阅读说明：本技术 在弱测量中对抗振幅型噪声的光谱数据处理方法及系统 (Spectral data processing method and system for resisting amplitude type noise in weak measurement ) 是由 黄靖正 黄朝政 李洪婧 曾贵华 于 2020-07-10 设计创作，主要内容包括：本发明提供了一种在弱测量中对抗振幅型噪声的光谱数据处理方法及系统,包括：噪声特性提取步骤：对测量得到的输出光谱以及输入光谱进行域变换,得到在变换域表述振幅型噪声特性的参数；滤波器设计步骤：根据输入光谱和提取的噪声特性参数,结合处理要求的选择设计滤波器；光谱数据滤波步骤：使用设计好的滤波器对光谱数据进行滤波处理和验证,得到噪声减弱后的光谱；计算弱测量结果步骤：利用噪声减弱后的光谱,通过弱测量系统光谱分析解算得出弱测量对待测参量的估计值；本发明实现对特定振幅型噪声的减弱,一定程度上克服了振幅型噪声对弱测量技术带来的不良影响。(The invention provides a spectral data processing method and a system for resisting amplitude type noise in weak measurement, which comprises the following steps: noise characteristic extraction step: carrying out domain transformation on the measured output spectrum and the input spectrum to obtain parameters for expressing the amplitude type noise characteristics in a transformation domain; a filter design step: according to the input spectrum and the extracted noise characteristic parameters, a filter is selected and designed in combination with the processing requirements; and (3) spectral data filtering: filtering and verifying the spectral data by using a designed filter to obtain a spectrum with reduced noise; and calculating a weak measurement result: calculating to obtain an estimated value of the weak measurement to the parameter to be measured by using the spectrum after the noise is weakened through the spectrum analysis of the weak measurement system; the invention realizes the weakening of the specific amplitude type noise and overcomes the adverse effect of the amplitude type noise on the weak measurement technology to a certain extent.)

1. A spectral data processing method for countering amplitude-type noise in weak measurement, characterized by comprising:

noise characteristic extraction step: carrying out domain transformation on the measured output spectrum and the input spectrum to obtain parameters for expressing the amplitude type noise characteristics in a transformation domain;

a filter design step: selecting and designing a filter according to the input spectrum and the extracted noise characteristic parameters in combination with the acquired processing requirements;

and (3) spectral data filtering: filtering and verifying the spectral data by using a designed filter to obtain a spectrum with reduced noise;

and calculating a weak measurement result: resolving and obtaining the value of the physical parameter to be measured through spectral analysis of a weak measurement system by using the spectrum after the noise is weakened;

the input spectrum is a signal light initial optical spectrum measured by a spectrometer from an optical signal source in advance;

The output spectrum is the optical spectrum of the signal light before the signal light finally enters the spectrometer after the signal light is subjected to the whole weak measurement process including selection before polarization, sensing of the parameter to be measured and selection after polarization;

the processing requirements in the filter design step comprise the filtering degree of noise, the limitation of the order of the filter and the limitation of passband distortion.

2. The method for processing spectral data of amplitude-type noise countering in weak measurement according to claim 1, characterized in that the noise characteristic extraction step includes:

noise characteristic extraction step M1.1: the input spectral data are n groups of numerical values with optical frequency and amplitude in one-to-one correspondence, wherein n is the number of optical frequency sampling points of the spectral data, the domain where the spectral data is located is marked as an omega domain, and the spectral distribution function is marked as P (omega); carrying out Fourier transformation on input spectrum data, transforming the input spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as P (x); when in [0, x ]_cut]Within the interval P (x) ≠ 0 at [ x ≠ 0_cut,∞]If the interval P (x) is 0, x will be_cutDefined as the cutoff threshold of the signal;

noise characteristic extraction step M1.2: the measured output spectrum data are n groups of numerical values with optical frequency and amplitude in one-to-one correspondence, wherein n is the number of optical frequency sampling points of the spectrum data, the domain where the spectrum data are located is marked as an omega domain, and the spectrum distribution function is marked as Q (omega); carrying out Fourier transformation on the output spectrum data, transforming the output spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as Q (x); when no noise is present, then at [0, x _cut]In the interval Q (x) ≠ 0 at [ x ≠ 0_cut,∞]Within the interval q (x) is 0; when noise is present, then at [0, x_cut]Outside the intervalIs [ x ] of_cut1,x_cut2]The interval also has Q (x) ≠ 0, where x_cut＜x_cut1＜x_cut2(ii) a X is to be_cut1And x_cut2Defined as the upper and lower cut-off thresholds, respectively, for noise.

3. The method for processing spectral data counteracting amplitude type noise in weak measurement according to claim 1, wherein the processing requirement in the filter designing step includes: the filtering degree of noise, the limitation of filter order and the limitation of passband distortion;

the noise filtering degree is the weakening degree of the corresponding amplitude of the noise before and after filtering on the transform domain of the optical spectrum;

the limitation of the filter order is the filtering calculation complexity, and the higher the filter order is, the higher the filtering calculation complexity is;

the limitation of the passband distortion is the limitation of the maximum ripple degree of the passband and whether the passband shape is required to be flat.

4. The method as claimed in claim 1, wherein the step of filtering the spectral data includes filtering the noisy acousto-optic spectral data in the ω -domain where the data spectrum is located according to the designed filter to obtain a spectrum with reduced noise.

5. The method for processing spectral data to combat amplitude type noise in weak measurement according to claim 1, wherein the spectral data filtering step further comprises: verifying the filtering effect in the x transform domain, if the filter does not meet the requirement, returning to the filter design step to adjust parameters or requirements, and redesigning;

the verification of the filtering effect comprises the steps of calculating a filtered x-domain spectrum and a noise-containing x-domain spectrum, and respectively calculating the response, ripple degree and filtering calculation complexity of a pass band and a stop band according to the filtered x-domain spectrum and the noise-containing x-domain spectrum; when the response, ripple degree and filtering calculation complexity of the pass band and the stop band meet preset conditions, calculating a weak measurement result; and when the responses, the ripple degrees and/or the filtering calculation complexity of the pass band and the stop band do not meet the preset conditions, repeatedly executing the steps from the filter design step to the step of filtering and processing the amplitude type noise until the responses, the ripple degrees and the filtering calculation complexity of the pass band and the stop band meet the preset conditions.

6. A spectral data processing system that counters amplitude-type noise in weak measurements, comprising:

a noise characteristic extraction module: carrying out domain transformation on the measured output spectrum and the input spectrum to obtain parameters for expressing the amplitude type noise characteristics in a transformation domain;

A filter design module: according to the input spectrum and the extracted noise characteristic parameters, a filter is selected and designed in combination with the processing requirements;

the spectral data filtering module: filtering and verifying the spectral data by using a designed filter to obtain a spectrum with reduced noise;

weak measurement result calculating module: calculating to obtain an estimated value of the weak measurement to the parameter to be measured by using the spectrum after the noise is weakened through the spectrum analysis of the weak measurement system;

the input spectrum is a signal light initial optical spectrum measured by a spectrometer from an optical signal source in advance;

the output spectrum is the optical spectrum of the signal light before the signal light finally enters the spectrometer after the signal light is subjected to the whole weak measurement process including selection before polarization, sensing of the parameter to be measured and selection after polarization;

the processing requirements in the filter design module include the degree of noise filtering, the limitation of filter order and the limitation of passband distortion.

7. The system according to claim 6, wherein the noise characteristics extraction module comprises:

noise characteristic extraction module M1.1: inputting spectral data as n with one-to-one correspondence of optical frequency and amplitude The group value is shown in the specification, wherein n is the number of optical frequency sampling points of the spectrum data, the domain where the spectrum data is located is marked as an omega domain, and the spectrum distribution function is marked as P (omega); carrying out Fourier transformation on input spectrum data, transforming the input spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as P (x); when in [0, x ]_cut]Within the interval P (x) ≠ 0 at [ x ≠ 0_cut,∞]If the interval P (x) is 0, x will be_cutDefined as the cutoff threshold of the signal;

noise characteristic extraction module M1.2: the measured output spectrum data are n groups of numerical values with optical frequency and amplitude in one-to-one correspondence, wherein n is the number of optical frequency sampling points of the spectrum data, the domain where the spectrum data are located is marked as an omega domain, and the spectrum distribution function is marked as Q (omega); carrying out Fourier transformation on the output spectrum data, transforming the output spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as Q (x); when no noise is present, then at [0, x_cut]In the interval Q (x) ≠ 0 at [ x ≠ 0_cut,∞]Within the interval q (x) is 0; when noise is present, then at [0, x_cut]Outside the interval [ x_cut1,x_cut2]The interval also has Q (x) ≠ 0, where x_cut＜x_cut1＜x_cut2(ii) a X is to be_cut1And x_cut2Defined as the upper and lower cut-off thresholds, respectively, for noise.

8. The system according to claim 6, wherein the filter design module processes the requirements including: the filtering degree of noise, the limitation of filter order and the limitation of passband distortion;

The noise filtering degree is the weakening degree of the corresponding amplitude of the noise before and after filtering on the transform domain of the optical spectrum;

the limitation of the filter order is the filtering calculation complexity, and the higher the filter order is, the higher the filtering calculation complexity is;

the limitation of the passband distortion is the limitation of the maximum ripple degree of the passband and whether the passband shape is required to be flat.

9. The system according to claim 6, wherein the spectral data filtering module comprises a filter obtained according to design, and the spectral data filtering module filters the acousto-optic spectral data containing noise in the ω domain where the data spectrum is located to obtain the spectrum with reduced noise.

10. The system according to claim 6, wherein the spectral data filtering module further comprises: verifying the filtering effect in the x transform domain, returning to a filter design module to adjust parameters or requirements if the filter does not meet the requirements, and redesigning;

the verification of the filtering effect comprises the steps of calculating a filtered x-domain spectrum and a noise-containing x-domain spectrum, and respectively calculating the response, ripple degree and filtering calculation complexity of a pass band and a stop band according to the filtered x-domain spectrum and the noise-containing x-domain spectrum; when the response, ripple degree and filtering calculation complexity of the pass band and the stop band meet preset conditions, calculating a weak measurement result; and when the responses, the ripple degree and/or the filtering calculation complexity of the pass band and the stop band do not meet the preset conditions, repeatedly triggering the filter design module to the filtering processing amplitude type noise module to execute the processing until the responses, the ripple degree and the filtering calculation complexity of the pass band and the stop band meet the preset conditions.

Technical Field

The invention relates to a technology for weakening noise in a weak measurement technology, in particular to a spectral data processing method and a system for resisting amplitude type noise in weak measurement, and more particularly to a spectral data processing method for reducing the influence of the amplitude type noise in the weak measurement process.

Background

Weak measurement techniques are a concept in quantum measurement. The weak measurement technology carries out strong measurement after the system evolves, and projects the state of the system to a certain state which is wanted by people, so that only weak influence is generated on the original system. Weak measurement techniques can yield results other than the eigenvalues of an operator under test at the expense of discarding a large number of measurements, i.e., multiple measurements, which can deviate significantly from the eigenvalues of the operator under test. For precise measurement, the method can realize high-precision and high-sensitivity measurement.

With the continuous development of weak measurement technology, technologies such as bias weak value amplification weak measurement, combined weak measurement and the like exist at present. The combined bias weak measurement realizes parameter measurement with extremely high sensitivity, and the practicability of the weak measurement is improved. However, the existing research on weak measurement is based on free space optical measurement, and the research on the optical fiber measurement system with wide practical application is lacked. There are some characteristics of fiber optic systems that are not free space: the influence of fiber twist, extrusion, dispersion, and electromagnetic magnetostriction effect of internal and external electromagnetic fields may all affect the weak measurement. For example, patent document CN110207822A discloses a high-sensitivity optical delay estimation system, which includes a light source module, a polarization state pre-modulation module, an optical delay sensing module, a combined spectrum detection module, and a data processing module. When applied to fiber optic systems, the method suffers from effects resulting from fiber imperfections, including amplitude-type noise. Amplitude-type noise can have an effect on the spectrum of the weak measurement and in turn on the sensitivity of the weak measurement scheme. The invention provides a light spectrum data processing method aiming at the characteristics of amplitude type noise, which can overcome the adverse effect of the amplitude type noise on a weak measurement technology to a certain extent.

The greatest advantage of weak measurement techniques over other measurement techniques is that sensitivity is significantly improved by post-polarization selection and analysis of spectral information. However, in the conventional weak measurement technical solution, the shape of the output spectrum is significantly changed by the amplitude type noise, so that the difficulty of reading the spectrum information is greatly increased, and in a serious case, the sensitivity is even lost.

Disclosure of Invention

In view of the defects in the prior art, the invention aims to provide a spectral data processing method and system for resisting amplitude type noise in weak measurement.

According to the invention, the spectral data processing method for resisting amplitude type noise in weak measurement comprises the following steps:

noise characteristic extraction step: carrying out domain transformation on the measured output spectrum and the input spectrum to obtain parameters for expressing the amplitude type noise characteristics in a transformation domain;

a filter design step: according to the input spectrum and the extracted noise characteristic parameters, a filter is selected and designed in combination with the processing requirements;

and (3) spectral data filtering: filtering and verifying the spectral data by using a designed filter to obtain a spectrum with reduced noise;

and calculating a weak measurement result: calculating to obtain an estimated value of the weak measurement to the parameter to be measured by using the spectrum after the noise is weakened through the spectrum analysis of the weak measurement system;

The input spectrum is a signal light initial optical spectrum measured by a spectrometer from an optical signal source in advance;

the output spectrum is the optical spectrum of the signal light before entering the spectrometer after the signal light undergoes the whole weak measurement process including selection before polarization, sensing of the parameter to be measured and selection after polarization.

The processing requirements in the filter design step comprise the filtering degree of noise, the limitation of the order of the filter and the limitation of passband distortion.

Preferably, the noise characteristic extracting step includes:

noise characteristic extraction step M1.1: the input spectral data are n groups of numerical values with optical frequency and amplitude in one-to-one correspondence, wherein n is the number of optical frequency sampling points of the spectral data, the domain where the spectral data is located is marked as an omega domain, and the spectral distribution function is marked as P (omega); carrying out Fourier transformation on input spectrum data, transforming the input spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as P (x); when in [0, x ]_cut]Within the interval P (x) ≠ 0 at [ x ≠ 0_cut,∞]If the interval P (x) is 0, x will be_cutDefined as the cutoff threshold of the signal;

noise characteristic extraction step M1.2: the measured output spectrum data are n groups of numerical values with optical frequency and amplitude in one-to-one correspondence, wherein n is the number of optical frequency sampling points of the spectrum data, the domain where the spectrum data are located is marked as an omega domain, and the spectrum distribution function is marked as Q (omega); carrying out Fourier transformation on the output spectrum data, transforming the output spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as Q (x); when no noise is present, then at [0, x _cut]In the interval Q (x) ≠ 0 at [ x ≠ 0_cut,∞]Within the interval q (x) is 0; when noise is present, then at [0, x_cut]Outside the interval [ x_cut1,x_cut2]The interval also has Q (x) ≠ 0, where x_cut＜x_cut1＜x_cut2(ii) a X is to be_cut1And x_cut2Defined as the upper and lower cut-off thresholds, respectively, for noise.

Preferably, the processing requirements in the filter design step include: the filtering degree of noise, the limitation of filter order and the limitation of passband distortion;

the noise filtering degree is the weakening degree of the corresponding amplitude of the noise before and after filtering on the transform domain of the optical spectrum;

the limitation of the filter order is the filtering calculation complexity, and the higher the filter order is, the higher the filtering calculation complexity is;

the limitation of the passband distortion is the limitation of the maximum ripple degree of the passband and whether the passband shape is required to be flat.

Preferably, the spectrum data filtering step includes filtering the noise-containing sound spectrum data in an ω domain where the data spectrum is located according to the designed filter to obtain a spectrum with reduced noise.

Preferably, the spectral data filtering step further comprises: verifying the filtering effect in the x transform domain, if the filter does not meet the requirement, returning to the filter design step to adjust parameters or requirements, and redesigning;

The verification of the filtering effect comprises the steps of calculating a filtered x-domain spectrum and a noise-containing x-domain spectrum, and respectively calculating the response, ripple degree and filtering calculation complexity of a pass band and a stop band according to the filtered x-domain spectrum and the noise-containing x-domain spectrum; when the response, ripple degree and filtering calculation complexity of the pass band and the stop band meet preset conditions, calculating a weak measurement result; and when the responses, the ripple degrees and/or the filtering calculation complexity of the pass band and the stop band do not meet the preset conditions, repeatedly executing the steps from the filter design step to the step of filtering and processing the amplitude type noise until the responses, the ripple degrees and the filtering calculation complexity of the pass band and the stop band meet the preset conditions.

According to the present invention, there is provided a spectral data processing system for countering amplitude type noise in weak measurement, comprising:

a noise characteristic extraction module: carrying out domain transformation on the measured output spectrum and the input spectrum to obtain parameters for expressing the amplitude type noise characteristics in a transformation domain;

a filter design module: according to the input spectrum and the extracted noise characteristic parameters, a filter is selected and designed in combination with the processing requirements;

the spectral data filtering module: filtering and verifying the spectral data by using a designed filter to obtain a spectrum with reduced noise;

Weak measurement result calculating module: calculating to obtain an estimated value of the weak measurement to the parameter to be measured by using the spectrum after the noise is weakened through the spectrum analysis of the weak measurement system;

the input spectrum is a signal light initial optical spectrum measured by a spectrometer from an optical signal source in advance;

the output spectrum is the optical spectrum of the signal light before entering the spectrometer after the signal light undergoes the whole weak measurement process including selection before polarization, sensing of the parameter to be measured and selection after polarization.

The processing requirements in the filter design module include the degree of noise filtering, the limitation of filter order and the limitation of passband distortion.

Preferably, the noise characteristic extraction module includes:

noise characteristic extraction module M1.1: the input spectral data are n groups of numerical values with one-to-one correspondence of optical frequency and amplitude valueIn the method, n is the number of optical frequency sampling points of the spectrum data, the domain where the spectrum data is located is marked as an omega domain, and the spectrum distribution function is marked as P (omega); carrying out Fourier transformation on input spectrum data, transforming the input spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as P (x); when in [0, x ]_cut]Within the interval P (x) ≠ 0 at [ x ≠ 0_cut,∞]If the interval P (x) is 0, x will be_cutDefined as the cutoff threshold of the signal;

Noise characteristic extraction module M1.2: the measured output spectrum data are n groups of numerical values with optical frequency and amplitude in one-to-one correspondence, wherein n is the number of optical frequency sampling points of the spectrum data, the domain where the spectrum data are located is marked as an omega domain, and the spectrum distribution function is marked as Q (omega); carrying out Fourier transformation on the output spectrum data, transforming the output spectrum data from an omega domain to an x domain, and recording a distribution function after transformation as Q (x); when no noise is present, then at [0, x_cut]In the interval Q (x) ≠ 0 at [ x ≠ 0_cut,∞]Within the interval q (x) is 0; when noise is present, then at [0, x_cut]Outside the interval [ x_cut1,x_cut2]The interval also has Q (x) ≠ 0, where x_cut＜x_cut1＜x_cut2(ii) a X is to be_cut1And x_cut2Defined as the upper and lower cut-off thresholds, respectively, for noise.

Preferably, the processing requirements in the filter design module include: the filtering degree of noise, the limitation of filter order and the limitation of passband distortion;

the noise filtering degree is the weakening degree of the corresponding amplitude of the noise before and after filtering on the transform domain of the optical spectrum;

the limitation of the filter order is the filtering calculation complexity, and the higher the filter order is, the higher the filtering calculation complexity is;

the limitation of the passband distortion is the limitation of the maximum ripple degree of the passband and whether the passband shape is required to be flat.

Preferably, the spectrum data filtering module includes a filter obtained according to design, and filters the noise-containing sound spectrum data in an ω domain where the data spectrum is located to obtain a spectrum with reduced noise.

Preferably, the spectral data filtering module further comprises: verifying the filtering effect in the x transform domain, returning to a filter design module to adjust parameters or requirements if the filter does not meet the requirements, and redesigning;

the verification of the filtering effect comprises the steps of calculating a filtered x-domain spectrum and a noise-containing x-domain spectrum, and respectively calculating the response, ripple degree and filtering calculation complexity of a pass band and a stop band according to the filtered x-domain spectrum and the noise-containing x-domain spectrum; when the response, ripple degree and filtering calculation complexity of the pass band and the stop band meet preset conditions, calculating a weak measurement result; and when the responses, the ripple degree and/or the filtering calculation complexity of the pass band and the stop band do not meet the preset conditions, repeatedly triggering the filter design module to the filtering processing amplitude type noise module to execute the processing until the responses, the ripple degree and the filtering calculation complexity of the pass band and the stop band meet the preset conditions.

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

1. the invention can recover the sensitivity of the weak measurement technology to the sensitivity without the noise under the amplitude type noise, and realize higher signal-to-noise ratio, thereby improving the robustness of the weak measurement technology;

2. The invention utilizes the transform domain analysis and processing of the optical frequency spectrum data to effectively weaken specific amplitude type noise in the data processing aspect and realize the quick and flexible compensation of the sensitivity. Because the physical structure of the front end of the system does not need to be modified, the technical implementation difficulty is reduced, and the technical reliability is improved.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:

FIG. 1 is a flow chart of the invention;

FIG. 2 is a block diagram of an embodiment of the present invention;

FIG. 3 is a light path diagram of a weak measurement system in an embodiment of the present invention;

FIG. 4 shows the optical spectrum (ω -domain) and sample transform domain spectrum (x-domain) after amplitude-type noise is added;

FIG. 5 is an optical spectrum (ω -domain) and sample transform domain spectrum (x-domain) of the filtered signal;

FIG. 6 comparison of weak measurement response curves before and after filtering (curve slope is proportional to sensitivity);

fig. 7 shows a weak measured response curve (curve slope is proportional to sensitivity) without amplitude type noise.

Detailed Description

The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.

First, the amplitude type noise and its behavior in the optical spectrum according to the present invention will be described. To analyze the effect of amplitude-type noise in the optical spectrum, a light-field signal E is taken into account_sAnd a noise signal E_nHaving the form:

E_s＝α(ω)e^jωt

where α (ω) is the intensity coefficient of the amplitude-type noise at the corresponding optical frequency, ω is the optical frequency, t is the time, t is the frequency_dIs the delay difference between the signal and the noise. When t is_dWhen the coherent time of the signal light is shorter than the coherence time of the signal light, noise interferes with the signal, and the intensity of the interference is proportional to the signal amplitude, and thus the noise is called amplitude-type noise.

When the signal light is a broad spectrum light, its spectrum P (ω) has the following form:

P(ω)＝|E_s+E_n,ω|²＝|α(ω)|²+|β|²+2α(ω)βcos(ωt_d)

its Fourier transform p (x) is:

wherein F₀(x)＝FT{α²(ω) } denotes the Fourier transform of the signal spectrum, F₁(x) FT { α (ω) } represents the fourier transform of the signal field intensity spectrum, and the second term represents the influence of amplitude-type noise on the signal light spectrum in the transform domain. From the relationship between field strength and intensity, we can know F₀(x) Is not less than F₁(x) The bandwidth of (c).