 18dB and its control switch S 0dB 、S 6dB 、S12dB In the form of negative feedback, one end is connected to the output of the amplifier and the other end is connected to the inverting input of the amplifier; the pseudo resistor R FB1 、R FB2 Connect the feedback node (inverting input) to the output terminal to provide voltage for the feedback node; the gain is determined by the number of capacitors in the capacitor array connected to the feedback loop. The low-noise gain adjustable analog front-end gain range is 1x, 2x, 4x, and 8x. The specific gain determination mode is as follows: Switch S 0dB 、S 6dB 、S 12dB Disconnect, capacitor C 18dB Connect to the feedback loop, the gain gear is 8 times; switch S 0dB 、S 6dB Disconnect, S 12dB Closed, capacitor C 12dB 、C 18dB Connect to the feedback loop, the gain gear is 4 times; switch S 0dB Disconnect, S 6dB 、S 12dB Closed, capacitor C 6dB 、C 12dB 、C 18dB Connect to the feedback loop, the gain gear is 2 times; switch S 0dB 、S 6dB 、S 12dB Closed, capacitor C 0dB 、C 6dB 、C 12dB 、C 18dB Connected to the feedback loop, the gain position is 1x; by adjusting the gain when the input signal size varies, the overall signal chain dynamic range can be extended by 18dB.

[0058] As shown in Figure 3, the output voltage adjustable charge pump includes a boost circuit, a sequence reference voltage generation circuit, and a logic control circuit. The boost circuit adopts a two-stage Dickson charge pump structure. The output of the first-stage Dickson charge pump supplies power to the second-stage Dickson charge pump. It can output a voltage of 6 to 13V per 1V, and can stably output a maximum bias voltage of 16V. The sequence reference voltage generation circuit consists of eight resistors, seven switches, and a decoder. The voltage control codeword output by the amplitude adjustment control module enters the sequence reference voltage generation circuit through the decoder to control the opening and closing of the seven switches, and the output reference voltage V REF ; The logic control circuit will reference voltage V REF With the feedback voltage V FB (1 / 16 of the charge pump output voltage or 1 / 8 of the output voltage) is compared. When V FB Greater than V REF When V FB Less than VREF When the boost circuit works, when it reaches steady state, V FB In V REF Vibrating nearby.

[0059] As shown in Figure 4, the analog-to-digital signal conversion circuit includes an anti-aliasing filter and a high-precision analog-to-digital converter. The high-precision analog-to-digital converter adopts a Delta Sigma ADC architecture and has seven sub-modules, including:

[0060] Operational amplifier, the core unit of the switched capacitor integrator;

[0061] The integrator with chopper is a first-stage integrator, which includes a chopper modulation circuit and an operational amplifier;

[0062] The switched capacitor integrator is a second to fifth stage integrator, with the operational amplifier as the main part, and the sampling capacitor and the switched capacitor connected across the input and output of the operational amplifier;

[0063] Gate voltage bootstrap switch, which is the input signal sampling switch;

[0064] The feedforward summing accumulator performs weighted summation of the outputs of the integrators at each level and outputs the result to the comparator;

[0065] The comparator is a single-bit comparator.

[0066] As shown in Figure 5, the working process of the amplitude detection module is as follows: the signal output by the analog-to-digital signal conversion circuit is converted from a single bit to a multi-bit signal through a second-order sinc filter, and the out-of-band noise shaping order is reduced; the filtered signal SIG IN By downsampling, the signal bit rate is reduced by four times, and the out-of-band noise is overlapped; then, the signal SIG SF After two 1-order downsampling, the final output is a 5-bit digital signal SIG DEC , and set the threshold (High V TH ,Low V TH ) and outputs the comparison results to the amplitude adjustment control module. Through low-order sinc filtering and progressive downsampling, the number of bits used for amplitude detection is minimized while ensuring sufficient signal-to-noise ratio and bandwidth for amplitude detection. The amplitude detection module uses a maximum of 6 bits.

[0067] The working process of the amplitude adjustment control module is as follows: When the output result of the amplitude detection module is that the input signal is higher than High V TH When the amplitude adjustment control module outputs a control code to reduce the analog front-end gain or lower the bias voltage provided by the charge pump to the MEMS; when the amplitude detection module outputs the result that the input signal is lower than Low V THWhen the amplitude detection module outputs the control code to increase the analog front-end gain or increase the bias voltage provided by the charge pump to the MEMS; when the amplitude detection module outputs the result that the input signal amplitude is between the high and low thresholds, the amplitude adjustment control module maintains the control code of the previous cycle and the circuit does not make any adjustments.

[0068] The workflow of analog front-end gain adjustment and charge pump electrostatic force feedback depth adjustment is as follows: the analog front-end gain gears are 1x, 2x, 4x, and 8x, with the initial setting being 8x; the charge pump output voltage is set to 1V per gear, with eight gears from 6V to 13V, and the initial setting being 13V. The control codeword has five digits, as shown in Figure 6. The first three digits represent electrostatic force adjustment, 000 for 6V, 001 for 7V, 010 for 8V, 011 for 9V, 100 for 10V, 101 for 11V, 110 for 12V, and 111 for 13V; the last two digits represent gain adjustment, 00 for 1x, 01 for 2x, 10 for 4x, and 11 for 8x. The initial value of the control codeword is 00000, corresponding to a charge pump output voltage of 6V and an analog front-end gain of 1x; when the difference result is higher than High V TH , indicating that the current signal amplitude is too large, exceeding the upper limit of the signal amplitude that the system can detect; when the difference result is lower than Low V TH , indicating that the current signal amplitude is too small, the amplitude adjustment control module adjusts the output control code to 00001, corresponding to the charge pump output voltage of 6V and the analog front-end gain of 2 times. If the difference result in the next cycle is still lower than Low V TH , then continue to adjust the control code to 00010, corresponding to the charge pump output voltage of 6V, and the analog front-end gain is 4 times; if the difference result is between the upper and lower thresholds, keep the control code at 00001; if the difference result is higher than the upper threshold, adjust the control code to 00000. When the analog front-end gain is 8 times, the difference result is still lower than the Low V TH At this time, the electrostatic force feedback adjustment will be started until the control code is adjusted to 11111, at which time the upper limit of the system sensitivity is reached.

[0069] Figure 7 shows the spectrum characteristics of this embodiment under normal operation. The sampling frequency is set to 2.5 MHz, the number of sampling points is 4096, the input signal frequency is 7324 Hz, and the input signal amplitude is 60 mV. The output digital signal is shown in Figure 6. Observing the spectrum and calculating performance indicators, the signal-to-noise ratio is approximately 84.5 dB-A, the effective number of bits is 13.8, and the input equivalent noise is -117.5 dBFS (assuming 2.4 Vpp as 0 dBFS) by subtracting the signal-to-noise ratio from the peak amplitude.

[0070] Figure 8 shows the variation trend of the system signal-to-noise ratio and acoustic overload point with the change of electrostatic force feedback depth in this embodiment. When the MEMS bias voltage is 8V, the system reaches the maximum acoustic overload point of 133dB.SPL Compared with fixed voltage bias, electrostatic force feedback increases the system's acoustic overload point by 11.4dB; when the MEMS bias voltage is 10.5V, the system reaches a maximum signal-to-noise ratio of 68.2dB-A. Compared with fixed voltage bias, electrostatic force feedback increases the system's signal-to-noise ratio by 6.2dB.

[0071] The above description of the embodiments is intended to facilitate understanding and application of the present invention by those skilled in the art. Those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without requiring creative effort. Therefore, the present invention is not limited to the above embodiments. Any improvements or modifications made by those skilled in the art based on the present disclosure should fall within the scope of protection of the present invention.

Claims

1. A high-dynamic capacitive digital microphone system based on electrostatic force feedback, comprising a MEMS microphone and an interface circuit chip thereof, wherein the MEMS microphone is used to convert variable sound pressure input from an external source into an audio analog signal, characterized in that: The interface circuit chip includes: A charge pump with adjustable output voltage to provide bias voltage for the MEMS microphone; Low-noise gain-adjustable analog front end for filtering and amplifying audio analog signals for differential output; The analog-to-digital conversion circuit is used to filter, sample, quantize and output digital codewords of the signals output by the analog front end; The amplitude detection module is used to reproduce the high-speed low-bit digital codeword into a low-speed high-bit signal, obtain the difference between the maximum and minimum values ​​of the signal, and compare the difference with the set upper and lower thresholds; An amplitude adjustment control module generates a voltage control codeword and a gain control codeword according to the comparison result output by the amplitude detection module, so as to adjust the output of the charge pump and the analog front end; The pulse density modulation module is used to modulate the digital codeword to output a pulse density modulation wave that matches the system input amplitude; The power module is used to provide operating voltage and current to each functional circuit module in the chip.

2. The high dynamic capacitance digital microphone system according to claim 1, wherein: The MEMS microphone's pickup capacitor consists of a highly flexible membrane, a perforated rigid backplate, and a cavity. The rigid backplate has dense perforations, which help transmit sound waves to the membrane, reduce air damping, and reduce noise. The membrane has ventilation holes, which quickly release pressure when the membrane vibrates, making it easier to vibrate. The cavity is formed by etching multiple different additional layers deposited on a silicon wafer to improve sensitivity. The sensitivity of a MEMS microphone refers to the microphone's sound-to-electricity conversion efficiency and is directly related to the bias voltage provided to the MEMS microphone. The higher the bias voltage, the higher the microphone sensitivity.

3. The high dynamic capacitance digital microphone system according to claim 1, wherein: The charge pump is used to provide electrostatic force feedback, and its feedback depth is adjusted by the voltage control codeword provided by the amplitude adjustment control module; the charge pump includes a boost circuit, a sequence reference voltage generation circuit and a logic control circuit, wherein the boost circuit adopts a two-stage Dickson charge pump structure, the sequence reference voltage generation circuit selects a reference voltage through the voltage control codeword, and the reference voltage enters the logic control circuit for comparison with the feedback voltage. The comparison result is used to control whether the boost circuit works or not, thereby realizing the function of adjustable output voltage.

4. The high dynamic capacitance digital microphone system according to claim 1, wherein: The analog front end adjusts the degree of amplification of the audio analog signal through the gain control codeword provided by the amplitude adjustment control module. The analog front end has multiple gain gears. By adjusting the gain when the input signal size is different, the dynamic range of the entire signal chain can be expanded.

5. The high dynamic capacitance digital microphone system according to claim 1, wherein: The analog front end is composed of three parts: a signal common-mode bias network, an amplifier circuit, and a feedback network. The signal common-mode bias network is located at the two input ends of the analog front end and is composed of two pseudo-resistors. One end of the pseudo-resistor is connected to the common-mode voltage and the other end is connected to the input end of the analog front end. The input signal entering the bias network will be biased by the pseudo-resistor to achieve high-pass filtering, filtering out signals below the audio frequency acceptable to the human ear; the main body of the amplifier circuit is two operational amplifiers, the inverting input ends of the two operational amplifiers are connected through capacitors, and the non-inverting input ends of the two operational amplifiers correspond to the two input ends of the analog front end; the feedback network is composed of a pseudo-resistor, a capacitor array and its control switch. The capacitor array and its control switch are connected across the two ends of the operational amplifier in the form of negative feedback. The pseudo-resistor connects the feedback node and the output end of the operational amplifier to provide voltage for the feedback node. The gain is determined by the number of capacitors in the capacitor array connected to the feedback loop.

6. The high dynamic capacitance digital microphone system according to claim 1, wherein: The analog-to-digital conversion circuit includes an anti-aliasing filter and an analog-to-digital converter, wherein the anti-aliasing filter is a low-pass filter, which is used to low-pass filter the signal output by the analog front end and suppress out-of-band signals to prevent noise from overlapping during sampling. After filtering, the signal is output to the analog-to-digital converter, which is used to sample, quantize and output digital codewords of the filtered signal.

7. The high dynamic capacitance digital microphone system according to claim 1, wherein: The amplitude detection module includes a sampling filter for signal reproduction and a digital module for comparing signal amplitudes. The sampling filter converts the high-speed, low-bit digital codeword samples output by the analog-to-digital conversion circuit into low-speed, high-bit signals. The sampling conversion process includes low-pass filtering and downsampling. The low-pass filtering uses a sinc filter to filter out out-of-band signals and prevent out-of-band signals from overlapping back into the band. Downsampling is achieved using an integrator, an accumulator-dump extractor, and a differentiator. After signal reproduction is completed through downsampling, the digital module uses the bandwidth lower limit as the detection frequency, obtains the maximum and minimum values ​​of the reproduced signal within the detection period, makes a difference, and compares the difference with the set upper and lower limit thresholds.

8. The high dynamic capacitance digital microphone system according to claim 1, wherein: When the comparison result shows that the difference is greater than the upper threshold, the amplitude adjustment control module will output the gain control codeword to reduce the gain of the analog front end, and the output voltage control codeword to reduce the bias voltage output by the charge pump, thereby reducing the sensitivity of the MEMS microphone and expanding the range of detectable signals. When the comparison result shows that the difference is less than the lower limit threshold, the amplitude adjustment control module will output the gain control codeword to increase the gain of the analog front end, and output the voltage control codeword to increase the bias voltage output by the charge pump, thereby improving the sensitivity of the MEMS microphone and downwardly expanding the detectable signal range; when the comparison result shows that the difference is between the upper and lower limit thresholds, the amplitude adjustment control module will maintain the gain control codeword and voltage control codeword output in the previous cycle.

9. The high dynamic capacitance digital microphone system according to claim 8, wherein: When the gain of the analog front end has been adjusted to the minimum, the comparison result is still that the difference is greater than the upper threshold. At this time, the amplitude adjustment control module will output the voltage control codeword to the charge pump for electrostatic force feedback adjustment, reducing the bias voltage provided by the charge pump to the MEMS microphone, thereby reducing the sensitivity of the MEMS microphone to recognize larger sound signals; when the gain of the analog front end has been adjusted to the maximum, the comparison result is still that the difference is less than the lower threshold. At this time, the amplitude adjustment control module will output the voltage control codeword to the charge pump for electrostatic force feedback adjustment, increasing the bias voltage provided by the charge pump to the MEMS microphone, thereby increasing the sensitivity of the MEMS microphone to recognize smaller sound signals.

10. The high dynamic capacitance digital microphone system according to claim 1, wherein: The pulse density modulation module adopts a digital Delta Sigma modulator, which modulates the digital codeword output by the analog-to-digital conversion circuit according to the control codeword, and generates a pulse density modulation wave that matches the system input amplitude to meet the format output required by the audio sensor.

Images