阅读说明：本技术 一种适用于质子磁旋进信号的噪声抑制系统及方法 (Noise suppression system and method suitable for proton magnetic precession signal ) 是由 刘欢 董浩斌 朱俊 袁志文 葛健 张艳丽 于 2019-08-29 设计创作，主要内容包括：本发明提供了一种适用于质子磁旋进信号的噪声抑制系统及方法,动态核极化传感器接收激励电路发出的激励信号,感应出质子磁旋进信号,并传递给调谐电路,调谐电路对信号进行调谐后,传递给放大电路,放大电路对信号进行放大后,窄带滤波器对信号进行带通滤波去除中心频带外的噪声后,将其传输给比较电路,比较电路对窄带滤波器处理后的信号进行整形成方波后,将其传递给FPGA进行后期信号处理,采集器与控制器连接,进行数据的采集与处理,控制器与激励电路连接,控制激励信号的开启与关闭,控制器与FPGA连接,FPGA对整形成方波后的质子磁旋进信号进行频率测量,并将所测量的频率值转化为磁场值,控制器与窄带滤波器相连接,调整其窄带中心频率。(The invention provides a noise suppression system and a method suitable for proton magnetic precession signals, wherein a dynamic nuclear polarization sensor receives an excitation signal sent by an excitation circuit, senses the proton magnetic precession signal and transmits the proton magnetic precession signal to a tuning circuit, the tuning circuit tunes the signal and transmits the signal to an amplifying circuit, the amplifying circuit amplifies the signal, a narrow-band filter band-pass filters the signal to remove noise outside a central frequency band and transmits the noise to a comparison circuit, the comparison circuit shapes the signal processed by the narrow-band filter into a square wave and transmits the square wave to an FPGA for post-period signal processing, a collector is connected with a controller for collecting and processing data, the controller is connected with the excitation circuit to control the on and off of the excitation signal, the controller is connected with the FPGA, and the FPGA carries out frequency measurement on the proton magnetic precession signal after shaping into the square wave, and the measured frequency value is converted into a magnetic field value, and the controller is connected with the narrow-band filter to adjust the narrow-band center frequency of the narrow-band filter.)

1. A noise suppression system suitable for proton magnetic precession signals is characterized by comprising a dynamic nuclear polarization weak magnetic sensor, an excitation circuit, a tuning circuit, an amplification circuit, a narrow band filter circuit, a comparison circuit, a collector, an FPGA and a controller, wherein the dynamic nuclear polarization sensor receives the excitation signal sent by the excitation circuit, senses the proton magnetic precession signal and transmits the proton magnetic precession signal to the tuning circuit, the tuning circuit tunes the signal and transmits the signal to the amplification circuit, the amplification circuit amplifies the signal and transmits the signal to the narrow band filter circuit, the comparison circuit and the collector respectively, the narrow band filter band-pass filters the signal to remove noise outside a central frequency band and transmits the signal to the comparison circuit, the comparison circuit shapes the signal processed by the narrow band filter into square waves and transmits the square waves to the FPGA for later-stage signal processing, and the collector is connected with the controller, the controller is connected with the excitation circuit and used for controlling the excitation signal to be turned on and off, the controller is connected with the FPGA, the FPGA carries out frequency measurement on the proton magnetic precession signal shaped into the square wave and converts the measured frequency value into a magnetic field value, and the controller is connected with the narrow-band filter and used for adjusting the narrow-band center frequency of the narrow-band filter.

2. The noise suppression system suitable for the proton magnetic precession signal according to claim 1, wherein the controller is configured to control the collector to collect environmental noise and instrument background noise when the excitation circuit is not driven, and generate the first discrete data; constructing a first space matrix according to the first discrete data, and solving the covariance and the corresponding weight of the first space matrix by adopting a Cholesky algorithm; the controller is also used for driving the excitation circuit to excite the sensor to output proton magnetic precession signals, waiting for a preset time after excitation is completed, and driving the collector to collect the proton magnetic precession signals to generate second discrete data; constructing a second spatial matrix according to the second discrete data, performing inverse transformation on the first spatial matrix, and performing multiplication operation on the first spatial matrix and the second spatial matrix to obtain a third spatial matrix; and performing singular value decomposition on the third spatial matrix by using an SVD algorithm to further remove noise and obtain a reconstructed proton magnetic precession signal.

3. The system according to claim 1, wherein the tuning circuit is configured to switch a capacitance value of a tuning capacitor connected in parallel with the sensor from zero to a theoretical tuning capacitance value corresponding to a frequency of the proton magnetic precession signal in a measurement environment under the driving of the controller; the amplifying circuit is used for amplifying environmental noise and instrument background noise when the sensor is not excited and is used for acquisition by the acquisition device; the device is also used for waiting for a preset time after the sensor is excited, amplifying the proton magnetic precession signal and collecting the proton magnetic precession signal by the collector; the narrow-band filter circuit is used for filtering the amplified proton magnetic precession signal by taking the frequency corresponding to the tuning capacitance value as a central frequency and shaping the proton magnetic precession signal by using the comparison circuit; and the comparison circuit is used for shaping the filtered reconstructed proton magnetic precession signal into a square wave, and is used for the FPGA to measure the frequency and convert the measured frequency value into a magnetic field value.

4. The noise suppression system suitable for the proton magnetic precession signal according to claim 1, wherein the collector is configured to collect, under driving of the controller, environmental noise and instrument background noise when the excitation circuit is not driven, and generate the first discrete data; the proton magnetic precession signal is also used for collecting the proton magnetic precession signal after the driving excitation circuit is driven by the controller to generate second discrete data.

5. The system of claim 1, wherein the excitation circuit is configured to excite the sensor to output the proton magnetic precession signal under the driving of the controller; the FPGA is used for measuring the frequency value of the reconstructed proton magnetic precession signal under the driving of the controller and converting the measured frequency value into a magnetic field value.

6. A noise suppression method suitable for proton magnetic precession signals is realized based on a noise suppression system suitable for proton magnetic precession signals, and is characterized by comprising the following steps:

s1, not exciting the sensor, and outputting noise signals mixed with background interference and instrument background interference;

s2, collecting the noise signal by using a collector to generate first discrete data;

s3, constructing a first spatial matrix m according to the first discrete data, and solving the covariance and the corresponding weight of the first spatial matrix m by adopting a Cholesky algorithm;

s4, exciting the sensor to output a proton magnetic precession signal, waiting for a preset time after excitation is finished, facilitating elimination of interference generated by circuit oscillation, and acquiring the proton magnetic precession signal by using an acquisition device to generate second discrete data;

s5, constructing a second spatial matrix y ═ y from the second discrete data₁,y₂,…,y_n]Performing inverse transformation on the first spatial matrix M, and performing multiplication with the transpose of the second spatial matrix y to obtain a third spatial matrix M: m ═ L^-1·y^TAt this time, M is the pre-whitened proton magnetic precession signal;

s6, performing singular value decomposition on the third spatial matrix M by adopting an SVD algorithm to further eliminate noise in a signal corresponding to the third spatial matrix M to obtain reconstruction data, wherein the reconstruction data is a proton magnetic precession signal after noise removal, inputting the reconstruction data into a narrow-band filter through a controller to further suppress interference outside a central frequency band to achieve the purpose of noise suppression, then transmitting the narrow-band filtered signal to a comparison circuit to be converted into a square wave, measuring the frequency of the square wave signal by using an FPGA, and converting the measured frequency value into a magnetic field value.

7. The method according to claim 6, wherein the step S3 specifically includes:

s31, amplifying the noise signal generated by the non-excited sensor by the amplifying circuit, and recording the amplified noise signal by the collector to obtain a first discrete data x ═ x₁,x₂,…,x_n]；

S32, constructing a new first space matrix m:

s33, solving a lower triangular matrix L with a diagonal element as a positive number by adopting a Cholesky algorithm so as to meet the condition: m ═ L^TL。

8. The method according to claim 6, wherein the step S6 specifically includes:

s61, according to the third space matrix M ═ M₁,M₂,…,M_n]And constructing a new space matrix A:

s62, making SVD to the above formula, then there is A ═ USV^T，

Where U is an orthogonal matrix of order m × m, S is a diagonal matrix of order m × n, V is an orthogonal matrix of order n × n, and A is decomposed:

in the formula, A₁A matrix formed for time sequence in the absence of noise interference, A₂Obtaining proton magnetic precession signal matrix A formed by time sequence after noise reduction for matrix formed by noise₁＝U₁S₁V₁ ^TThe signals are further input into a program-controlled narrow-band filter to further filter noise signals outside the central frequency band of the proton magnetic precession signals, then the signals after narrow-band filtering are transmitted to a comparison circuit to be converted into square waves, the frequency of the square wave signals is measured by using an FPGA, and the measured frequency value is converted into a magnetic field value.

Technical Field

The invention relates to the technical field of weak magnetic field measurement, in particular to a noise suppression system and method suitable for proton magnetic precession signals.

Background

The proton precession type magnetometer is a magnetic field measuring instrument for measuring slowly changing weak magnetic field or constant weak magnetic field, and its sensor, i.e. proton precession type sensor, is an inductance element. The measurement principle is that a certain excitation condition is utilized to enable protons in a solution in which an inductor is positioned to be in an activated state, and the protons can do Larmor precession motion around a stable external magnetic field, namely an earth magnetic field after the excitation condition is removed to generate a proton magnetic precession signal, wherein the precession frequency of the proton magnetic precession signal is in direct proportion to an external magnetic field; therefore, the proton magnetic precession signal is induced by the inductor, amplified, shaped and measured in frequency to obtain the external magnetic field value. Compared with other magnetic field measurement technologies, the proton precession magnetometer has the characteristics of high precision, high sensitivity and the like, and is widely applied to the fields of space detection, near-surface detection, ocean detection, geomagnetic field measurement, military technologies and the like. Because the signal-to-noise ratio of the proton magnetic precession signal is an important factor for measuring the frequency measurement accuracy, in order to increase the signal-to-noise ratio of the proton magnetic precession signal output by the sensor and improve the frequency measurement accuracy, noise suppression needs to be performed on the signal to a certain extent so as to improve the performance of the instrument.

At present, most proton precession magnetometers mainly adopt proton magnetic precession signal noise suppression means which mainly comprise a narrow-band filtering method (hardware), an autocorrelation method (software) and a singular value decomposition method (software). However, the following problems still exist in the design of hardware circuit or algorithm: 1) a narrow-band filter is added, certain narrow-band noise is introduced, and the noise suppression of signals is not facilitated; 2) the autocorrelation algorithm and the SVD algorithm can have a certain effect when the external environment noise is relatively low, but cannot perform noise reduction work when the background noise is large. Therefore, once the instrument is in a high-interference environment, spectrum confusion is easy to occur to cause signal imbalance, so that the instrument cannot work normally.

Disclosure of Invention

The technical problem to be solved by the present invention is to provide a noise suppression system and method suitable for proton precession signals to solve the above technical defects, aiming at the technical problems existing in the design of the sensor tuning algorithm of the existing proton precession magnetometer.

A noise suppression system suitable for proton magnetic precession signals comprises a dynamic nuclear polarization weak magnetic sensor, an excitation circuit, a tuning circuit, an amplifying circuit, a narrow-band filtering circuit, a comparison circuit, a collector, an FPGA and a controller, wherein the dynamic nuclear polarization sensor receives the excitation signal sent by the excitation circuit, senses the proton magnetic precession signal and transmits the proton magnetic precession signal to the tuning circuit, the tuning circuit tunes the signal and transmits the signal to the amplifying circuit, the amplifying circuit amplifies the signal and transmits the signal to the narrow-band filtering circuit, the comparison circuit and the collector respectively, the narrow-band filter band-pass filters the signal to remove noise outside a central frequency band and transmits the signal to the comparison circuit, the comparison circuit shapes the signal processed by the narrow-band filter into a square wave and transmits the square wave to the FPGA for post-stage signal processing, and the collector is connected with the controller to collect and process data, the controller is connected with the excitation circuit and used for controlling the excitation signal to be turned on and off, the controller is connected with the FPGA, the FPGA carries out frequency measurement on the proton magnetic precession signal shaped into the square wave and converts the measured frequency value into a magnetic field value, and the controller is connected with the narrow-band filter and used for adjusting the narrow-band center frequency of the narrow-band filter.

Further, the controller is used for controlling the collector, collecting the environmental noise and the instrument background noise when the excitation circuit is not driven, and generating first discrete data; constructing a first space matrix according to the first discrete data, and solving the covariance and the corresponding weight of the first space matrix by adopting a Cholesky algorithm; the controller is also used for driving the excitation circuit to excite the sensor to output proton magnetic precession signals, waiting for a preset time after excitation is completed, and driving the collector to collect the proton magnetic precession signals to generate second discrete data; constructing a second spatial matrix according to the second discrete data, performing inverse transformation on the first spatial matrix, and performing multiplication operation on the first spatial matrix and the second spatial matrix to obtain a third spatial matrix; and performing singular value decomposition on the third spatial matrix by using an SVD algorithm to further remove noise and obtain a reconstructed proton magnetic precession signal.

Further, the tuning circuit is used for switching the capacitance value of a tuning capacitor connected with the sensor in parallel from zero to a theoretical tuning capacitance value corresponding to the proton magnetic precession signal frequency in the measurement environment under the driving of the controller; the amplifying circuit is used for amplifying environmental noise and instrument background noise when the sensor is not excited and is used for acquisition by the acquisition device; the device is also used for waiting for a preset time after the sensor is excited, amplifying the proton magnetic precession signal and collecting the proton magnetic precession signal by the collector; the narrow-band filter circuit is used for filtering the amplified proton magnetic precession signal by taking the frequency corresponding to the tuning capacitance value as a central frequency and shaping the proton magnetic precession signal by using the comparison circuit; and the comparison circuit is used for shaping the filtered reconstructed proton magnetic precession signal into a square wave, and is used for the FPGA to measure the frequency and convert the measured frequency value into a magnetic field value.

Further, the collector is used for collecting environmental noise and instrument background noise when the excitation circuit is not driven under the driving of the controller to generate first discrete data; the proton magnetic precession signal is also used for collecting the proton magnetic precession signal after the driving excitation circuit is driven by the controller to generate second discrete data.

Further, the excitation circuit is used for exciting the sensor to output a proton magnetic precession signal under the driving of the controller; the FPGA is used for measuring the frequency value of the reconstructed proton magnetic precession signal under the driving of the controller and converting the measured frequency value into a magnetic field value.

A noise suppression method suitable for proton magnetic precession signals is realized based on a noise suppression system suitable for proton magnetic precession signals, and comprises the following steps:

s1, not exciting the sensor, and outputting noise signals mixed with background interference and instrument background interference;

s2, collecting the noise signal by using a collector to generate first discrete data;

s3, constructing a first spatial matrix m according to the first discrete data, and solving the covariance and the corresponding weight of the first spatial matrix m by adopting a Cholesky algorithm;

s4, exciting the sensor to output a proton magnetic precession signal, waiting for a preset time after excitation is finished, facilitating elimination of interference generated by circuit oscillation, and acquiring the proton magnetic precession signal by using an acquisition device to generate second discrete data;

s5, constructing a second spatial matrix y ═ y from the second discrete data₁,y₂,…,y_n]Performing inverse transformation on the first spatial matrix M, and performing multiplication with the transpose of the second spatial matrix y to obtain a third spatial matrix M: m ═ L^-1·y^TAt this time, M is the pre-whitened proton magnetic precession signal;

s6, performing singular value decomposition on the third spatial matrix M by adopting an SVD algorithm to further eliminate noise in a signal corresponding to the third spatial matrix M to obtain reconstruction data, wherein the reconstruction data is a proton magnetic precession signal after noise removal, inputting the reconstruction data into a narrow-band filter through a controller to further suppress interference outside a central frequency band to achieve the purpose of noise suppression, then transmitting the narrow-band filtered signal to a comparison circuit to be converted into a square wave, measuring the frequency of the square wave signal by using an FPGA, and converting the measured frequency value into a magnetic field value.

Further, step S3 specifically includes:

s31, amplifying the noise signal generated by the non-excited sensor by the amplifying circuit, and recording the amplified noise signal by the collector to obtain a first discrete data x ═ x₁,x₂,…,x_n]；

S32, constructing a new first space matrix m:

s33, solving a lower triangular matrix L with a diagonal element as a positive number by adopting a Cholesky algorithm so as to meet the condition: m ═ L^TL。

Further, step S6 specifically includes:

s61, according to the third space matrix M ═ M₁,M₂,…,M_n]And constructing a new space matrix A:

s62, making SVD to the above formula, then there is A ═ USV^T，

Where U is an orthogonal matrix of order m × m, S is a diagonal matrix of order m × n, V is an orthogonal matrix of order n × n, and A is decomposed:

in the formula, A₁A matrix formed for time sequence in the absence of noise interference, A₂Obtaining proton magnetic precession signal matrix A formed by time sequence after noise reduction for matrix formed by noise₁＝U₁S₁V₁ ^TThe signals are further input into a program-controlled narrow-band filter to further filter noise signals outside the central frequency band of the proton magnetic precession signals, then the signals after narrow-band filtering are transmitted to a comparison circuit to be converted into square waves, the frequency of the square wave signals is measured by using an FPGA, and the measured frequency value is converted into a magnetic field value.

Compared with the prior art, the invention has the beneficial effects that:

1. adopting a mode of combining a Cholesky algorithm and an SVD algorithm to further inhibit unknown noise attached to the proton magnetic precession signal on a software level;

2. and a program-controlled narrow-band filter circuit is adopted to further suppress noise outside a central frequency band of the proton magnetic precession signal on a hardware level.

Drawings

The invention will be further described with reference to the accompanying drawings and examples, in which:

FIG. 1 is a block diagram of a noise suppression system for proton magnetic precession signals according to the present invention;

FIG. 2 is a flow chart of a noise suppression method for a proton magnetic precession signal according to the present invention;

FIG. 3 is a graph showing the comparison of the effects of the embodiment of the present invention.

Detailed Description

For a more clear understanding of the technical features, objects and effects of the present invention, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

A noise suppression system suitable for proton magnetic precession signals comprises a dynamic nuclear polarization weak magnetic sensor, an excitation circuit, a tuning circuit, an amplifying circuit, a narrow band filter circuit, a comparison circuit, a collector, an FPGA and a controller, wherein the dynamic nuclear polarization weak magnetic sensor receives the excitation signal sent by the excitation circuit, induces the proton magnetic precession signal and transmits the proton magnetic precession signal to the tuning circuit, the tuning circuit tunes the signal and transmits the signal to the amplifying circuit, the amplifying circuit amplifies the signal and transmits the signal to the narrow band filter circuit, the comparison circuit and the collector respectively, the narrow band filter band-pass filters the signal and transmits the signal to the comparison circuit, the comparison circuit shapes the signal into square and transmits the square to the FPGA for post-stage signal processing, the collector is connected with the controller for data acquisition and processing, and the controller is connected with the excitation circuit, the controller is connected with the FPGA to measure the frequency of the proton magnetic precession signal and convert the measured frequency value into a magnetic field value, and the controller is connected with the narrow-band filter to adjust the narrow-band center frequency of the narrow-band filter.

The controller is used for controlling the collector, collecting environmental noise and instrument background noise when the excitation circuit is not driven, and generating first discrete data; constructing a first space matrix according to the first discrete data, and solving the covariance and the corresponding weight of the first space matrix by adopting a Cholesky algorithm; the driving circuit is used for driving the excitation sensor to output a proton magnetic precession signal, waiting for a preset time after excitation is finished, and driving the collector to collect the proton magnetic precession signal to generate second discrete data; constructing a second spatial matrix according to the second discrete data, performing inverse transformation on the first spatial matrix, and performing multiplication operation on the first spatial matrix and the second spatial matrix to obtain a third spatial matrix; and performing singular value decomposition on the third spatial matrix by using an SVD algorithm to further remove noise and obtain a reconstructed proton magnetic precession signal.

The tuning circuit is used for switching the capacitance value of a tuning capacitor connected with the sensor in parallel from zero to a theoretical tuning capacitance value corresponding to the proton magnetic precession signal frequency in a measuring environment under the driving of the controller; the amplifying circuit is used for amplifying environmental noise and instrument background noise when the sensor is not excited and is used for acquisition by the acquisition device; the device is also used for waiting for a preset time after the sensor is excited, amplifying the proton magnetic precession signal and collecting the proton magnetic precession signal by the collector; the narrow-band filter circuit is used for filtering the amplified proton magnetic precession signal by taking the frequency corresponding to the tuning capacitance value as a central frequency and shaping the proton magnetic precession signal by using the comparison circuit; and the comparison circuit is used for shaping the filtered reconstructed proton magnetic precession signal into a square wave, and is used for the FPGA to measure the frequency and convert the measured frequency value into a magnetic field value. The collector is used for collecting environmental noise and instrument background noise when the excitation circuit is not driven under the driving of the controller to generate first discrete data; the proton magnetic precession signal is also used for collecting the proton magnetic precession signal after the driving excitation circuit is driven by the controller to generate second discrete data. The excitation circuit is used for exciting the sensor to output a proton magnetic precession signal under the driving of the controller; the FPGA is used for measuring the frequency value of the reconstructed proton magnetic precession signal under the driving of the controller and converting the measured frequency value into a magnetic field value.

A noise suppression method for proton magnetic precession signals is implemented based on a noise suppression system for proton magnetic precession signals, as shown in FIG. 2, and includes:

s1, not exciting the sensor, and outputting noise signals mixed with background interference and instrument background interference;

s2, collecting the noise signal by using a collector to generate first discrete data;

s3, constructing a first spatial matrix m according to the first discrete data, and solving the covariance and the corresponding weight of the first spatial matrix m by adopting a Cholesky algorithm;

s31, amplifying the noise signal generated by the non-excited sensor by the amplifying circuit, and recording the amplified noise signal by the collector to obtain a first discrete data x ═ x₁,x₂,…,x_n]；

S32, constructing a new first space matrix m:

s33, solving a lower triangular matrix L with a diagonal element as a positive number by adopting a Cholesky algorithm so as to meet the condition:

m＝L^T·L。

s4, exciting the sensor to output a proton magnetic precession signal, waiting for a preset time after excitation is finished, facilitating elimination of interference generated by circuit oscillation, and acquiring the proton magnetic precession signal by using an acquisition device to generate second discrete data;

s5, constructing a second spatial matrix y ═ y from the second discrete data₁,y₂,…,y_n]Performing inverse transformation on the first spatial matrix M, and performing multiplication with the transpose of the second spatial matrix y to obtain a third spatial matrix M: m ═ L^-1·y^TAt this time, M is the pre-whitened proton magnetic precession signal;

and S6, performing singular value decomposition on the third space matrix M by adopting an SVD (singular value decomposition) algorithm to further eliminate noise, and obtaining reconstructed data.

S61, according to the third space matrix M ═ M₁,M₂,…,M_n]And constructing a new space matrix A:

s62, when SVD is performed on the formula, the following steps are performed:

A＝USV^T，

in the formula, U is an orthogonal matrix of order m × m, S is a diagonal matrix of order m × n, and V is an orthogonal matrix of order n × n. A can be decomposed into:

in the formula, A₁A matrix formed for time sequence in the absence of noise interference, A₂Is a matrix formed by noise, so that a proton magnetic precession signal matrix A formed by time sequence after noise reduction can be obtained₁The estimated formula of (c) is:

A₁＝U₁S₁V₁ ^T，

and further inputting the signals into a program-controlled narrow-band filter to further filter noise signals outside the central frequency band of the proton magnetic precession signals, transmitting the narrow-band filtered signals to a comparison circuit to be converted into square waves, measuring the frequency of the square wave signals by using an FPGA (field programmable gate array), and converting the measured frequency value into a magnetic field value.

As shown in fig. 3, the spectra are sequentially a spectrum acquired by the collector without being processed by the proton magnetic precession signal amplified by the amplifying circuit, a spectrum processed by the conventional Auto Correlation algorithm (Auto Correlation) of the proton magnetic precession signal, and a spectrum composed of reconstructed data obtained by processing the proton magnetic precession signal by the Cholesky & SVD algorithm of the present invention under the background of strong noise; the comparison shows that the frequency spectrum processed by the Cholesky & SVD algorithm is more vivid and the signal-to-noise ratio is higher.

Compared with the existing tuning algorithm, the method adopts a mode of combining the Cholesky algorithm and the SVD algorithm to further inhibit the unknown noise attached to the proton magnetic precession signal on a software level; and a program-controlled narrow-band filter circuit is adopted to further suppress noise outside a central frequency band of the proton magnetic precession signal on a hardware level.

In conclusion, the invention realizes the improvement of the signal-to-noise ratio of the proton magnetic precession signal by combining the Cholesky algorithm and the SVD algorithm, effectively overcomes the defects of poor noise reduction effect, easy signal distortion and the like of the existing noise suppression algorithm in a strong interference environment, effectively enhances the environmental adaptability of the proton magnetic precession sensor, can improve the precision of measuring the frequency of the proton magnetic precession signal in the later period, is applied to instruments depending on the proton magnetic precession sensor, such as a proton magnetic precession magnetometer, an optical pump magnetometer, a nuclear magnetic resonance proton precession signal imager and the like, and effectively improves the performance of the instruments.

While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.