Fluorescence data denoising demodulation method and related device

Abstract

This invention relates to the field of quantum precision measurement, and discloses a fluorescence data denoising and demodulation method and related equipment. The fluorescence data denoising and demodulation method proposed in this solution can move the center line of fluorescence digital signal data at different carrier frequencies to near the zero axis, effectively reducing data offset and ensuring the accuracy of quantum precision measurement.

Description

Technical Field

[0001] This invention relates to the field of quantum precision measurement technology, specifically to a fluorescence data noise reduction and demodulation method, a fluorescence data noise reduction and demodulation device, a quantum sensing device, and a storage medium. Background Technology

[0002] Nitrogen-vacancy (NV) centers in diamond are among the most studied solid-state spin center systems, exhibiting immense potential for applications in quantum precision measurement due to their millisecond-level spin coherence time at room temperature, atomic-level spatial resolution, and optically readable quantum properties. Quantum sensing technology based on NV centers is primarily achieved through the optically detected magnetic resonance (ODMR) method. Its core principle involves initializing the NV center's spin state to its ground state using a laser, manipulating it with a microwave field, and obtaining spin state information by detecting changes in fluorescence signal intensity, thereby enabling precise measurement of physical quantities such as magnetic fields, temperature, and electric fields.

[0003] In quantum sensing systems based on NV centers, a photodetector is used to acquire the photofluorescence signal generated by the diamond NV centers under the action of excitation light and a microwave field, and converts the optical signal into an analog electrical signal output. An analog-to-digital converter (ADC) acquires the analog signal output from the photodetector, converting the continuous analog signal into a discrete digital signal for subsequent processing such as data filtering, demodulation, and magnetic field calculation. In practical applications, to obtain the fluorescence response of the NV centers at different microwave frequencies, it is usually necessary to scan the microwave frequency or use frequency-modulated microwaves, with the ADC continuously acquiring the fluorescence signal during frequency changes.

[0004] However, during microwave frequency changes, the simulated fluorescence signal data acquired by the ADC exhibits a zero-axis shift. Specifically, as shown in the attached diagram... Figure 1 and Figure 2As shown, within different frequency ranges (the figure only controls the switching between two frequencies, f1 and f2, for illustration, where f1 = 2.83 GHz and f2 = 2.8711 GHz), the acquired analog data fluctuates around its respective center line, and these center lines are offset from the zero axis to varying degrees, none of them coinciding with the zero axis. This zero-axis offset phenomenon is related to several factors, including the transient response during microwave frequency switching, the operating characteristics of the photodetector, and the non-ideal characteristics of the ADC sampling circuit. When analog data with zero-axis offset directly enters subsequent signal processing stages such as filtering and phase-locked demodulation, the offset is introduced into the calculation process, leading to errors in the demodulation results, and thus affecting the accuracy and precision of the magnetic field measurement. Therefore, how to eliminate or suppress the zero-axis offset in the ADC acquired data and improve the accuracy of signal processing is an important problem to be solved in current NV color center quantum sensing technology. Summary of the Invention

[0005] This invention proposes a fluorescence data denoising and demodulation method, a fluorescence data denoising and demodulation device, a quantum sensing device, and a storage medium to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A fluorescence data denoising and demodulation method, applied to a sensing process based on diamond NV centers and using photodetector magnetic resonance, includes: S1. Select several NV axes in the diamond NV color center as sampling axes, and let two pairs of magnetic characterization frequencies of each sampling axis be used as data sampling points, and record them as the left sampling point and the right sampling point of the sampling axis. The sampling object is the fluorescence digital signal data at each data sampling point. During sampling, the two-point alternating sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point. S2. Preprocessing the fluorescent digital signal data, the preprocessing includes: -Based on the fluorescence digital signal data obtained from the same side data sampling point of the same sampling axis in the previous sampling, the mean value is calculated to obtain the zero axis offset value and stored; - Perform a subtraction operation between the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the current sampling and the zero axis offset value obtained from the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the previous sampling, and obtain the preprocessed fluorescence digital signal data of the current sampling. S3. Perform real-time filtering and demodulation on the obtained preprocessed current subfluorescence digital signal data.

[0007] In a preferred design of the fluorescence data denoising and demodulation method described above, the signal duration of the fluorescence digital signal data used for averaging calculation at each data sampling point is an integer number of microwave modulation cycles.

[0008] In the fluorescence data denoising and demodulation method described above, in a preferred design, the signal duration of the obtained preprocessed fluorescence digital signal data is not less than the filtering duration.

[0009] In a preferred design of the fluorescence data denoising and demodulation method described above, the fluorescence digital signal data used for averaging does not contain microwave switching transient interference data.

[0010] Another aspect of this application also introduces a fluorescence data noise reduction and demodulation device, applied to a process based on diamond NV color centers and using photodetector magnetic resonance sensing, comprising: Signal sampling module: Select several NV axes in the diamond NV center as sampling axes, and let two pairs of magnetic characterization frequencies of each sampling axis be used as data sampling points, and record them as the left sampling point and the right sampling point of the sampling axis. The sampling object is the fluorescence digital signal data at each data sampling point. During sampling, the two-point alternating sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point. Signal preprocessing module: performs preprocessing on the fluorescence digital signal data, the preprocessing including: -Based on the fluorescence digital signal data obtained from the same side data sampling point of the same sampling axis in the previous sampling, the mean value is calculated to obtain the zero axis offset value and stored; - Perform a subtraction operation between the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the current sampling and the zero axis offset value obtained from the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the previous sampling, and obtain the preprocessed fluorescence digital signal data of the current sampling. Noise reduction and demodulation module: performs real-time filtering and demodulation on the preprocessed current subfluorescent digital signal data.

[0011] In a preferred design of the fluorescence data noise reduction and demodulation device described above, in the signal preprocessing module, the signal duration of the fluorescence digital signal data used for averaging calculation at each data sampling point is an integer number of microwave modulation cycles.

[0012] In a preferred design of the fluorescence data noise reduction and demodulation device described above, the signal duration of the preprocessed fluorescence digital signal data obtained in the signal preprocessing module is not less than the filtering duration.

[0013] In a preferred design of the fluorescence data noise reduction and demodulation device described above, the fluorescence digital signal data used for averaging in the signal preprocessing module does not contain microwave switching transient interference data.

[0014] Another aspect of this application describes a quantum sensing device comprising: Diamond NV color centers are diamond blocks containing ensemble NV color centers. The laser module is used to transmit excitation light to the diamond NV color center; A microwave module is used to radiate modulated microwaves to the diamond NV color center; The photoelectric detection module is used to collect the fluorescence generated by the diamond NV color center under the dual action of excitation light and modulated microwave; The quantum sensing device, as described above, is characterized in that it further comprises: The PID module is used to track two pairs of spin resonant frequencies along at least one NV axis in the diamond NV center and give frequency adjustment signals in turn. The microwave module receives the frequency adjustment signals in turn and switches the carrier frequency to the corresponding spin resonant frequency. In the fluorescence data noise reduction and demodulation device described above, the sampling object of the signal sampling module is the fluorescence digital signal data output by the photoelectric detection module.

[0015] In a preferred design of the quantum sensing device described above, the signal duration of the fluorescent digital signal data at each data sampling point is equal to a filtering duration.

[0016] In a preferred design of the quantum sensing device described above, a magnetizer is also included, which is used to concentrate the magnetic field and form a magnetizing magnetic field parallel to the first NV axis of the diamond NV center; the PID module is used to track two pairs of spin resonant frequencies along the first NV axis of the diamond NV center and sequentially provide frequency adjustment signals.

[0017] In another aspect, this application also describes a storage medium storing a computer program that, when executed by data processing hardware, enables the operation of the fluorescence data noise reduction and demodulation method described above.

[0018] Compared with the prior art, the beneficial effects of the present invention are: the fluorescence data noise reduction and demodulation method proposed in this solution can move the center line of fluorescence digital signal data at different carrier frequencies to near the zero axis, effectively reducing the data offset and ensuring the accuracy of quantum precision measurement. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a frequency-time relationship diagram when the carrier switches between frequency f1 and frequency f2. Figure 2 To and Figure 1 Corresponding ADC sampling result diagram; Figure 3 This is a flowchart illustrating the fluorescence data noise reduction and demodulation method in Example 1; Figure 4 This is the ODMR spectrum obtained by tracking the resonance frequencies along the four NV axes in Example 1; Figure 5 This is a system block diagram of the fluorescence data noise reduction and demodulation device in Example 2; Figure 6 This is a system block diagram of the diamond NV color center quantum device in Example 3; Figure 7 This is a schematic diagram of the hardware structure of the storage medium in Embodiment 4. Detailed Implementation

[0021] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, one or more embodiments are now described with reference to the accompanying drawings, wherein similar reference numerals are used throughout the text to refer to similar components. In the following description, numerous specific details are set forth for purposes of explanation in order to provide a more thorough understanding of one or more embodiments. However, it will be apparent that one or more embodiments may be practiced in various circumstances without these specific details, and the various embodiments may be combined with and referenced to each other without contradiction.

[0023] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0024] Quantum sensing technology based on NV centers primarily achieves measurements through optically detected magnetic resonance (ODMR). The core measurement principle of ODMR is as follows: The electron spin ground state of the NV center is a spin triplet, containing three energy levels: m_s=0 and m_s=±1. Under zero magnetic field conditions, the m_s=±1 state is degenerate; however, under the influence of an external magnetic field, due to the Zeeman effect, the m_s=+1 and m_s=-1 states undergo energy level splitting, the magnitude of which is proportional to the strength of the external magnetic field. During measurement, the NV center is first irradiated with a green laser with a wavelength of approximately 532 nm to initialize the electron spin state to the m_s=0 ground state; then, a frequency-tunable microwave field is applied. When the microwave frequency precisely matches the energy level difference between the m_s=0 and m_s=±1 states, a resonant transition occurs in the NV center, and some electrons are flipped to the m_s=±1 state. Because the red fluorescence intensity emitted by the NV color center in the m_s=0 state is significantly stronger than that in the m_s=±1 state, the fluorescence intensity detected at the resonance frequency will decrease significantly, forming a resonance valley. The location of the resonance frequency can be determined by scanning the microwave frequency and monitoring the change in fluorescence intensity. Under the influence of an external magnetic field, the m_s=±1 state splits into two energy levels, corresponding to two spin resonance frequencies symmetrical about the central resonance frequency. The difference between these two frequencies is linearly related to the strength of the external magnetic field. By measuring the frequency difference and combining it with the electron spin gyromagnetic ratio, the magnitude of the external magnetic field can be accurately calculated.

[0025] In the above measurement process, a photodetector is used to collect the photoluminescence signal generated by the diamond NV center under the action of excitation light and microwave field, and convert the optical signal into an analog electrical signal output. An analog-to-digital converter (ADC) collects the analog signal output from the photodetector, converting the continuous analog signal into a discrete digital signal for subsequent data filtering, demodulation, and magnetic field calculation. In practical applications, to obtain the fluorescence response of the NV center at different microwave frequencies, it is usually necessary to scan the microwave frequency and modulate the microwave, with the ADC continuously collecting the fluorescence signal during frequency changes.

[0026] However, during microwave frequency changes, the analog fluorescence signal acquired by the ADC exhibits a zero-axis offset. Specifically, within different frequency ranges, the acquired analog data fluctuates around its respective center line, with these center lines deviating to varying degrees from the zero axis. This zero-axis offset is related to several factors, including the transient response during microwave frequency switching, the operating characteristics of the photodetector, and the non-ideal characteristics of the ADC sampling circuit. When analog data with this zero-axis offset is directly input into subsequent signal processing stages such as filtering and phase-locked demodulation, the offset is introduced into the calculation process, leading to errors in the demodulation results and consequently affecting the accuracy and precision of the magnetic field measurement.

[0027] Considering the above problems, this plan proposes some solutions.

[0028] Regarding the term "magnetic characterization frequency" used in this application, it should be specifically noted that this term is a broad concept. In this application, unless specifically defined as "spin resonance frequency," "magnetic characterization frequency" should be understood to include, but is not limited to, the following types of frequencies: spin resonance frequency (the zero-crossing point of the ODMR curve, a characteristic frequency easily tracked by the PID module), frequency in the linear region of the ODMR curve (the region on both sides of the resonance peak that changes approximately linearly), frequency at the point of maximum slope (the position where the slope of the ODMR curve is the largest, which usually has high sensitivity), and other characteristic frequencies that can be used to calculate the magnetic field. Those skilled in the art will understand that, under different measurement schemes and application scenarios, appropriate magnetic characterization frequencies can be selected as data sampling points according to actual needs.

[0029] The term "spin resonance frequency" used in this application specifically refers to the resonance transition frequency between the m_s=0 state and the m_s=±1 state in the ODMR spectrum, i.e., the zero-crossing point of the ODMR curve. At this frequency, the fluorescence signal is at the zero point of the ODMR curve, making it easy for the PID module to track and lock onto, and it is a commonly used reference frequency in quantum precision measurement. In some embodiments of this application, to explicitly define the measurement accuracy and tracking characteristics of the device, the data sampling point is limited to the "spin resonance frequency".

[0030] The term "zero-axis offset" used in this application refers to the numerical value corresponding to the phenomenon where the ADC output data deviates from the zero axis on the spectral line. Specifically, due to factors such as the transient response during microwave frequency switching, the operating characteristics of the photodetector, and the non-ideal characteristics of the ADC sampling circuit, the center line of the data acquired by the ADC when acquiring fluorescence signals is not located at the zero axis position, but rather has a certain offset relative to the zero axis. This offset is the "zero-axis offset value." This solution preprocesses the fluorescence digital signal data to eliminate this zero-axis offset value, so that the processed data center line coincides with the zero axis.

[0031] Several examples are listed below to illustrate the relevant solutions. Example 1

[0032] This example introduces a fluorescence data denoising and demodulation method applied to a sensing process based on diamond NV centers and using photodetector magnetic resonance, as shown in the attached figure. Figure 3 As shown, it includes steps S1-S3, wherein: S1. Select several NV axes in the diamond NV color center as sampling axes, and let two pairs of magnetic characterization frequencies of each sampling axis be used as data sampling points, and record them as the left sampling point and the right sampling point of the sampling axis. The sampling object is the fluorescence digital signal data at each data sampling point. During sampling, the two-point alternating sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point.

[0033] Specifically, the diamond NV center contains multiple NV axes of various crystal orientations. This scheme selects several NV axes as sampling axes. For each sampling axis, under the influence of an external magnetic field, due to the Zeeman effect, the m_s=+1 and m_s=-1 states corresponding to that NV axis undergo energy level splitting, forming two magnetic characterization frequencies symmetrical about the center frequency. This scheme denotes these two paired magnetic characterization frequencies as the "left sampling point" and "right sampling point" of that sampling axis, respectively, where the left sampling point corresponds to a lower frequency and the right sampling point corresponds to a higher frequency. For example, the left sampling point can correspond to the transition frequency between the m_s=0 state and the m_s=-1 state, and the right sampling point can correspond to the transition frequency between the m_s=0 state and the m_s=+1 state.

[0034] Regarding the selection of the number of sampling axes, for example, in single-crystal diamond containing ensemble NV centers, there are four NV axes. This solution can select three or four NV axes as sampling axes, such as in vector magnetic field measurement scenarios. For example, in some other application scenarios, only one or two NV axes can be selected as sampling axes to reduce data processing complexity and improve curve data contrast. The number of sampling axes can be flexibly selected according to actual measurement requirements.

[0035] The specific process of alternating dual-point sampling is illustrated using four NV axes (denoted as NV axis 1, NV axis 2, NV axis 3, and NV axis 4) as sampling axes, as shown in the attached diagram. Figure 4 As shown: Each NV axis has two data sampling points: a left sampling point (L) and a right sampling point (R). Therefore, there are a total of eight data sampling points across the four NV axes: NV axis 1 - left sampling point (1L), NV axis 1 - right sampling point (1R), NV axis 2 - left sampling point (2L), NV axis 2 - right sampling point (2R), NV axis 3 - left sampling point (3L), NV axis 3 - right sampling point (3R), NV axis 4 - left sampling point (4L), and NV axis 4 - right sampling point (4R). During alternating sampling, each sampling axis is sampled in a preset order. For example, the sampling sequence can be designed as follows: NV axis 1 - left sampling point (1L) → NV axis 1 - right sampling point (1R) → NV axis 2 - left sampling point (2L) → NV axis 2 - right sampling point (2R) → NV axis 3 - left sampling point (3L) → NV axis 3 - right sampling point (3R) → NV axis 4 - left sampling point (4L) → NV axis 4 - right sampling point (4R) → return to NV axis 1 - left sampling point (1L) → ..., and so on.

[0036] For example, when only one NV axis is selected as the sampling axis, this sampling axis has only two data sampling points, namely the left sampling point (L) and the right sampling point (R). When sampling at two points alternately, the sampling sequence can be designed as: left sampling point (L) → right sampling point (R) → left sampling point (L) → right sampling point (R) → ..., and so on. Since the sampling process is very fast, the time interval between two adjacent sampling points (such as two consecutive left sampling points or two consecutive right sampling points) is short. Therefore, the changes in the fluorescent digital signal data at two sampling points on the same side are minimal, and their zero-axis offset values ​​are almost equal, which helps to improve the accuracy of zero-axis offset correction.

[0037] Of course, the above sampling order is only an example, and those skilled in the art can design other sampling orders according to actual needs.

[0038] When sensing diamond NV centers using the ODMR method, the fluorescence signal generated by the diamond NV centers is selectively collected by a photodetector, converted from an optical signal to an analog electrical signal, and then acquired by an ADC. This analog signal is converted from a continuous analog signal into a discrete digital signal for subsequent processing such as data filtering, demodulation, and magnetic field calculation. This digital signal is the fluorescence digital signal data discussed here. During the dual-point alternating sampling process, the carrier frequency of the microwave module switches accordingly based on the currently sampled data point, and the ADC synchronously acquires the fluorescence digital signal data at that carrier frequency.

[0039] The modulation-demodulation-based measurement scheme uses modulated microwaves to irradiate the diamond NV color center, and then demodulates the fluorescence signal based on the modulated signal. Typically, frequency modulation (FM) or amplitude modulation (AM) signals are used to modulate the carrier wave. The purpose is to shift the low-frequency measurement signal to a higher frequency through FM or AM modulation, which facilitates subsequent demodulation by a lock-in amplifier, effectively suppresses low-frequency noise and DC drift, and improves the signal-to-noise ratio.

[0040] Regarding the zero-axis offset mentioned above, its specific manifestation is as follows: during the acquisition time period corresponding to different carrier frequencies, the fluorescent digital signal data will fluctuate around their respective center lines, and there are different degrees of offset between the center lines and the zero axis at different carrier frequencies. For example, the center lines at different carrier frequencies are located above or below the zero axis, forming the following... Figure 1 The data in the middle are all deviating from the zero axis.

[0041] The zero-axis offset problem occurs when the frequency changes. Specifically, the fluorescent digital signal data that generates the zero-axis offset is produced when at least two modulated microwaves with different carrier frequencies act on the diamond NV center. For example, in practical applications of NV center quantum sensing, in order to achieve real-time continuous measurement of dynamically changing magnetic fields, it is usually necessary to make the output frequency of the microwave source track the spin resonance frequency of the NV center in real time. For this purpose, a proportional-integral-derivative (PID) control module is often used to perform closed-loop adjustment of the microwave source output frequency (carrier frequency) to achieve automatic tracking and locking of the magnetic characterization frequency. For example, in an application scenario where the frequency is tracked and locked based on a PID control module, the magnetic characterization frequency can be specifically selected as the spin resonance frequency. In this scheme, only one carrier frequency of the modulated microwave acts on the diamond NV color center at the same time. The calculation of external physical quantities requires fluorescence data at a pair of spin resonance frequencies within a short time. Therefore, the carrier frequency of the modulated microwave will switch back and forth between the pair of spin resonance frequencies. This inevitably leads to the acquisition of fluorescence digital signal data with zero axis offset when the ADC is used to continuously acquire the output signal of the photodetector.

[0042] Of course, the application scenarios listed here are only one example of the applicable scenarios for the method proposed in this solution. Other scenarios that meet the same problem can also use the method proposed in this solution. For example, in some fixed-frequency measurement schemes, the carrier frequency of the modulated microwave will switch rapidly between two preset fixed frequency points. These two frequency points are usually set at positions with large slopes on both sides of the spin resonance peak of the NV color center. When the external magnetic field changes, the spin resonance frequency of the NV color center will shift, causing the position of the resonance peak relative to the two fixed carrier frequency points to change, thereby changing the difference in fluorescence response at the two frequency points. By demodulating this difference signal, the current position of the spin resonance frequency can be calculated in reverse, thereby obtaining the magnetic field value. In this scheme, the carrier frequency of the modulated microwave is a preset fixed value that does not change in real time with the magnetic field. Instead, the shift of the resonance frequency is passively sensed by the differential response of the fluorescence signal at the two frequencies. This is different from the aforementioned scheme where the PID control module tracks and locks the resonance frequency in real time, but the noise reduction and demodulation method of the fluorescence data recorded in this scheme is also applicable.

[0043] S2. Preprocessing the fluorescent digital signal data, the preprocessing includes: -Based on the fluorescence digital signal data obtained from the same side data sampling point of the same sampling axis in the previous sampling, the mean value is calculated to obtain the zero axis offset value and stored; - Perform a subtraction operation between the fluorescence digital signal data obtained from the sampling point on the same side of the same sampling axis in the current sampling and the zero axis offset value obtained from the fluorescence digital signal data obtained from the sampling point on the same side of the previous sampling axis, to obtain the preprocessed fluorescence digital signal data for the current sampling.

[0044] Step S2 is the key step in solving the aforementioned zero-axis offset problem. Its core idea is to use the method of one-to-one correspondence of the same data sampling points of the front and rear wheels to correct the zero-axis offset.

[0045] The key to this scheme lies in the constraint of "one-to-one correspondence between sampling points of the front and rear wheels". Specifically, when sampling a certain data sampling point (such as NV axis 1-left sampling point 1L) for the current wheel, its zero axis offset value is calculated based on the fluorescent digital signal data collected at the same data sampling point (NV axis 1-left sampling point 1L) of the previous wheel; subsequently, the fluorescent digital signal data collected by the current wheel is subtracted from the zero axis offset value to complete the zero axis offset correction.

[0046] The definition of "front and rear wheels" depends on the sampling period. For a specific data sampling point, the time interval between two consecutive arrivals at that sampling point is one sampling period. For example, taking the aforementioned selection of four NV axes and eight data sampling points as an example, if the sampling duration of each data sampling point is T, then completing one round of acquisition of eight data sampling points requires 8T time; correspondingly, for a specific data sampling point (such as 1L), the time interval between two consecutive arrivals at that sampling point is 8T, which is the sampling period of that data sampling point, the previous sampling is the previous round, and the current sampling is the current round.

[0047] The "one-to-one correspondence" constraint ensures the accuracy and timeliness of the zero-axis offset calculation. Accuracy is reflected in the fact that the previous and current rounds collect fluorescence signals from the same data sampling points. These sampling points correspond to the same carrier frequency and physical conditions, therefore the calculated zero-axis offset accurately reflects the degree to which the current data deviates from the zero axis. Timeliness is reflected in the fact that, due to the short time interval between rounds (usually one sampling period), changes in the external environment (such as temperature drift, laser power fluctuations, etc.) have a small impact on the zero-axis offset, allowing the calculated zero-axis offset to track changes in the offset in a timely manner.

[0048] Regarding the specific process of obtaining the zero-axis offset value through mean calculation, for example, taking N fluorescent digital signal data points obtained from the same side of the same sampling axis in the previous sampling as an example, these N data points are summed and then divided by the number of data points N to obtain the mean result, which is the zero-axis offset value. Preferably, the signal duration of the fluorescent digital signal data used for mean calculation is an integer number of microwave modulation cycles. This is because the fluorescent digital signal contains a modulation signal component. When the data segment covers an integer number of modulation cycles, the positive and negative half-cycles of the modulation signal component can cancel each other out, thereby extracting the DC offset component, i.e., the zero-axis offset value, more accurately. Of course, in practical applications, data segments with non-integer number of modulation cycles can also be used to calculate the zero-axis offset value. Although the accuracy may be reduced, it is still applicable in scenarios where the accuracy requirement for the offset value is not high.

[0049] Regarding the data source for the mean calculation, for example, the zero-axis offset value can be obtained by averaging multiple data points from the previous single sampling. This is a preferred implementation of this solution. Since the data from the previous single sampling is the most recent data, the zero-axis offset information it contains is closest to the zero-axis offset of the current data, thus the calculated zero-axis offset value is the most accurate.

[0050] In this embodiment, the fluorescent digital signal data used to calculate the zero-axis offset value comes from the fluorescent digital signal data obtained from the same side data sampling point of the same sampling axis in the previous sampling. This is a preferred embodiment of the present invention. The technical advantages of using the previous data are: on the one hand, the previous data is the most recent historical data, and the zero-axis offset information contained in it is closest to the zero-axis offset of the current data, so the calculated zero-axis offset value is the most accurate; on the other hand, using only the previous data has low data storage overhead and low computational load, which is conducive to achieving high-speed real-time processing.

[0051] It should be noted that those skilled in the art will understand that the phrase "previous time" is a preferred limitation on the timing of data sourcing, not an exclusive limitation. From a technical perspective, zero-axis offset arises from factors such as the transient response during microwave frequency switching, the operating characteristics of photodetectors, and the non-ideal characteristics of the ADC sampling circuit. The data offset caused by these factors exhibits relatively stable characteristics over a short period. Therefore, within a certain time range, the zero-axis offset values ​​of the fluorescent digital signal data acquired from the same sampling point on the same side of the same sampling axis in the previous few sampling iterations all show good correlation with the zero-axis offset value of the current data.

[0052] Based on the above technical principles, those skilled in the art can obtain the zero-axis offset value using the following variant implementation methods. These variant implementation methods are substantially the same as the preferred implementation methods of the present invention in terms of technical means and technical effects, and belong to the equivalent implementation methods of the present invention: Method 1: Perform mean calculation on the fluorescent digital signal data obtained from the same side of the same sampling axis in the second, third, or Nth sampling (N is an integer greater than 1).

[0053] Method 2: Combine fluorescent digital signal data obtained from the same sampling points on the same side of the same sampling axis in the previous multiple samplings (such as the first two, three, etc.) for calculation. For example, the mean of the first two data points can be used as the zero-axis offset value, or the weighted average of the first three data points can be used as the zero-axis offset value. This method can further reduce random errors and improve the calculation accuracy of the zero-axis offset value by statistically processing multiple historical data.

[0054] Method 3: Use the mean or median of multiple historical data points within a time window as the zero-axis offset value. For example, a sliding time window can be set, and the mean or median of the fluorescence digital signal data obtained at the same sampling point on the same side of the same sampling axis within the window can be calculated as the zero-axis offset value for the current data correction.

[0055] It is important to emphasize that regardless of the specific implementation method described above, the core principle remains the same: using fluorescence digital signal data acquired from sampling points on the same side of the same historical sampling axis to estimate the zero-axis offset value, and then using this data for correction of the current data. This core technical feature differs from the preferred implementation only in the time range of the historical data selection; the data processing techniques (averaging or similar operations), the purpose of data processing (obtaining the zero-axis offset value), and the effect of data processing (achieving zero-axis offset correction) are essentially the same. Therefore, the aforementioned implementation methods should not be considered as departing from the technical concept of this invention, but rather should be understood as equivalent features to the technical feature of "fluorescence digital signal data acquired from sampling points on the same side of the same sampling axis in the previous iteration."

[0056] Those skilled in the art should understand that, in practical applications, a balance can be struck between the calculation accuracy of the zero-axis offset value, data storage overhead, and calculation real-time performance, and the above-described implementation method or variant implementation method can be flexibly selected according to the specific needs of the scenario.

[0057] The specific process of the difference operation involves subtracting the fluorescence digital signal data acquired at the same sampling point on the same side of the current sampling axis from the pre-stored zero-axis offset value point by point to obtain the pre-processed fluorescence digital signal data for the current sampling. Preferably, the difference operation can be performed point by point, that is, the difference operation is immediately performed with the zero-axis offset value after each data point is acquired and the result is output, without waiting for the entire sampling cycle to be completed, thus ensuring real-time performance.

[0058] Regarding the timing of zero-axis offset calculation, for example, the zero-axis offset value of the current data can be calculated and stored in parallel while the current data is being used for difference correction, so that it can be used by the next data sampling point on the same side of the same sampling axis. Specifically, when the current round of data for a certain data sampling point arrives, the zero-axis offset value calculated in the previous round for that sampling point is first read from the storage unit, difference correction is performed on the current round of data, and the result is output to the filtering and demodulation stage; at the same time, the mean value is calculated on the current round of data to obtain the updated zero-axis offset value, which is then written to the storage unit, overwriting or updating the historical zero-axis offset value of that sampling point. This parallel processing method enables dynamic updating of the zero-axis offset value without blocking the data processing flow, ensuring the real-time performance of the measurement.

[0059] To avoid the impact of transient interference data during microwave switching on the calculation of the zero-axis offset, the transient interference data can be skipped before the averaging operation, for example. At the instant the carrier frequency of the modulated microwave switches, the microwave source output is not yet stable, which may cause abnormal fluctuations in the fluorescent digital signal data. Skipping this data can improve the accuracy of the zero-axis offset calculation. The following method can be used as an example to skip transient interference data: Manual implementation method: First, determine the duration of the transition signal corresponding to the transient interference data during microwave switching. This duration can be obtained by engineers through observation and experience of the actual signal. Then, set a time threshold that is slightly longer than the duration of the transition signal. Delay the acquisition of valid data from the microwave switching time point by this time threshold, thereby skipping the transient interference data.

[0060] Automatic implementation: After microwave frequency switching, the signal variance or standard deviation within a sliding time window is automatically calculated. When the variance or standard deviation exceeds a preset threshold, the data within that time period is determined to be transient interference data. When the variance or standard deviation drops below the preset threshold and remains stable, the transient interference data is considered to have ended, and valid data collection begins. This method requires no manual intervention and can automatically identify and skip transient interference data.

[0061] Regarding the signal duration of the obtained preprocessed fluorescence digital signal data, preferably, it can be set to be no less than the signal duration required for filtering (i.e., the filtering duration) to meet the data volume requirements of the filtering algorithm and ensure the filtering effect. Of course, in practical applications, the signal duration setting can be flexibly adjusted according to specific needs. For example, in the case of pursuing the limit of rapid detection, the signal duration of the fluorescence digital signal data used for averaging can be set to one microwave modulation cycle, and the signal duration of the obtained preprocessed fluorescence digital signal data can be set to be equal to one filtering duration, thereby maximizing the detection speed while ensuring the preprocessing effect.

[0062] S3. Perform real-time filtering and demodulation on the obtained preprocessed fluorescent digital signal data.

[0063] Step S3 involves filtering and demodulating the preprocessed fluorescence digital signal data to extract the effective signal for magnetic field calculation. This step is a standard processing step in the NV color center quantum sensing measurement process, and its specific implementation can be selected and adjusted according to actual application requirements.

[0064] The purpose of filtering is to suppress noise components in the signal and improve the signal-to-noise ratio (SNR). For example, at the front end of the electrical link, a low-pass filter can be used to filter the preprocessed fluorescent digital signal data. The cutoff frequency of the filter can be set to, for example, several kilohertz to tens of kilohertz to suppress high-frequency noise. After demodulation, a low-pass filter can also be used to remove high-frequency harmonic components and noise generated during demodulation, retaining only the DC component. Of course, the filtering method is not limited to low-pass filtering; in practical applications, band-pass filtering, high-pass filtering, or other types of filters can be selected according to the signal characteristics. Furthermore, in the digital post-processing stage, composite filtering methods such as wavelet transform and adaptive filtering can be superimposed to further improve the SNR.

[0065] Regarding demodulation processing, its purpose is to extract the response component related to the modulated signal from the modulated fluorescence signal. Taking frequency-modulated microwave as an example, the carrier frequency of the modulating microwave can be set around the zero-field splitting frequency of the NV color center (approximately 2.87 GHz), specifically falling within the linear region of the ODMR spectrum (e.g., 2.85-2.89 GHz) to obtain a larger demodulation slope. The modulation frequency can be set in the range of hundreds of hertz to thousands of hertz, preferably around 1 kHz, to balance noise suppression and response speed. Of course, the above parameters and methods are only examples, and can be flexibly adjusted according to specific measurement requirements and hardware conditions in practical applications.

[0066] The timing relationship between filtering and demodulation can be exemplified as follows: first, the preprocessed current fluorescence digital signal data is filtered, and then demodulated. The resulting DC voltage signal after demodulation can be directly used for magnetic field inversion calculations. Of course, the specific order and combination of filtering and demodulation can be adjusted according to the actual scheme.

[0067] For example, in a dual-point alternating sampling scenario, filtering and demodulation of the data segment can be performed immediately after the preprocessing of the fluorescence digital signal data of each sampling point is completed, without waiting for the entire sampling cycle to complete. This point-by-point processing method can effectively ensure the real-time performance of the measurement and is particularly suitable for applications where the external magnetic field changes rapidly and a high sampling rate is required. Example 2

[0068] Corresponding to Example 1, this example proposes a corresponding fluorescence data noise reduction and demodulation device, applied to a process based on diamond NV color centers and using photodetector magnetic resonance (PDM) for sensing. See Appendix. Figure 5 It includes a signal sampling module, a signal preprocessing module, and a noise reduction and demodulation module.

[0069] In this example, the signal sampling module is used to select several NV axes in the diamond NV center as sampling axes, and let two pairs of magnetic characterization frequencies of each sampling axis be used as data sampling points, and denoted as the left sampling point and the right sampling point of the sampling axis. The sampling object is the fluorescence digital signal data at each data sampling point. During sampling, the two-point alternating sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point.

[0070] The signal sampling module is one of the core components of this device. Its function is to select the sampling axis, set the data sampling points, and perform alternating two-point sampling of the fluorescence digital signal data. Specifically, the signal sampling module selects several NV axes in the diamond NV center as sampling axes, and designates two pairs of magnetic characterization frequencies of each sampling axis as data sampling points, which are denoted as the left and right sampling points of that sampling axis. During the sampling process, alternating two-point sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point.

[0071] For example, the signal sampling module may include an ADC (Analog-to-Digital Converter) and its control unit as necessary components. The ADC is used to convert the input analog electrical signal into a discrete digital signal, and its control unit is used to control the sampling timing, data transmission, and other operations of the ADC. For example, the sampling rate of the ADC can be set to the range of several kilohertz to several megahertz, preferably several hundred kilohertz to several megahertz, to meet the acquisition requirements of NV color center fluorescence signals; the resolution of the ADC can be, for example, 12-bit, 14-bit, or 16-bit, preferably 16-bit, to improve sampling accuracy. For example, the ADC can be implemented using a main control chip with integrated ADC, such as the STM32H750VBT6. This type of chip integrates multiple high-precision SAR-ADCs and supports features such as buffering, DMA (Direct Memory Access), and hardware oversampling, enabling high-speed analog signal acquisition.

[0072] For example, the signal sampling module may further include a photodetector. The photodetector receives the fluorescence signal generated by the diamond NV color center and converts the optical signal into an analog electrical signal output. For example, the photodetector may be a photomultiplier tube (PMT), an avalanche photodiode (APD), a multi-pixel photon counter (MPPC / SiPM), or a photodiode (PD). Preferably, for the fluorescence signal emitted by the NV color center (wavelength range of approximately 600-850 nm), a photodetector with high quantum efficiency corresponding to this wavelength range can be used. For example, a photomultiplier tube has a gain of up to 10^6 to 10^7 times, suitable for detecting extremely weak fluorescence signals; an avalanche photodiode has a fast response speed (nanosecond level), suitable for scenarios requiring fast time-resolved measurements. Of course, the type and parameters of the photodetector can be flexibly selected based on factors such as the actual fluorescence signal intensity, measurement accuracy requirements, and cost.

[0073] Exemplarily, the signal sampling module may further include a microwave source. The microwave source generates a modulated microwave signal and radiates it to the diamond NV center via an antenna or microwave transmission line. Exemplarily, the output frequency range of the microwave source may cover approximately 2.0 GHz to 3.5 GHz, preferably approximately 2.7 GHz to 3.1 GHz, to cover the zero-field splitting frequency (approximately 2.87 GHz) and the linear region of the ODMR spectrum of the NV center. Exemplarily, the microwave source may support frequency modulation (FM) or amplitude modulation (AM), with the modulation frequency exemplarily ranging from hundreds of hertz to thousands of hertz, preferably set around 1 kHz. Exemplarily, the microwave signal output by the microwave source may be amplified by a power amplifier and radiated to the diamond NV center via a microwave transmission structure such as an antenna or copper wire.

[0074] For example, the signal sampling module may further include a sampling control unit. The sampling control unit controls the timing of the alternating dual-point sampling, enabling cyclic sampling of each sampling axis. For example, the sampling control unit can control the microwave source to switch to the corresponding carrier frequency according to a preset sampling sequence table, simultaneously triggering the ADC to acquire fluorescence signal data at that frequency. For example, the sampling control unit can further manage the selection of the number of sampling axes, configuring the number of NV axes participating in sampling (e.g., one, two, three, or four NV axes) according to actual measurement requirements. For example, the sampling control unit can be implemented using an FPGA or microcontroller, which generates precise sampling control signals based on the clock signal and sampling parameters to ensure the accuracy and repeatability of the sampling timing.

[0075] The interconnection and collaboration between the components are exemplarily described as follows: The photodetector receives the fluorescence signal generated by the diamond NV color center, converts the optical signal into an analog electrical signal, and transmits this analog electrical signal to the ADC for sampling. Under the control of the control unit, the ADC converts the continuous analog signal into discrete fluorescent digital signal data. The modulated microwave signal generated by the microwave source is amplified by the power amplifier and radiated to the diamond NV color center. The carrier frequency of the microwave source is switched according to the instructions of the sampling control unit. The sampling control unit coordinates the timing of the microwave source and the ADC to achieve alternating sampling at two points according to a preset sampling sequence. Of course, the connection method of the above components can be flexibly adjusted according to the actual system architecture.

[0076] Of course, the specific composition of the signal sampling module is not limited to the example above. Other combinations of components capable of achieving alternating dual-point acquisition of fluorescent digital signal data can also be applied to this device. For example, if the signal sampling module does not include a photodetector or microwave source, the module can be mounted as an independent sampling unit in an existing diamond NV color center device, which itself already includes components such as a photodetector and a microwave source.

[0077] In this example, the signal preprocessing module is used to preprocess the fluorescence digital signal data. The preprocessing includes: first, performing a mean operation on the fluorescence digital signal data obtained from the same sampling point on the same sampling axis in the previous sampling to obtain a zero-axis offset value and storing it; second, performing a difference operation between the fluorescence digital signal data obtained from the current sampling point on the same sampling axis and the pre-stored zero-axis offset value obtained from the fluorescence digital signal data obtained from the same sampling point on the same sampling axis in the previous sampling, to obtain the preprocessed current fluorescence digital signal data.

[0078] The signal preprocessing module is a key component of this device for zero-axis offset correction. Its function is to perform zero-axis offset correction by using a one-to-one correspondence between the sampling points of the front and rear wheels.

[0079] For example, the signal preprocessing module may include a data processing unit and a storage unit as necessary components. The data processing unit is used to perform operations such as averaging, subtraction, and data caching; the storage unit is used to cache the received fluorescent digital signal data, zero-axis offset values, and intermediate results during the calculation process. For example, the data processing unit may be implemented using an FPGA (Field-Programmable Gate Array), DSP (Digital Signal Processor), ARM (Reduced Instruction Set Computer Microprocessor), or a high-performance microcontroller. Preferably, the data processing unit may be implemented using an FPGA, which has strong parallel processing capabilities and high real-time performance, making it suitable for processing high-speed acquired fluorescent digital signal data. For example, the FPGA may be a Xilinx ZYNQ series chip, which integrates FPGA logic resources and an ARM processor core, enabling hardware and software co-processing and providing both high-speed data acquisition and real-time signal processing capabilities. For example, the storage unit may be RAM (Random Access Memory) or an internal cache for temporarily storing fluorescent digital signal data and zero-axis offset values.

[0080] The implementation of the one-to-one correspondence between front and rear wheel sampling points can be exemplarily adopted as follows: The data processing unit maintains an independent processing channel and storage space for each data sampling point. When the current wheel data for a certain data sampling point arrives, the data processing unit first identifies the identifier of the data sampling point (such as the NV axis number and left / right side identifier), reads the zero axis offset value calculated in the previous round from the storage unit, and uses it to perform subtraction correction on the current round data; at the same time, the data processing unit performs a mean operation on the current round data, calculates the updated zero axis offset value, and writes it to the storage unit, overwriting or updating the historical zero axis offset value of the sampling point for use in the next round. For example, the storage unit can be divided into multiple storage areas, each corresponding to a data sampling point, used to store the latest calculated zero axis offset value for that sampling point.

[0081] Regarding the implementation of the averaging function, the following method can be used as an example: the data processing unit accumulates and sums multiple fluorescent digital signal data points acquired from the same sampling point on the same side of the same sampling axis in the previous sampling, and then divides by the number of data points to obtain the average result, i.e., the zero-axis offset value. For example, the averaging operation can be implemented through hardware logic circuits such as accumulators and dividers, or it can be executed in a microprocessor through a software program. Preferably, a pipelined accumulator can be used in an FPGA to implement high-speed averaging to improve processing efficiency. For example, the data processing unit can determine the duration of the data participating in the averaging operation according to preset parameters, and preferably, the duration of the data can be set to an integer number of microwave modulation cycles.

[0082] The implementation of the zero-axis offset value storage function can be exemplarily described as follows: the storage unit allocates an independent storage address or storage area for each data sampling point. For example, in a scenario where four NV axes are selected, resulting in a total of eight data sampling points, the storage unit can divide the space into eight independent storage areas, corresponding to the eight data sampling points 1L, 1R, 2L, 2R, 3L, 3R, 4L, and 4R, respectively. After the data processing unit completes the mean calculation for a certain data sampling point, it writes the zero-axis offset value into the corresponding storage area; when the next round of data for that data sampling point arrives, it reads the zero-axis offset value from the corresponding storage area for difference correction.

[0083] Regarding the implementation of the difference correction function, the following method can be used as an example: The data processing unit subtracts the zero-axis offset value read from the storage unit from each fluorescence digital signal data point acquired at the same sampling point on the same side of the current sampling axis to obtain the preprocessed fluorescence digital signal data for the current sampling. For example, the difference operation can be implemented by a subtractor circuit or executed by a software program. Preferably, the data processing unit can perform point-by-point difference processing, that is, perform the difference operation and output immediately after receiving each data point, without waiting for the entire sampling cycle to complete data acquisition, thus ensuring real-time performance.

[0084] Regarding the timing and update mechanism of the zero-axis offset value, for example, the zero-axis offset value of the current data can be calculated and stored in parallel while the current data is used for difference correction, so that it can be used by the next data sampling point on the same side of the same sampling axis. Specifically, when the current round of data for a certain data sampling point arrives, the data processing unit first reads the zero-axis offset value calculated in the previous round from the storage unit, performs difference correction on the current round of data, and outputs the preprocessed fluorescent digital signal data; at the same time, the data processing unit performs mean calculation on the current round of data, calculates the updated zero-axis offset value, and writes it into the storage unit. This parallel processing method can realize the dynamic update of the zero-axis offset value without blocking the data processing flow, ensuring the real-time performance of the measurement.

[0085] For example, the signal preprocessing module may further include a transient interference data skipping unit. This unit is used to identify and skip transient interference data generated by microwave frequency switching. For example, the transient interference data skipping unit can be implemented using a time threshold judgment method: the data processing unit records the microwave switching time point, determines the starting time point of valid data according to a preset time threshold, and performs mean calculation from this starting time point, thereby skipping transient interference data. For example, the transient interference data skipping unit can also be implemented using a variance or standard deviation judgment method: the data processing unit calculates the signal variance or standard deviation within a sliding time window, and when the variance or standard deviation exceeds a preset threshold, it is determined to be transient interference data; when the variance or standard deviation drops below the preset threshold and remains stable, it is determined that the transient interference data has ended.

[0086] For example, the signal preprocessing module may further include an offset value optimization unit. This unit is used to improve the accuracy of the zero-axis offset value. For example, the offset value optimization unit may be implemented using a sliding window update method: the data processing unit continuously updates the zero-axis offset value of each data sampling point as data acquisition progresses. For example, the offset value optimization unit may also be implemented using a median calculation method: the data points used for mean calculation are sorted and the median value is taken as the zero-axis offset value to improve the ability to resist outliers.

[0087] The interconnection and collaboration between the components are exemplarily as follows: The signal preprocessing module receives the fluorescence digital signal data output by the signal sampling module through a data interface, and the data is first stored in the storage unit for buffering; the data processing unit reads the zero-axis offset value of the previous round of the sampling point from the storage unit according to the identifier of the current data sampling point, calculates the difference between the current round data and the zero-axis offset value to obtain the preprocessed fluorescence digital signal data, and outputs it to the noise reduction and demodulation module; at the same time, the data processing unit uses the current round data to calculate the updated zero-axis offset value and writes it into the storage unit. Of course, the connection method and data processing flow of the above components can be flexibly adjusted according to the actual system architecture.

[0088] Of course, the specific composition of the signal preprocessing module is not limited to the example above. Other component combinations that can realize one-to-one correspondence between front and rear wheel sampling points, mean calculation and difference correction functions can also be applied to this device.

[0089] In this example, the noise reduction and demodulation module is used to perform real-time filtering and demodulation on the acquired preprocessed fluorescent digital signal data. The noise reduction and demodulation module is an integral part of this device for signal filtering and demodulation processing. Its function is to filter the preprocessed fluorescent digital signal data to suppress noise and demodulate it to extract the effective signal.

[0090] For example, the noise reduction and demodulation module may include a filtering unit and a demodulation unit as necessary components. The filtering unit is used to filter the preprocessed fluorescent digital signal data to suppress noise components; the demodulation unit is used to demodulate the filtered signal to extract the response components related to the modulated signal.

[0091] Regarding the implementation of the filtering unit, a digital filter can be used, as an example. The digital filter can be, for example, a low-pass filter, a band-pass filter, a high-pass filter, or other types of filters. Preferably, a low-pass filter can be used to filter out high-frequency noise. For example, the cutoff frequency of the low-pass filter can be set to several kilohertz to tens of kilohertz to suppress high-frequency noise components. For example, the digital filter can be implemented using an FIR (Finite Impulse Response) filter or an IIR (Infinite Impulse Response) filter. Preferably, an FIR filter can be used in an FPGA or DSP, which has linear phase characteristics and low signal distortion. For example, the filtering unit may also include a demodulation filter to filter out high-frequency harmonic components and noise generated during demodulation, retaining only the DC component.

[0092] The demodulation unit can be implemented, for example, using a digital lock-in amplifier (LLP). The LLP demodulates the signal by multiplying the input signal and the reference signal, then using a low-pass filter to extract the component with the same frequency and phase as the reference signal. For example, the LLP can be implemented using an FPGA, with the reference signal generated by the data processing unit based on the modulation signal parameters. Alternatively, the demodulation unit can be implemented using correlation demodulation, square-law demodulation, or other demodulation methods.

[0093] For example, the noise reduction and demodulation module can use integrated chips such as the ZYNQ-7010 as core components. These chips integrate FPGA and ARM processor cores, and are equipped with high-sampling-rate ADCs and DACs, enabling high-speed sampling and output of analog and digital signals, as well as hardware and software collaborative processing. For example, such systems can achieve a sampling rate of up to 125 MS / s and a data transmission bandwidth of hundreds of MiB / s, featuring high integration, ease of operation, and high phase-locked loop accuracy.

[0094] For example, the noise reduction and demodulation module may further include a composite filtering unit. This unit is used to further improve the signal-to-noise ratio during the digital post-processing stage. For example, the composite filtering unit may employ wavelet transform, adaptive filtering, or other methods to further process the demodulated signal. For example, the composite filtering unit may automatically select filtering parameters based on signal characteristics to improve the filtering effect.

[0095] For example, the noise reduction and demodulation module may further include a signal output unit. This unit is used to output the demodulated DC voltage signal for subsequent magnetic field inversion calculations. For example, the signal output unit may use a DAC (digital-to-analog converter) to convert the digital signal into an analog voltage signal for output, or it may directly output a digital signal for use by a subsequent digital processing system.

[0096] The interconnection and collaboration between the components are exemplarily as follows: The noise reduction and demodulation module receives preprocessed fluorescent digital signal data output from the signal preprocessing module via a data interface. The filtering unit first filters the data to suppress noise. The filtered data is then input to the demodulation unit for demodulation. The demodulated DC signal is output through the signal output unit. For example, the filtering and demodulation units can be processed sequentially, i.e., filtering followed by demodulation; other processing sequences can also be used depending on the actual solution. Of course, the connection methods and processing flows of the above components can be flexibly adjusted according to the actual system architecture.

[0097] Of course, the specific composition of the noise reduction and demodulation module is not limited to the example above. Other combinations of components that can achieve signal filtering and demodulation functions can also be applied to this device. Example 3

[0098] This example introduces a diamond NV color center quantum device based on ODMR technology, as shown in the attached image. Figure 6 As shown, it includes a diamond NV color center 1, a laser module 2, a microwave module 3, a photoelectric detection module 4, a processor 5, several PID modules 6, and a fluorescence data noise reduction and demodulation device 7.

[0099] In this example, the diamond NV color center 1 is a diamond block containing an ensemble NV color center; the laser module 2 is used to transmit excitation light to the diamond NV color center; the microwave module 3 is used to radiate modulated microwaves to the diamond NV color center; the photoelectric detection module 4 is used to collect the fluorescence generated by the diamond NV color center 1 under the dual action of excitation light and modulated microwave; the processor 5 is used for system control and data processing; several PID modules 6 are used to track two pairs of spin resonance frequencies along at least one NV axis in the diamond NV color center 1 and sequentially provide frequency adjustment signals, the microwave module 3 sequentially receives the frequency adjustment signals and switches the carrier frequency to the corresponding spin resonance frequency; the fluorescence data noise reduction and demodulation device 7 is a device with the same function as mentioned in the previous embodiment.

[0100] It is important to note that in this embodiment, the data sampling point is selected as the spin resonance frequency, i.e., the zero-crossing point of the ODMR curve. This is because in diamond NV center quantum devices based on ODMR technology, the PID module needs to accurately track the spin resonance frequency to achieve accurate measurement of the magnetic field. The spin resonance frequency is a characteristic frequency that the PID module can easily track; at this frequency, the fluorescence signal changes significantly, facilitating frequency locking. Of course, in other embodiments, those skilled in the art can also set the data sampling point to other types of magnetic characterization frequencies, such as the linear region frequency of the ODMR curve or the frequency of the maximum slope, according to actual needs. These changes should all be considered equivalent implementations of the present invention.

[0101] The diamond NV color center 1 is the core sensing element of this device, and is a diamond block containing an ensemble of NV color centers. Exemplarily, the diamond NV color center 1 can be prepared by chemical vapor deposition (CVD) or high-temperature high-pressure (HPHT). Exemplarily, the size of the diamond block can be in the range of approximately 200 μm to 500 μm, preferably in the range of approximately 300 μm to 400 μm. Exemplarily, the NV concentration in the diamond NV color center 1 can be in the range of approximately 100 ppb to 1000 ppb (ppb is one part per billion), preferably in the range of approximately 100 ppb to 500 ppb, to achieve higher measurement sensitivity. Exemplarily, the crystal orientation of the diamond can be selected as {100} or {111}, preferably, in specific application scenarios where the magnetic field direction is fixed, the {100} or {111} crystal orientation can be selected to facilitate the angle design between the magnetic field direction and the NV axis. For example, the longitudinal relaxation time T1 of the diamond NV color center 1 is approximately on the order of several milliseconds, and the transverse relaxation time T2 is approximately on the order of tens to hundreds of microseconds. Of course, the specific parameters of the diamond NV color center 1 can be flexibly selected according to the actual application requirements.

[0102] Laser module 2 is used to transmit excitation light to the diamond NV center 1 to achieve spin polarization and fluorescence excitation of the NV center's electrons. For example, laser module 2 can use a green laser with a wavelength of approximately 532 nm, because the NV center has a strong absorption efficiency for light at a wavelength of 532 nm, enabling efficient spin polarization. For example, the power stability of the laser can be controlled within ±1% to reduce measurement errors caused by laser power fluctuations. For example, laser module 2 may also include optical components such as a focusing lens and a polarizing lens to improve beam quality and excitation efficiency. For example, the laser can be transmitted to the diamond NV center 1 via spatial light transmission or fiber optic transmission. If fiber optic transmission is used, laser module 2 may also include devices such as fiber optic couplers. For example, the diamond NV center 1 can be fixed to the end of the fiber optic cable or set up independently. Of course, the specific composition and optical path structure of laser module 2 can be flexibly designed according to actual application requirements.

[0103] Microwave module 3 is used to radiate modulated microwaves to the diamond NV center 1 to achieve manipulation and resonant excitation of the NV center's electron spin. Exemplarily, microwave module 3 may include components such as a microwave source, a power amplifier, a microwave switch, and a microwave antenna. Exemplarily, the output frequency range of the microwave source may cover approximately 2.0 GHz to 3.5 GHz, preferably approximately 2.7 GHz to 3.1 GHz, to cover the zero-field splitting frequency (approximately 2.87 GHz) and the linear region of the ODMR spectrum of the NV center. Exemplarily, the microwave source may support frequency modulation (FM) or amplitude modulation (AM), and the modulation frequency may exemplary be in the range of hundreds of hertz to thousands of hertz, preferably set around approximately 1 kHz. Exemplarily, the power amplifier is used to amplify the power of the microwave signal to meet the microwave field strength required for spin manipulation of the NV center. Exemplarily, the microwave switch is used to control the on / off state of the microwave signal to achieve pulsed microwave output. For example, the microwave antenna can be a microstrip antenna, a loop copper wire, or other types of microwave transmission structures. Preferably, a microstrip antenna can be used, which has the characteristics of compact structure and easy integration. Of course, the specific composition and structure of the microwave module 3 can be flexibly designed according to the actual application requirements.

[0104] The photodetector module 4 is used to collect the fluorescence generated by the diamond NV color center 1 under the dual action of excitation light and modulated microwave. Exemplarily, the photodetector module 4 may include a photodetector and an optical path collection structure. Exemplarily, the photodetector may be a photomultiplier tube (PMT), an avalanche photodiode (APD), a multi-pixel photon counter (MPPC / SiPM), or a photodiode (PD). Preferably, for the fluorescence signal emitted by the NV color center (wavelength range of approximately 600 nm to 850 nm), a photodetector with high quantum efficiency corresponding to this wavelength range can be used. Exemplarily, a photomultiplier tube has a gain of up to 10^6 to 10^7 times, suitable for detecting extremely weak fluorescence signals; an avalanche photodiode has a fast response speed (nanosecond level), suitable for scenarios requiring fast time-resolved measurements. Exemplarily, the optical path collection structure may include optical elements such as filters, lenses, and mirrors. Exemplarily, a filter is used to filter stray light (such as scattered light from the excitation light), allowing only light in the fluorescence band of the NV color center to pass through, thereby improving the signal-to-noise ratio. For example, lenses or mirrors can be used to improve fluorescence collection efficiency. Of course, the specific composition and optical path structure of the photodetector module 4 can be flexibly designed according to the actual application requirements.

[0105] Processor 5 is used for system control and data processing. Exemplarily, processor 5 can be implemented using an FPGA (Field-Programmable Gate Array), DSP (Digital Signal Processor), ARM (Reduced Instruction Set Computing Microprocessor), or a high-performance microcontroller. Exemplarily, the functions of processor 5 may include: control of laser module 2 (e.g., laser switching, power adjustment), control of microwave module 3 (e.g., frequency setting, modulation parameter setting), control of photoelectric detection module 4 (e.g., sampling parameter setting), control of PID module 6 (e.g., tracking parameter setting), as well as data acquisition, processing, storage, and communication. Exemplarily, processor 5 can run an embedded operating system or bare-metal program to achieve coordinated control and data flow management of the various modules. Exemplarily, processor 5 can interact with a host computer or external system via communication interfaces such as USB, serial port, or Ethernet. Of course, the specific type and functional configuration of processor 5 can be flexibly selected according to actual application requirements.

[0106] Several PID modules 6 are used to track two pairs of spin resonant frequencies along at least one NV axis in the diamond NV color center 1 and sequentially provide frequency adjustment signals. Exemplarily, the number of PID modules 6 depends on the number of NV axes to be tracked and the number of pairs of spin resonant frequencies. For example, in applications where only two spin resonant frequencies along one NV axis need to be tracked, one or two PID modules 6 can be used; in applications such as vector magnetic field measurement where eight spin resonant frequencies along all four NV axes need to be tracked, eight PID modules 6 can be used. Exemplarily, the PID modules 6 can be implemented in hardware, such as using analog circuits or digital hardware circuits (FPGA internal logic) to implement proportional-integral-derivative operations; exemplaryly, the PID modules 6 can also be implemented in software, such as running a PID control algorithm program in the processor 5. Preferably, FPGA hardware implementation can provide faster response speed and higher control accuracy. For example, the working process of PID module 6 is as follows: For a certain NV axis, there are two pairs of spin resonance frequencies (corresponding to the two energy levels m_s=+1 and m_s=-1). PID module 6 continuously adjusts the carrier frequency of the microwave source according to the feedback of the fluorescence signal, so that it locks on the two spin resonance frequencies in turn, thereby realizing the real-time continuous measurement of the dynamically changing magnetic field.

[0107] The fluorescence data noise reduction and demodulation device 7 is the same device as mentioned in the previous embodiments, comprising a signal sampling module, a signal preprocessing module, and a noise reduction and demodulation module. The signal sampling module samples the fluorescence digital signal data output by the photodetector module 4, and is used to perform dual-point alternating sampling of paired spin resonant frequencies along each NV axis. The signal preprocessing module is used to perform zero-axis offset correction using a one-to-one correspondence between the front and rear wheel sampling points. The noise reduction and demodulation module is used to perform real-time filtering and demodulation of the preprocessed data. The specific composition and working principle of the fluorescence data noise reduction and demodulation device 7 are as described in the previous embodiments and will not be repeated here.

[0108] Overall Collaboration: For example, the collaborative relationships among the components of this device are as follows: Laser module 2 transmits excitation light to the diamond NV center 1, achieving spin polarization of the NV center electrons; Microwave module 3, under the control of PID module 6, radiates modulated microwaves to the diamond NV center 1, with the carrier frequency switching between paired spin resonant frequencies; The diamond NV center 1 generates fluorescence under the combined action of the excitation light and modulated microwaves; Photodetector module 4 collects the fluorescence signal and converts it into an electrical signal; Fluorescence data noise reduction and demodulation device 7 performs alternating dual-point sampling, zero-axis offset correction, filtering, and demodulation on the electrical signal, outputting a demodulated signal; Processor 5 is responsible for the overall coordination control and data processing of the system. Of course, the above collaborative relationships can be flexibly adjusted according to the actual application scenario.

[0109] Based on the above embodiments, the device may further include a magnetizer, by way of example. The magnetizer is used to concentrate the magnetic field and form a focused magnetic field parallel to the first NV axis of the diamond NV center. In this scheme, the PID module is used to track two pairs of spin resonant frequencies along the first NV axis of the diamond NV center and sequentially provide frequency adjustment signals.

[0110] The function of the magnetic concentrator is to focus and guide an external magnetic field in a specific direction, making the magnetic field direction parallel to the first NV axis. For example, the magnetic concentrator can be made of a highly permeable material (such as permalloy, ferrite, etc.), and its shape can be designed according to actual application requirements, such as conical, cylindrical, or other structures capable of focusing the magnetic field. For example, the magnetic concentrator can be placed around or near the diamond NV color center 1, forming a closed or semi-closed magnetic circuit structure.

[0111] When the external magnetic field direction is parallel to the first NV axis, the magnetic field projection sensed along the first NV axis is the largest (equal to the total magnetic field), resulting in the largest splitting of its corresponding ODMR resonance peak and the highest ODMR curve contrast. Conversely, the other three NV axes, due to their angle with the magnetic field direction, experience smaller magnetic field projections, leading to relatively weaker ODMR signals. This configuration ensures the highest magnetic field measurement sensitivity and best signal quality along the first NV axis, thus improving measurement accuracy and signal-to-noise ratio. For example, this scheme is suitable for applications requiring high-precision measurement of magnetic field components in specific directions, such as geomagnetic navigation, magnetic field anomaly detection, and uniaxial magnetic field measurement.

[0112] Based on the above embodiments, by way of example, the signal duration of the fluorescent digital signal data at each data sampling point is equal to a filtering duration.

[0113] This embodiment provides a preferred solution under extreme conditions, aiming to maximize measurement speed. Specifically, the signal duration of the fluorescent digital signal data at each data sampling point is equal to a filtering duration. This is the minimum data length requirement for the filtering algorithm to work effectively. If the data duration is further shortened, the filtering effect will decrease significantly, thereby affecting the quality of the demodulated signal.

[0114] In this extreme case, the total duration of the fluorescence digital signal data at each data sampling point is equal to one filtering duration, which is the shortest data length for effective filtering and demodulation. Compared to using longer filtering durations, this scheme can significantly shorten the data processing time for a single sampling point, thereby effectively improving measurement speed.

[0115] This scheme does not impose any particular limitation on the signal duration of the fluorescent digital signal data used for averaging. Since the filtering duration is usually longer than the microwave modulation period, the data used for averaging can be flexibly selected from the data of the same sampling points in the previous round. For example, any segment of the data in the previous round can be selected for averaging, and the data duration can be one or more microwave modulation periods. Preferably, using data from multiple modulation periods for averaging can reduce random errors and improve the accuracy of the zero-axis offset value.

[0116] For example, this extreme solution is suitable for applications with extremely high real-time requirements, such as AC current tracking and other scenarios where the external magnetic field changes rapidly. In these scenarios, the magnetic field changes rapidly, requiring the measurement system to have an extremely high response speed. In this case, using the shortest data length solution can effectively capture rapidly changing magnetic field signals. Of course, while this solution improves measurement speed, if the data duration used for averaging is short, the calculation accuracy of the zero-axis offset value may be slightly lower. In practical applications, a trade-off between measurement speed and accuracy can be struck according to specific requirements. Example 4

[0117] Corresponding to the fluorescence data denoising and demodulation methods above, see attached... Figure 7 As shown, one embodiment of this application also proposes a storage medium 100, which can be a non-transitory computer-readable storage medium. Exemplarily, the storage medium 100 may include a hard disk, optical disk, USB flash drive, flash memory, magnetic storage, or other forms of non-transitory computer-readable storage medium. A computer program 101 is stored on the storage medium 100. When executed by data processing hardware, the computer program 101 is capable of implementing the operation of the fluorescence data denoising and demodulation method described above. The computer program 101 contains one or more computer-readable instructions that, when executed by a processor, can perform one or more steps according to the fluorescence data denoising and demodulation method described above.

[0118] The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0119] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0121] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0122] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method of fluorescence data denoising demodulation, applied to a process based on diamond NV color centers and using optically detected magnetic resonance for sensing, characterized in that, Include: S1. Select several NV axes in the diamond NV color center as sampling axes, and let two pairs of magnetic characterization frequencies of each sampling axis be used as data sampling points, and record them as the left sampling point and the right sampling point of the sampling axis. The sampling object is the fluorescence digital signal data at each data sampling point. During sampling, the two-point alternating sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point. S2. Preprocessing the fluorescent digital signal data, the preprocessing includes: -Based on the fluorescence digital signal data obtained from the same side data sampling point of the same sampling axis in the previous sampling, the mean value is calculated to obtain the zero axis offset value and stored; - Perform a subtraction operation between the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the current sampling and the zero axis offset value obtained from the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the previous sampling, and obtain the preprocessed fluorescence digital signal data of the current sampling. S3. Perform real-time filtering and demodulation on the obtained preprocessed current subfluorescence digital signal data.

2. The fluorescence data denoising demodulation method of claim 1, wherein, For each data sampling point, the signal duration of the fluorescent digital signal data used for averaging is an integer number of microwave modulation cycles.

3. The fluorescence data denoising demodulation method of claim 1, wherein, The signal duration of the obtained preprocessed fluorescent digital signal data is not less than the filtering duration.

4. The fluorescence data noise reduction and demodulation method according to claim 1, characterized in that, The fluorescent digital signal data used for averaging does not include microwave switching transient interference data.

5. A fluorescence data noise reduction and demodulation device, applied to a process based on diamond NV color centers and using photodetector magnetic resonance sensing, characterized in that... Include: Signal sampling module: Select several NV axes in the diamond NV center as sampling axes, and let two pairs of magnetic characterization frequencies of each sampling axis be used as data sampling points, and record them as the left sampling point and the right sampling point of the sampling axis. The sampling object is the fluorescence digital signal data at each data sampling point. During sampling, the two-point alternating sampling is performed on each sampling axis in turn to obtain the fluorescence digital signal data at each data sampling point. Signal preprocessing module: performs preprocessing on the fluorescence digital signal data, the preprocessing including: -Based on the fluorescence digital signal data obtained from the same side data sampling point of the same sampling axis in the previous sampling, the mean value is calculated to obtain the zero axis offset value and stored; - Perform a subtraction operation between the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the current sampling and the zero axis offset value obtained from the fluorescence digital signal data obtained from the data sampling point on the same side of the same sampling axis in the previous sampling, and obtain the preprocessed fluorescence digital signal data of the current sampling. Noise reduction and demodulation module: performs real-time filtering and demodulation on the preprocessed current subfluorescent digital signal data.

6. The fluorescence data noise reduction and demodulation device according to claim 5, characterized in that, In the signal preprocessing module, for each data sampling point, the signal duration of the fluorescent digital signal data used for averaging is an integer number of microwave modulation cycles.

7. The fluorescence data noise reduction and demodulation device according to claim 5, characterized in that, In the signal preprocessing module, the signal duration of the preprocessed fluorescent digital signal data is not less than the filtering duration.

8. The fluorescence data noise reduction and demodulation device according to claim 5, characterized in that, In the signal preprocessing module, the fluorescent digital signal data used for averaging does not include microwave switching transient interference data.

9. A quantum sensing device, comprising: Diamond NV color centers are diamond blocks containing ensemble NV color centers. The laser module is used to transmit excitation light to the diamond NV color center; A microwave module is used to radiate modulated microwaves to the diamond NV color center; The photoelectric detection module is used to collect the fluorescence generated by the diamond NV color center under the dual action of excitation light and modulated microwave; The quantum sensing device, as described above, is characterized in that it further comprises: The PID module is used to track two pairs of spin resonant frequencies along at least one NV axis in the diamond NV center and give frequency adjustment signals in turn. The microwave module receives the frequency adjustment signals in turn and switches the carrier frequency to the corresponding spin resonant frequency. The fluorescence data noise reduction and demodulation device according to any one of claims 5-8, wherein the sampling object of the signal sampling module is the fluorescence digital signal data output by the photoelectric detection module.

10. The quantum sensing device according to claim 9, characterized in that, It also includes a magnetic concentrator for focusing the magnetic field and forming a magnetic field parallel to the first NV axis of the diamond NV center; and a PID module for tracking two pairs of spin resonant frequencies along the first NV axis of the diamond NV center and sequentially providing frequency adjustment signals.

11. The quantum sensing device according to claim 9, characterized in that, The signal duration of the fluorescent digital signal data at each data sampling point is equal to the filtering duration.

12. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by data processing hardware, enables the operation of the fluorescence data noise reduction and demodulation method as described in any one of claims 1-4.

Images