High-frequency sampling signal processing method based on high-density array pressure sensor

Abstract

The invention relates to the technical field of sensor signal processing, discloses a high-frequency sampling signal processing method based on a high-density array pressure sensor, and aims to solve the problems of insufficient accuracy, safety and real-time performance of a system under high-density nodes and high sampling frequency, and the method comprises the following steps: S1, carrying out row-column cross addressing scanning and activating a preset pressure threshold node; s2, the independent analog-to-digital converters are assembled in a partition mode, and crosstalk is eliminated through time interleaving sampling; s3, the signal is enhanced through a three-stage link of pre-amplification, noise suppression and filtering, and the noise is reduced; s4, the voltage-controlled temperature-compensated crystal oscillator is matched with the phase-locked loop and the digital delay line to realize clock synchronization; s5, encrypting transmission data through a national cryptographic SM4 algorithm; s6, three-level verification is carried out to guarantee data validity; and S7, performing moving average updating on the null drift baseline and performing compensation output. According to the invention, stable sampling of 10000Hz-50000Hz is realized, the signal-to-noise ratio is greater than or equal to 90dB, the bit error rate is less than or equal to 10 <-10 >, the effective data retention rate is greater than or equal to 99.5%, and the comprehensive performance of the system is improved.

Description

Technical Field

[0001] This invention relates to the field of sensor signal processing technology, and more specifically to a high-frequency sampling signal processing method based on a high-density array pressure sensor. Background Technology

[0002] High-density array pressure sensors are widely used in industrial monitoring, medical equipment, and human-computer interaction due to their ability to acquire spatially distributed pressure information. However, as application scenarios increasingly demand higher accuracy, speed, and data integrity in pressure sensing, existing high-density array pressure sensor acquisition systems are gradually revealing numerous technical bottlenecks, making it difficult to meet the comprehensive needs of practical applications.

[0003] Currently, mainstream high-density array pressure sensors have reached node scales of thousands or even tens of thousands. Under high-frequency sampling requirements, traditional serial sampling architectures struggle to handle the data acquisition tasks of massive numbers of nodes. Serial sampling requires sequentially acquiring signals node by node, resulting in excessively long overall sampling cycles and making real-time capture of dynamic pressure signals impossible. When multiple sampling channels operate simultaneously, the electrical signals of adjacent channels interfere with each other, distorting the sampled data. Noise interference is another factor limiting the sampling accuracy of sensor systems. The signals from high-density array pressure sensors are often weak, especially in scenarios monitoring minute pressure changes. Traditional noise suppression methods can only suppress noise within a specific frequency range and cannot achieve wide-band, low-amplitude noise elimination. In multi-channel parallel sampling systems, existing systems often use ordinary crystal oscillators to provide clock signals, which have insufficient frequency stability and lack effective real-time correction mechanisms. As sampling time progresses, the phase difference between the clocks of different channels gradually increases, leading to significant errors in subsequent data processing and analysis results. High-density array pressure sensors generate a large amount of sampled data during operation. Traditional data transmission methods often use unencrypted bus transmission, lacking effective encryption mechanisms, thus compromising data security. Meanwhile, in high-speed data transmission, factors such as signal attenuation and electromagnetic interference can easily lead to an increased transmission error rate, failing to meet the requirements for high reliability.

[0004] In summary, existing high-density array pressure sensor acquisition systems face problems such as signal crosstalk, noise interference, and insufficient clock synchronization accuracy under high-density node and high sampling frequency conditions, making it difficult to simultaneously meet the comprehensive requirements of accuracy, safety, and real-time performance. There is an urgent need to propose an innovative high-frequency sampling signal processing method to break through the existing technical bottlenecks and promote the further application and development of high-density array pressure sensors in various fields. Summary of the Invention

[0005] The present invention aims to provide a high-frequency sampling signal processing method based on a high-density array pressure sensor, in order to solve the problem that existing high-density array pressure sensor acquisition systems are unable to simultaneously meet the comprehensive requirements of accuracy, safety and real-time performance under the conditions of high-density nodes and high sampling frequency.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A high-frequency sampling signal processing method based on a high-density array pressure sensor includes: S1, uses row and column interleaving addressing to scan the array and select nodes within the preset pressure threshold for activation; S2, divide the active nodes into several blocks, each block is configured with an independent analog-to-digital converter, and the sampling of nodes in the block is started sequentially with a preset small time offset using a time-staggered sampling method; S3 performs noise suppression and signal enhancement. A three-stage processing link of "preamplifier-noise suppression-filtering" is built at the front end of the analog-to-digital conversion. The preamplifier adopts an instrumentation amplifier architecture. The equivalent input noise is reduced in a low-temperature environment by an interferometer, and an active bandpass filter is used for filtering. S4 uses a voltage-controlled temperature-compensated crystal oscillator to generate a reference clock. A phase-locked loop is used to keep the phase difference between the sampling clock of each group and the reference clock within a preset range. The phase error is detected periodically, and when the error exceeds the limit, it is corrected in real time through a digital delay line. S5, group the sampling data of each zone and encrypt the transmission in electronic codebook mode using the national cryptographic SM4 algorithm; S6, the receiving end first performs a first-level cyclic redundancy check, performs a second-level timestamp continuity check on the data that passes the first-level check, and marks abnormal data whose adjacent timestamp differences exceed a preset threshold. The data packets marked as abnormal perform a third-level spatial consistency check. S7 establishes a zero-drift baseline for nodes, performs zero-pressure sampling on all nodes, records the baseline value, updates the baseline value using a moving average algorithm, subtracts the updated baseline value from the real-time sampled value to complete drift compensation, and outputs the compensated data to the host computer through a high-speed serial interface.

[0007] The principles and advantages of this scheme are as follows: In practical applications, node selection and activation can reduce the amount of parallel data and lower the system processing load, while ensuring that sampling is only performed on effective pressure areas, thus improving sampling targeting; hierarchical parallel and time-interleaved sampling can eliminate crosstalk within blocks, and combined with independent analog-to-digital converters, it can achieve high-frequency stable sampling, avoiding the rate bottleneck of traditional sampling architectures; the three-level signal processing link can effectively suppress noise and enhance weak signals, ensuring a high signal-to-noise ratio and providing high-quality signals for subsequent data processing; adaptive clock synchronization can maintain the clock consistency of the entire array, avoiding sampling data errors caused by clock deviations. Time misalignment ensures the accuracy of data in the time dimension; encrypted transmission mechanism, combining national cryptographic algorithms with frequency hopping spread spectrum, enhances data transmission security and anti-interference capabilities, and reduces the bit error rate; multi-level validity verification filters out abnormal data at each level, significantly improving the retention rate of valid data and reducing the interference of invalid data on subsequent analysis; online drift compensation can correct zero drift of sensor nodes in real time, avoiding measurement errors caused by drift accumulation; and finally, high-speed output ensures data real-time performance. This comprehensive approach solves multiple problems of existing systems in terms of sampling efficiency, signal quality, data security, clock synchronization, data validity, and drift control.

[0008] Preferably, as an improvement, in S1, the array size is 256 rows × 256 columns, the total number of nodes is 65536, and the preset pressure threshold is 0.5kPa to 2kPa; S1 further includes: S11, the number of activated nodes is controlled to be between 5% and 20% of the total number of nodes; S12, set the scan period to 50μs~200μs.

[0009] Technical benefits: Facilitates high-density pressure sensing while accurately identifying nodes within the effective pressure range, reduces the amount of parallel data, and ensures that the entire array scan is completed before subsequent high-frequency sampling cycles.

[0010] Preferably, as an improvement, in step S2, the active node is divided into 8 to 32 blocks, with a preset micro-time offset of 125 ns to 1 μs; step S2 further includes: S21, set the analog-to-digital converter resolution to 16-bit and the sampling frequency to 10000Hz~50000Hz; S22 controls the sampling clock jitter to be less than 20psRMS.

[0011] Technical benefits: It facilitates balancing the data volume and processing efficiency of each block, eliminates signal crosstalk during sampling at each node within the block, and ensures high stability of the sampling time base.

[0012] Preferably, as an improvement, S3 includes: S31 digitally controls the gain of the preamplifier to 20dB~40dB with a step size of 2dB and an input impedance ≥100MΩ. S32, through feedback control, reduces the equivalent input noise to [value missing]. The signal amplitude is monitored in real time. When the signal amplitude exceeds 80% of the full scale, the preamplifier gain is automatically reduced by 2dB to 5dB. S33, filter passband 1Hz~25kHz, passband ripple ≤0.5dB, stopband attenuation ≥60dB@0.1Hz / 100kHz; S34 performs peak detection on the amplitude of the amplified and filtered signal. If the peak value exceeds 90% of the full scale of the analog-to-digital converter, a gain adjustment command is triggered to ensure that the signal-to-noise ratio is stable at ≥90dB.

[0013] Technical benefits: It facilitates the transmission of weak signals without attenuation, avoids signal saturation, suppresses low-frequency drift noise and high-frequency electromagnetic interference, stably maintains a high signal-to-noise ratio, and provides a high-quality signal source for subsequent data processing.

[0014] Preferably, as an improvement, in S4, the preset range of the phase difference is ±100ps, the period for detecting the phase error is every 1ms to 5ms, and the digital delay line takes effect in the next sampling period after real-time correction is completed.

[0015] Technical benefits: It facilitates ensuring high synchronization between the sampling clock of each zone and the reference clock, guarantees the time consistency of multi-channel sampling data, promptly detects and quickly corrects clock synchronization deviations, and ensures that the entire array maintains continuous clock synchronization during high-frequency sampling.

[0016] Preferably, as an improvement, S5 includes: S51, group the sampling data of each block into 128-bit groups; S52 uses the national cryptographic SM4 algorithm for electronic codebook encryption, with a key length of 128 bits. S53, the encrypted data is transmitted at a rate of 1Gbps to 5Gbps through the low-voltage differential signal bus; S54, superimposed pseudo-random frequency hopping spread spectrum on the transmission link, with a frequency hopping rate of 1MHz to 10MHz; S55, controls the transmission error rate to be less than 10. - ¹².

[0017] Technical benefits: It effectively prevents the theft or tampering of sampled data during transmission, meets the real-time transmission requirements of large amounts of data under high-frequency sampling, and improves the anti-interference capability and security of data transmission.

[0018] Preferably, as an improvement, in S6, the first-level cyclic redundancy check polynomial adopts CRC-32; the abnormal data for the second-level timestamp continuity check is data whose difference between adjacent timestamps exceeds the range of 995μs to 1005μs; the third-level spatial consistency check uses the adjacent node pressure gradient threshold of 0.2kPa / cm for judgment, and data exceeding the threshold are discarded.

[0019] Technical benefits: It facilitates a significant increase in the retention rate of effective data, ensuring that the data ultimately used for analysis is highly reliable and effective.

[0020] Preferably, as an improvement, in S7, the zero-pressure sampling period is 10s to 60s, the window length for updating the baseline value using the moving average algorithm is 100 to 1000 sampling points, and the output rate of the high-speed serial interface is 100MB / s to 500MB / s.

[0021] Technical benefits: Facilitates timely updates of node zero-drift baselines, improves baseline stability, and quickly transmits compensated and effective data to the host computer.

[0022] Preferably, as an improvement, in S7, the output data format is 16-bit raw pressure data + 32-bit timestamp + 8-bit block number, totaling 56 bits per sampling point.

[0023] Technical benefits: Improves standardization, facilitates data parsing and processing by the host computer, and reduces the complexity of data interaction. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of a high-frequency sampling signal processing method based on a high-density array pressure sensor according to the present invention. Detailed Implementation

[0025] The following detailed description illustrates the specific implementation method: The basic implementation examples are as follows: Figure 1 As shown: A high-frequency sampling signal processing method based on a high-density array pressure sensor includes: S1 employs a row-column interleaved addressing method to scan the array. The array size is 256 rows × 256 columns, with a total of 65,536 nodes. Nodes within a preset pressure threshold (0.5 kPa to 2 kPa) are selected for activation. Only nodes with pressure values ​​within this range are activated. The 256 row × 256 column array size enables high-density pressure sensing, covering a wider monitoring area; the preset pressure threshold of 0.5 kPa to 2 kPa accurately filters out nodes within the effective pressure range, preventing invalid nodes from consuming system resources. S1 also includes: S11 controls the number of active nodes to 5% to 20% of the total number of nodes, thereby significantly reducing the amount of parallel data and reducing the system's computation and transmission pressure.

[0026] S12, with a scan period of 50μs to 200μs, ensures that the full array scan is completed before the subsequent high-frequency sampling period, avoiding the impact of scanning time on sampling real-time performance and providing time guarantee for high-frequency sampling.

[0027] S2, the active nodes are divided into several groups. In this embodiment, the active nodes are preferably divided into 8 to 32 groups. Each group is equipped with an independent analog-to-digital converter. A time-interleaved sampling method is used to sequentially start sampling of nodes within the group with a preset small time offset, which is 125ns to 1μs. The division into 8 to 32 groups can be flexibly adapted according to the number of active nodes, balancing the data volume and processing efficiency of each group; the small time offset of 125ns to 1μs can effectively eliminate signal crosstalk during sampling of each node within the group, ensuring the independence of the sampling signal. S2 also includes: S21 sets the analog-to-digital converter resolution to 16-bit to achieve high-precision signal quantization, reduce quantization errors, and improve pressure measurement accuracy; the sampling frequency is set to 10000Hz~50000Hz to meet the requirements for capturing dynamic pressure signals and is suitable for scenarios with rapid pressure changes.

[0028] S22 controls the sampling clock jitter to be less than 20psRMS, ensuring high stability of the sampling time base, avoiding sampling time deviation caused by clock fluctuations, and further ensuring the accuracy of the sampling data.

[0029] S3 performs noise suppression and signal enhancement by establishing a three-stage processing link of "preamplifier-noise suppression-filtering" at the front end of the analog-to-digital conversion. The preamplifier adopts an instrumentation amplifier architecture, and an interferometer is used to reduce the equivalent input noise in a low-temperature environment. An active bandpass filter is used for filtering. S3 includes: The S31 digitally controls the gain of the preamplifier from 20dB to 40dB in 2dB steps, allowing for flexible adjustment based on signal strength. This ensures that weak signals are effectively amplified while strong signals remain unsaturated. With an input impedance ≥100MΩ, the high input impedance reduces signal attenuation during transmission, guaranteeing distortion-free transmission of weak pressure signals.

[0030] S32 introduces a noise suppression circuit, employing an interferometer to achieve noise suppression at a low temperature of 4.2K. Feedback control is used to reduce the equivalent input noise to [a lower level]. Simultaneously, the signal amplitude is monitored in real time. When the signal amplitude exceeds 80% of full scale, the preamplifier gain is automatically reduced by 2dB to 5dB. This reduces the equivalent input noise to [value missing]. The following approach, combined with a real-time gain adjustment mechanism, can significantly reduce noise interference with the signal and prevent signal saturation.

[0031] In this embodiment, the interferometer is a quantum interference device (SQUID). A quantum interference device (SQUID) is an ultra-high sensitivity magnetic or electrosensitive device based on the quantum tunneling effect (Josephson effect). In the high-frequency sampling signal processing flow of this document, it is mainly used to reduce the equivalent input noise, providing a low-noise basis for subsequent signal amplification, filtering and analog-to-digital conversion. It is the key hardware support for achieving the technical indicator of "signal-to-noise ratio ≥90dB".

[0032] The S33 filter has a passband of 1Hz to 25kHz, a passband ripple of ≤0.5dB, and a stopband attenuation of ≥60dB@0.1Hz / 100kHz. With a passband range of 1Hz to 25kHz, it can accurately preserve the pressure signal frequency band while suppressing low-frequency drift noise below 1Hz and high-frequency electromagnetic interference above 25kHz, further purifying the signal.

[0033] S34 performs peak detection on the amplified and filtered signal amplitude. If the peak value exceeds 90% of the full-scale range of the analog-to-digital converter (ADC), a gain adjustment command is triggered to ensure a stable signal-to-noise ratio (SNR) of ≥90dB. The linkage between peak detection and gain adjustment commands allows for real-time monitoring of the signal amplitude, ensuring that the ADC always operates within its optimal range and stably maintains a high SNR of ≥90dB, providing a high-quality signal source for subsequent data processing.

[0034] S4 uses a voltage-controlled temperature-compensated crystal oscillator to generate the reference clock, with a frequency stability of ±0.01ppm. A phase-locked loop (PLL) maintains the phase difference between the sampling clock of each zone and the reference clock within a preset range of ±100ps, ensuring high synchronization between the sampling clocks of each zone and the reference clock. This avoids timing misalignment of sampling data due to clock phase deviation and guarantees the time consistency of multi-channel sampling data. Phase error is detected periodically at intervals of 1ms to 5ms. When the error exceeds the limit, real-time correction is performed via a digital delay line to promptly detect clock synchronization deviations and prevent their accumulation. The digital delay line correction takes effect in the next sampling cycle after completion, quickly correcting clock deviations and ensuring continuous clock synchronization throughout the array during high-frequency sampling. This is suitable for long-term continuous high-frequency sampling scenarios, improving the long-term operational stability of the system.

[0035] S5 involves grouping the sampled data from each zone and encrypting it using the national cryptographic algorithm SM4 in electronic codebook mode for transmission. S5 includes: S51 groups the sampled data of each zone into 128-bit groups; the 128-bit data grouping method can adapt to the encryption block size of the national cryptographic SM4 algorithm and improve encryption efficiency.

[0036] S52 uses the national cryptographic SM4 algorithm for electronic codebook encryption with a key length of 128 bits. The 128-bit key of the national cryptographic SM4 algorithm complies with national cryptographic standards, has high encryption strength, and can effectively prevent the sampled data from being stolen or tampered with during transmission.

[0037] S53, after encryption, transmits data at a rate of 1Gbps to 5Gbps via a low-voltage differential signal bus; the high-speed transmission rate of 1Gbps to 5Gbps can meet the real-time transmission requirements of large amounts of data under high-frequency sampling and avoid data accumulation.

[0038] S54 superimposes pseudo-random frequency hopping spread spectrum on the transmission link, with a frequency hopping rate of 1MHz to 10MHz; the pseudo-random frequency hopping spread spectrum of 1MHz to 10MHz can reduce the probability of the transmitted signal being intercepted and improve the anti-interference capability and security of data transmission.

[0039] S55, controls the transmission error rate to be less than 10. - ¹², less than 10 - The low transmission error rate of ¹² ensures the integrity of the sampled data during transmission, reducing data loss or distortion caused by transmission errors.

[0040] S6. The receiving end first performs a first-level cyclic redundancy check. The first-level cyclic redundancy check polynomial adopts CRC-32. The first-level cyclic redundancy check of CRC-32 polynomial has a high detection probability (≥99.999%), which can quickly filter out data that has errors during transmission and ensure the integrity of data transmission.

[0041] A second-level timestamp continuity check is performed on the data that passes the first-level check. The timestamp accuracy is 1μs. If the difference between adjacent timestamps exceeds the range of 995μs to 1005μs, it is marked as abnormal. This is to accurately identify data with abnormal timestamps and avoid invalid data caused by time synchronization problems.

[0042] Data packets marked as abnormal undergo a three-level spatial consistency check. A pressure gradient threshold of 0.2 kPa / cm between adjacent nodes is used for judgment, and data exceeding the threshold is discarded. The pressure gradient threshold of 0.2 kPa / cm conforms to the physical pressure distribution law and can effectively remove abnormal data that does not conform to spatial distribution logic. Through the three-level check and progressive screening mechanism, the effective data retention rate is greatly improved (effective data retention rate ≥99.5% after three-level check), ensuring that the data finally used for analysis has high reliability and effectiveness.

[0043] S7. Establish node zero-drift baseline, perform zero-pressure sampling on all nodes, and record the baseline value. The zero-pressure sampling period is 10s to 60s, which facilitates timely updating of node zero-drift baseline and avoids measurement errors caused by long-term drift accumulation.

[0044] The baseline value is updated using a moving average algorithm with an update window length of 100 to 1000 sampling points. This facilitates the smoothing of baseline value fluctuations, improves baseline stability, and reduces the impact of transient interference on the baseline.

[0045] The drift compensation is completed by subtracting the updated baseline value from the real-time sampled value. The compensated data is then output to the host computer via a high-speed serial interface with an output rate of 100MB / s to 500MB / s. This high-speed output rate allows for the rapid transmission of the compensated data to the host computer, preventing data stagnation and meeting the needs of real-time data processing and analysis. This provides timely data support for subsequent applications (such as pressure distribution visualization and abnormal pressure alarms).

[0046] The output data format is 16-bit raw pressure data + 32-bit timestamp + 8-bit zone number, totaling 56 bits per sampling point. The 16-bit raw pressure data fully preserves the accuracy information of the pressure measurement, providing an accurate basis for pressure value calculation; the 32-bit timestamp accurately records the acquisition time of each sampling point, facilitating subsequent analysis of the temporal variation patterns of the pressure signal and is suitable for time-series tracing of dynamic pressure events; the 8-bit zone number quickly locates the zone to which the sampled data belongs, facilitating data traceability and fault diagnosis. When data anomalies occur, the corresponding sampling zone can be quickly identified, improving system maintenance efficiency; the fixed 56-bit data format has a high degree of standardization, facilitating upper-level computer parsing and processing, and reducing the complexity of data interaction.

[0047] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A high-frequency sampling signal processing method based on a high-density array pressure sensor, characterized in that, include: S1, uses row and column interleaving addressing to scan the array and select nodes within the preset pressure threshold for activation; S2, divide the active nodes into several blocks, each block is configured with an independent analog-to-digital converter, and the sampling of nodes in the block is started sequentially with a preset small time offset using a time-staggered sampling method; S3 performs noise suppression and signal enhancement. A three-stage processing link of "preamplifier-noise suppression-filtering" is built at the front end of the analog-to-digital conversion. The preamplifier adopts an instrumentation amplifier architecture. The equivalent input noise is reduced in a low-temperature environment by an interferometer, and an active bandpass filter is used for filtering. S4 uses a voltage-controlled temperature-compensated crystal oscillator to generate a reference clock. A phase-locked loop is used to keep the phase difference between the sampling clock of each group and the reference clock within a preset range. The phase error is detected periodically, and when the error exceeds the limit, it is corrected in real time through a digital delay line. S5, group the sampling data of each zone and encrypt the transmission in electronic codebook mode using the national cryptographic SM4 algorithm; S6, the receiving end first performs a first-level cyclic redundancy check, performs a second-level timestamp continuity check on the data that passes the first-level check, and marks abnormal data whose adjacent timestamp differences exceed a preset threshold. The data packets marked as abnormal perform a third-level spatial consistency check. S7 establishes a zero-drift baseline for nodes, performs zero-pressure sampling on all nodes, records the baseline value, updates the baseline value using a moving average algorithm, subtracts the updated baseline value from the real-time sampled value to complete drift compensation, and outputs the compensated data to the host computer through a high-speed serial interface.

2. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that, In S1, the array has a size of 256 rows × 256 columns, a total of 65,536 nodes, and a preset pressure threshold of 0.5 kPa to 2 kPa. S1 further includes: S11, the number of activated nodes is controlled to be between 5% and 20% of the total number of nodes; S12, set the scan period to 50μs~200μs.

3. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that, In step S2, the activated nodes are divided into 8 to 32 blocks, with a preset micro-time offset of 125 ns to 1 μs; step S2 also includes: S21, set the analog-to-digital converter resolution to 16-bit and the sampling frequency to 10000Hz~50000Hz; S22 controls the sampling clock jitter to be less than 20psRMS.

4. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that, S3 includes: S31 digitally controls the gain of the preamplifier to 20dB~40dB with a step size of 2dB and an input impedance ≥100MΩ. S32, through feedback control, reduces the equivalent input noise to [value missing]. The signal amplitude is monitored in real time. When the signal amplitude exceeds 80% of the full scale, the preamplifier gain is automatically reduced by 2dB to 5dB. S33, filter passband 1Hz~25kHz, passband ripple ≤0.5dB, stopband attenuation ≥60dB@0.1Hz / 100kHz; S34 performs peak detection on the amplitude of the amplified and filtered signal. If the peak value exceeds 90% of the full scale of the analog-to-digital converter, a gain adjustment command is triggered to ensure that the signal-to-noise ratio is stable at ≥90dB.

5. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that: In S4, the preset range of phase difference is ±100ps, the period for detecting phase error is every 1ms to 5ms, and the digital delay line takes effect in the next sampling period after real-time correction is completed.

6. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that, S5 includes: S51, group the sampling data of each block into 128-bit groups; S52 uses the national cryptographic SM4 algorithm for electronic codebook encryption, with a key length of 128 bits. S53, the encrypted data is transmitted at a rate of 1Gbps to 5Gbps through the low-voltage differential signal bus; S54, superimposed pseudo-random frequency hopping spread spectrum on the transmission link, with a frequency hopping rate of 1MHz to 10MHz; S55, controls the transmission error rate to be less than 10. - ¹².

7. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that: In S6, the first-level cyclic redundancy check polynomial adopts CRC-32; the abnormal data for the second-level timestamp continuity check are data whose difference between adjacent timestamps exceeds the range of 995μs to 1005μs; the third-level spatial consistency check uses the adjacent node pressure gradient threshold of 0.2kPa / cm for judgment, and data exceeding the threshold are discarded.

8. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that: In S7, the zero-pressure sampling period is 10s to 60s, the window length for updating the baseline value using the moving average algorithm is 100 to 1000 sampling points, and the output rate of the high-speed serial interface is 100MB / s to 500MB / s.

9. The high-frequency sampling signal processing method based on a high-density array pressure sensor according to claim 1, characterized in that: In S7, the output data format is 16-bit raw pressure data + 32-bit timestamp + 8-bit block number, totaling 56 bits per sampling point.

Images