An implementation device and method of a single-photon spectrometer

Abstract

The application belongs to the technical field of optical measurement, and discloses an implementation device and method of a single-photon spectrometer, which comprises a spectral wavelength scanning module, collects a wide-spectrum light signal, carries out wave separation processing, and calculates a photon counting rate between adjacent wavelengths; wavelength steps are dynamically adjusted according to the photon counting rate, adaptive scanning of spectral changes is realized, and a single-spectrum light signal is output; a self-focusing detection module converts the single-spectrum light signal into a photon event through a preset array channel; in a preset time window, the variance and Shannon entropy of each photon event are calculated, and a local uncertainty index is constructed; spectral domain self-focusing and resampling are carried out according to the local uncertainty index, and photon counting data are obtained; high-precision, high-stability and high-efficiency spectral light signal detection and visual analysis are realized.

Description

Technical Field

[0001] This invention relates to the field of optical measurement technology, and more specifically, to a device and method for implementing a single-photon spectrometer. Background Technology

[0002] Existing devices and methods for implementing single-photon spectrometers mainly suffer from the following problems:

[0003] Currently, single-photon spectrometers, as important instruments for achieving high-sensitivity analysis of extremely weak light signals, are widely used in quantum optics, optical communication, biofluorescence detection, and material spectral characterization. Existing single-photon spectrometers typically employ broadband light sources, scanning different center wavelengths in a fixed-step manner using tunable optical filters, and recording photon counts at each step using a single-photon detector to obtain spectral distribution information. However, existing technologies have the following shortcomings:

[0004] Existing single-photon spectrometers generally employ a fixed-wavelength step scanning strategy, failing to consider the intensity differences and rates of change of the spectral signal across different wavelength bands. When the spectrum changes sharply in certain regions, a fixed step size may miss local details; while in regions with gentle spectral changes, it wastes measurement time resources. Due to the lack of a real-time adaptive step size adjustment mechanism based on the rate of change of photon counts, existing systems struggle to balance spectral resolution and measurement efficiency.

[0005] Existing systems typically use observed photon counts directly during spectral scanning without dynamically correcting for fluctuations in light source power, changes in optical transmittance, or differences in detector quantum efficiency, thus limiting the accuracy of spectral reconstruction. Especially under weak signal or high-throughput conditions, the lack of joint calibration of detection efficiency and light source flux easily introduces systematic errors.

[0006] The sampling integration time of existing single-photon spectrometers is mostly set to a fixed value, and cannot be dynamically allocated according to the rate of spectral change or the amount of local information. This results in insufficient sampling in high-information regions and redundant sampling in low-information regions, making it difficult to achieve the optimal balance between measurement time and spectral resolution.

[0007] Under high photon flux conditions, single-photon detectors suffer from a dead-time effect, meaning that no new photons can be recorded within a recovery period after a photon event is triggered. Current technologies generally lack systematic dead-time correction for this effect, leading to significantly lower photon counts and consequently affecting the recovery of the true photon flux and the accurate estimation of spectral parameters.

[0008] In view of this, the present invention proposes a device and method for implementing a single-photon spectrometer to solve the above problems. Summary of the Invention

[0009] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution:

[0010] An apparatus for implementing a single-photon spectrometer, comprising:

[0011] The spectral wavelength scanning module acquires broadband optical signals, performs wavelength division processing, calculates the photon count change rate between adjacent wavelengths, dynamically adjusts the wavelength step according to the photon count change rate, realizes adaptive scanning for spectral changes, and outputs a monospectral optical signal.

[0012] The self-focusing detection module converts single-spectral light signals into photon events through a preset array channel; within a preset time window, it calculates the variance and Shannon entropy of each photon event to construct a local uncertainty index; based on the local uncertainty index, it performs spectral domain self-focusing and resampling to obtain photon counting data.

[0013] The main control analysis and calibration module establishes a Poisson distribution statistical model of photon counting based on photon counting data, and performs adaptive allocation of timing resources based on the maximization of Fisher information, calculates the optimal sampling time of the array channel, and performs dead time correction to obtain the true photon count value.

[0014] The dark count compensation module identifies the dark count and background noise of the array channel based on the real photon count value. It acquires the dark spectral template through no light input and performs adaptive subtraction compensation during measurement. The adaptive subtraction compensation is used to adaptively subtract photon events caused by dark count and background noise, realizing joint noise reduction in the wavelength domain and time domain, and generating a net spectral count matrix.

[0015] The human-computer interaction display module plots the wavelength-power curve of the broadband optical signal based on the net spectral counting matrix and allows users to manually calibrate it.

[0016] Specifically, the method for acquiring broadband optical signals and performing wavelength division processing includes:

[0017] The broadband optical signal emitted by the light source under test is input to the spectral wavelength scanning module via fiber optic coupling. After being modulated by a pre-optical filter, it is input to the beam splitter. The beam splitter includes a tunable optical filter, a Fabry-Perot resonator, or a grating, which is used to perform spectral separation on the input broadband optical signal and separate the light energy of different wavelength components to form a wavelength-splitting beam.

[0018] Specifically, the method for obtaining the photon count change rate includes:

[0019] The initial wavelength step is preset, driving the tunable optical filter to gradually change within the preset wavelength range, thereby outputting single-spectral optical signals with different center wavelengths in the time series; the photon count is recorded synchronously at each wavelength step point, and the rate of change of photon count between adjacent wavelength step points is calculated.

[0020] Specifically, the method for outputting a single-spectral optical signal includes:

[0021] Based on the photon count change rate, the next wavelength step is dynamically adjusted. The dynamically adjusted wavelength step is used to drive the tunable optical filter to scan to the next center wavelength. At the same time, the photon count value is recorded. The photon count change rate is repeatedly calculated and the step size is adjusted until the scanning of the entire wavelength range is completed, and a single-spectrum optical signal is output.

[0022] Specifically, the method for constructing the local uncertainty index includes:

[0023] The single-spectral optical signal is converted into photon events through a preset array channel. Each photon event records the photon arrival time and the corresponding array channel number. Within a preset time window, the photon events of each array channel are counted to obtain the photon event count within the preset time window.

[0024] The statistical variance of photon events counted by each array channel within a preset time window is calculated. The distribution of photon events within the preset time window is regarded as a probability distribution. Shannon entropy is calculated, and the variance of each array channel is weighted and fused with Shannon entropy to construct a local uncertainty index.

[0025] Specifically, the method for acquiring the photon counting data includes:

[0026] Based on the local uncertainty index calculated by each array channel within a preset time window, the single-spectrum optical signal is subjected to self-focusing processing in the spectral domain. The self-focusing processing includes a preset local uncertainty index threshold and adjusting the sampling density according to the preset local uncertainty index threshold.

[0027] For wavelength regions where the local uncertainty index is greater than the preset local uncertainty index threshold, the sampling density is increased; for wavelength regions where the local uncertainty index is less than or equal to the preset local uncertainty index threshold, the sampling density is decreased. Based on the self-focusing processing results, the spectral signal is resampled. The resampling is achieved by adjusting the number of photon count samplings at each wavelength step point, and finally, the photon count data is obtained.

[0028] Specifically, the method for obtaining the actual photon count value includes:

[0029] Based on photon counting data, a Poisson distribution statistical model for photon counting is established. The Poisson distribution statistical model predetermines the expected photon count value at each wavelength step by the incident photon flux density, detection efficiency, and sampling integration time. The photon counting data follows a Poisson distribution.

[0030] By maximizing Fisher information, the optimal sampling integration time for each wavelength step is determined under the preset total measurement time budget constraint. The sampling integration time is initialized using an approximate allocation criterion and serves as the baseline for event-triggered local scheduling. Simultaneously, array channel dead time correction is performed to obtain the true photon count value.

[0031] Specifically, the method for generating the net spectral count matrix includes:

[0032] Under the condition of not receiving broadband optical signals, dark photon events are collected through each array channel of the single photon detection linear array. The dark spectral template is obtained through cumulative statistics. The real photon count value of each array channel is compared and analyzed with the dark spectral template. The photon events caused by dark count and background noise are adaptively subtracted to obtain the noise-corrected net photon count value.

[0033] Adaptive subtraction compensation is performed jointly in the wavelength and time domains. The net photon count values ​​after adaptive subtraction compensation for each array channel at each wavelength step point are organized into a matrix form to form a net spectral count matrix.

[0034] Specifically, the method for plotting the wavelength-power curve of the broadband optical signal and allowing users to manually calibrate it includes:

[0035] Based on the net spectral count matrix, a wavelength-power curve of the broadband optical signal is plotted with wavelength as the abscissa and power as the ordinate. The power is obtained by converting the net photon count value at each wavelength step point, combined with the photon energy and sampling integration time of the corresponding wavelength. The human-computer interaction display module also includes a manual calibration function, which allows users to adjust system parameters by providing an adjustable parameter interface and display the adjusted wavelength-power curve in real time.

[0036] A method for implementing a single-photon spectrometer includes:

[0037] S1. Acquire broadband optical signals and perform wavelength division processing to calculate the photon count change rate between adjacent wavelengths; dynamically adjust the wavelength step according to the photon count change rate to achieve adaptive scanning of spectral changes and output monospectral optical signals.

[0038] S2. Convert the single-spectral optical signal into photon events through a preset array channel; within a preset time window, calculate the variance and Shannon entropy of each photon event to construct a local uncertainty index; perform spectral self-focusing and resampling based on the local uncertainty index to obtain photon counting data.

[0039] S3. Based on the photon counting data, establish a Poisson distribution statistical model for photon counting, and adaptively allocate timing resources according to the criterion of maximizing Fisher information. Calculate the optimal sampling time of the array channel, and simultaneously perform dead time and nonlinear correction to obtain the true photon count value.

[0040] S4. Based on the real photon count value, identify the dark count and background noise of the array channel, acquire the dark spectral template through no light input, and perform adaptive subtraction compensation in the measurement. The adaptive subtraction compensation is used to adaptively subtract photon events caused by dark count and background noise, realize joint noise reduction in the wavelength domain and time domain, and generate a net spectral count matrix.

[0041] S5. Based on the net spectral count matrix, plot the wavelength-power curve of the broadband optical signal and allow users to manually calibrate it.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] By calculating the photon count change rate at adjacent wavelength step points and dynamically adjusting the step size, the scanning step size can adaptively adjust according to spectral changes. This allows for increased sampling density in regions of rapid spectral change and decreased sampling density in regions of gradual spectral change, thus capturing spectral features more accurately. Dynamic step size adjustment ensures high resolution in key spectral regions while avoiding wasting timing resources in flat regions, achieving optimized allocation of timing resources and balancing resolution and efficiency. Recording photon counts and calculating the change rate in real time at each wavelength step point allows for sensitive response to spectral changes, reducing local information loss and improving the integrity of the spectral signal and the accuracy of subsequent data processing. The system can adaptively adjust the scanning strategy for spectral signals from different light sources and with different intensity distributions, enhancing its applicability and flexibility.

[0044] By modeling with Poisson distribution, correcting for incident photon flux and detection efficiency, and correcting for dead time, more accurate true photon counts can be obtained, thereby improving the reliability and accuracy of single-spectral optical signals. Combining the Fisher information maximization principle and local event-triggered scheduling, the sampling integration time for each wavelength step can be adaptively allocated, increasing the sampling density at key wavelengths and saving time measurement resources in flat regions, thus balancing spectral resolution and measurement efficiency. Incident photon flux and detection efficiency can be corrected in real time during measurement, and the sampling integration time can be dynamically adjusted. The system can adapt to light source fluctuations, environmental changes, and differences in spectral characteristics, achieving more stable and reliable spectral measurements. Initializing the sampling integration time using an approximate allocation criterion and using it as a local scheduling baseline provides a simple and easy-to-implement engineering method while ensuring that the measurement results are close to the optimal allocation of Fisher information. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the device structure for implementing a single-photon spectrometer according to the present invention.

[0046] Figure 2 This is a schematic diagram of the implementation method of a single-photon spectrometer according to the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Example 1: Please refer to Figure 1 As shown, this embodiment provides an apparatus for implementing a single-photon spectrometer, including:

[0049] The spectral wavelength scanning module acquires broadband optical signals, performs wavelength division processing, calculates the photon count change rate between adjacent wavelengths, dynamically adjusts the wavelength step according to the photon count change rate, realizes adaptive scanning for spectral changes, and outputs a monospectral optical signal.

[0050] The self-focusing detection module converts single-spectral light signals into photon events through a preset array channel; within a preset time window, it calculates the variance and Shannon entropy of each photon event to construct a local uncertainty index; based on the local uncertainty index, it performs spectral domain self-focusing and resampling to obtain photon counting data.

[0051] The main control analysis and calibration module establishes a Poisson distribution statistical model of photon counting based on photon counting data, and performs adaptive allocation of timing resources based on the maximization of Fisher information, calculates the optimal sampling time of the array channel, and performs dead time correction to obtain the true photon count value.

[0052] The dark count compensation module identifies the dark count and background noise of the array channel based on the real photon count value. It acquires the dark spectral template through no light input and performs adaptive subtraction compensation during measurement. The adaptive subtraction compensation is used to adaptively subtract photon events caused by dark count and background noise, realizing joint noise reduction in the wavelength domain and time domain, and generating a net spectral count matrix.

[0053] The human-computer interaction display module plots the wavelength-power curve of the broadband optical signal based on the net spectral counting matrix and allows users to manually calibrate it.

[0054] Methods for acquiring broadband optical signals and performing wavelength division processing include:

[0055] The broadband optical signal emitted by the light source under test is input to the spectral wavelength scanning module via fiber optic coupling, and then modulated by a pre-optical filter before being input to the beam splitter to improve the beam uniformity and signal stability. The beam splitter includes a tunable optical filter, a Fabry-Perot resonator, or a grating, which is used to perform spectral separation on the input broadband optical signal and separate the light energy of different wavelength components to form a wavelength-splitting beam.

[0056] Methods for obtaining the rate of change in photon count include:

[0057] A preset initial wavelength step drives a tunable optical filter to gradually change within a preset wavelength range, thereby outputting single-spectral optical signals with different center wavelengths in a time series. At each wavelength step point, the photon count is recorded synchronously, and the rate of change of photon count between adjacent wavelength step points is calculated to characterize the change of light intensity gradient of the spectral signal between adjacent wavelengths.

[0058] The rate of change in photon count is: ;in, Indicates the first The rate of change of photon count at each wavelength step point; Indicates the first The center wavelength corresponding to each wavelength step point, that is, the center wavelength corresponding to each scan step in the spectral scanning process is... ; Indicates the first The center wavelength corresponding to each wavelength step point; Indicates at the center wavelength The photon count value measured below; Indicates at the center wavelength The photon count value measured below; This represents the change in photon count between two adjacent wavelength step points; Indicates the first Wavelength stepping at each step point, i.e., adjacent step points arrive The wavelength interval; Indicates the index of the wavelength step point;

[0059] For example, the single-photon spectrometer of this invention measures a broadband visible-near-infrared light signal. By controlling the center wavelength through a tunable optical filter, a local range of 450nm-600nm is selected within the spectral measurement range of 400nm-900nm for dynamic scanning testing. A preset initial wavelength step is used. With a wavelength of 5nm, the tunable optical filter begins scanning at a starting wavelength of 450nm, gradually adjusting the wavelength output, and synchronously recording the photon count value at each wavelength step point.

[0060] The step points and corresponding photon counts in the experiment are shown in the table below:

[0061]

[0062] Photon count change rate When wavelength steps At 5nm, we get:

[0063] First step (450-455nm): ;

[0064] Third step (460-465nm): ;

[0065] Fifth step forward (470-475nm): ;

[0066] This shows that the photon count change rate is the largest and the spectral change is the most dramatic in the 460-465nm region.

[0067] Methods for outputting monospectral optical signals include:

[0068] Based on the rate of change in photon count, the wavelength step is dynamically adjusted for the next step. The dynamically adjusted wavelength step is as follows: ;in, Indicates the number after dynamic adjustment Wavelength stepping at each step point; This indicates the preset initial wavelength step, which is the initial reference for wavelength step adjustment; This represents the sensitivity coefficient, used to adjust the response of the step size to the rate of change in photon count;

[0069] The dynamically adjusted wavelength step is used to drive the tunable optical filter to scan to the next center wavelength, while recording the photon count value. The photon count change rate is repeatedly calculated and the step size is adjusted until the entire wavelength range is scanned, and a monospectral light signal is output.

[0070] This invention addresses the following technical problems of existing technologies: Current single-photon spectrometers typically employ fixed-wavelength step scanning of broadband optical signals, but the intensity variations of the spectral signal differ across different wavelength bands. Fixed stepping can easily miss crucial information in regions of steep spectral changes, while wasting timing resources in flat regions. Existing technologies lack dynamic adjustment for photon count changes between adjacent wavelengths and a real-time adaptive mechanism, resulting in low scanning efficiency and difficulty in balancing spectral signal resolution with timing resource utilization. During spectral scanning, a fixed step size may prevent the capture of local spectral details, affecting the accuracy of subsequent spectral signal processing (such as photon count statistics, noise correction, and spectral reconstruction).

[0071] The advantages over existing technologies are as follows: By calculating the photon count change rate at adjacent wavelength step points and dynamically adjusting the step size, the scanning step size can be adaptively adjusted according to spectral changes. This allows for increased sampling density in regions of rapid spectral change and decreased sampling density in regions of slow spectral change, thus capturing spectral features more accurately. Dynamic step size adjustment ensures high resolution in key spectral regions while avoiding wasting timing resources in flat regions, achieving optimized allocation of timing resources and balancing resolution and efficiency. Recording photon counts and calculating the change rate in real time at each wavelength step point allows for sensitive response to spectral changes, reducing local information loss and improving the integrity of the spectral signal and the accuracy of subsequent data processing (such as photon count statistics, noise correction, and resampling). It also enables adaptive adjustment of the scanning strategy for spectral signals from different light sources and with different intensity distributions, enhancing the system's applicability and flexibility.

[0072] Methods for constructing local uncertainty indices include:

[0073] The single-spectral optical signal is converted into photon events through a preset array channel. Each photon event records the photon arrival time and the corresponding array channel number. Within a preset time window, the photon events of each array channel are counted to obtain the photon event count within the preset time window. The width of the preset time window can be adaptively set according to the photon counting rate and the desired time resolution.

[0074] It should be noted that the preset array channel is a single-photon detection line array. Each channel of the single-photon detection line array independently detects the incident single-spectral light signal and outputs a pulse signal. Each detected photon pulse is defined as a photon event, and the photon arrival time and corresponding channel number are recorded to form a photon event sequence.

[0075] The statistical variance of photon events counted by each array channel within a preset time window is calculated to characterize the intensity of random fluctuations in photon events. The distribution of photon events within the preset time window is regarded as a probability distribution, and Shannon entropy is calculated to characterize the statistical uncertainty of photon events. The variance of each array channel and Shannon entropy are weighted and fused to construct a local uncertainty index.

[0076] Methods for acquiring photon counting data include:

[0077] Based on the local uncertainty index calculated by each array channel within a preset time window, the single-spectrum optical signal is subjected to self-focusing processing in the spectral domain. The self-focusing processing includes a preset local uncertainty index threshold and adjusting the sampling density according to the preset local uncertainty index threshold.

[0078] It should be noted that the spectral domain is formed by a discrete wavelength sequence obtained by stepping a broadband optical signal within a preset wavelength range using a spectral wavelength scanning module. At each wavelength step point, the single-spectral optical signal is detected and photon events are counted by the single-photon detection line array of the preset array channel, forming a photon count sequence for the corresponding wavelength point. Arranging the photon counts at each wavelength step point constitutes the discrete spectral domain used for spectral domain self-focusing processing.

[0079] For wavelength regions where the local uncertainty index is greater than the preset local uncertainty index threshold, the sampling density is increased to improve the local spectral resolution; for wavelength regions where the local uncertainty index is less than or equal to the preset local uncertainty index threshold, the sampling density is reduced to save measurement time resources.

[0080] The spectral signal is resampled based on the self-focusing processing results. The resampling is achieved by adjusting the number of photon count samplings at each wavelength step point, and finally obtaining the photon count data.

[0081] Methods for obtaining the actual photon count include:

[0082] Based on photon counting data, a Poisson distribution statistical model for photon counting is established. The Poisson distribution statistical model predetermines the expected photon count value at each wavelength step by the incident photon flux density, detection efficiency, and sampling integration time. The photon counting data follows a Poisson distribution.

[0083] It should be noted that the incident photon flux density is the number of photons incident on the single-photon detector linear array per unit area per unit time, reflecting the optical signal intensity at different wavelength step points. This parameter can be obtained during the system initialization phase through calibration measurements of the light source output power, optical transmittance, and optical path loss, and can be corrected in real time by preset array channels during the measurement process to compensate for incident changes caused by light source fluctuations.

[0084] Detection efficiency represents the photoelectric conversion efficiency of a single-photon detector linear array at various wavelengths, i.e., the probability that an incident photon is effectively detected and generates a counting pulse. Detection efficiency can be obtained from the detector's quantum efficiency curve or through experimental calibration, and can be dynamically corrected according to changes in environmental parameters such as temperature and bias voltage to maintain model accuracy.

[0085] The sampling integration time is the length of time the system takes to perform cumulative counting of photon events at each wavelength step. The allocation of the sampling integration time is adaptively adjusted based on the total measurement time budget constraint and the Fisher information maximization criterion. The system can first initialize the sampling time based on an approximate allocation criterion, and then dynamically adjust it during the measurement process according to the spectral change rate or local uncertainty index to achieve a balance between maximizing global information and utilizing measurement time resources.

[0086] By maximizing Fisher information, the optimal sampling integration time for each wavelength step is determined within the preset total measurement time budget. Within the preset total measurement time budget, the goal is to select the sampling integration time that maximizes Fisher information.

[0087] Fisher's information is: ;in, Indicates Fisher's information content; Indicates at the center wavelength The expected photon count, i.e. the average count, is determined by the incident photon flux density, detection efficiency, and sampling integration time. The weighting factors represent Fisher information; This represents the spectral parameters to be estimated, used to characterize the spectral shape of a single-spectral optical signal;

[0088] The sampling integration time is initialized using an approximate allocation criterion and serves as the baseline for event-triggered local scheduling to achieve a balance between maximizing global information (maximizing Fisher information) and timing resource utilization; at the same time, array channel dead time correction is performed to obtain the true photon count value.

[0089] The actual photon count is: ;in, Indicates the first The actual photon count values ​​of each array channel; Indicates the first The observed photon count values ​​for each array channel; The dead time of the array channel refers to the time interval during which the detector needs to recover after each photon event is triggered, during which the detector cannot record new photons; This represents the total dead time accumulated by the observed photon count within a unit sampling time. Indicates the index of the array channel;

[0090] This invention addresses the following technical problems in existing technologies: In existing technologies, photon counting for single-photon detectors typically uses observed values ​​directly, without considering the impact of light source fluctuations, detector quantum efficiency differences, or environmental factors (such as temperature and bias voltage) on the counting, leading to decreased measurement accuracy. In existing technologies, the sampling integration time is usually a fixed value, unable to be dynamically allocated according to the spectral change rate or local spectral information content. This may result in insufficient sampling in wavelength regions with high information content, or wasted timing resources in wavelength regions with low information content. Traditional technologies do not systematically correct for detector dead time, leading to severely understated counting at high photon fluxes, affecting the accurate estimation of spectral parameters.

[0091] Compared to existing technologies, the advantages are as follows: By modeling with Poisson distribution, correcting incident photon flux and detection efficiency, and correcting dead time, more accurate true photon counts can be obtained, thereby improving the reliability and accuracy of single-spectral optical signals. Combining the Fisher information maximization principle and local event-triggered scheduling, the sampling integration time for each wavelength step point can be adaptively allocated, increasing the sampling density at key wavelength points and saving time measurement resources in flat regions, thus balancing spectral resolution and measurement efficiency. Incident photon flux and detection efficiency can be corrected in real time during measurement, and the sampling integration time can be dynamically adjusted. The system can adapt to light source fluctuations, environmental changes, and differences in spectral characteristics, achieving more stable and reliable spectral measurements. Initializing the sampling integration time using an approximate allocation criterion and using it as a local scheduling baseline provides a simple and easy-to-implement engineering method, while ensuring that the measurement results are close to the optimal allocation of Fisher information.

[0092] Methods for generating the net spectral count matrix include:

[0093] Under conditions where no broadband optical signal is received (i.e., no light input), dark photon events are collected through each array channel of the single-photon detector linear array. The dark spectral template is obtained through cumulative statistics, reflecting the inherent noise characteristics of the detector and background ambient light interference. The actual photon count value of each array channel is compared and analyzed with the dark spectral template, and the photon events caused by dark count and background noise are adaptively subtracted to obtain the noise-corrected net photon count value.

[0094] Adaptive subtraction compensation is performed jointly in the wavelength and time domains. The net photon count values ​​after adaptive subtraction compensation for each array channel at each wavelength step point are organized into a matrix form, forming a net spectral count matrix. In the net spectral count matrix, the rows represent the wavelength step points, and the columns represent the array channels.

[0095] Methods for plotting wavelength-power curves of broadband optical signals and allowing users to manually calibrate include:

[0096] Based on the net spectral count matrix, a wavelength-power curve of the broadband optical signal is plotted with wavelength as the abscissa and power as the ordinate. Power is obtained by converting the net photon count at each wavelength step point, combined with the photon energy of the corresponding wavelength and the sampling integration time. The conversion process of spectral power is based on the relationship between photon energy and cumulative count. Specifically, the spectral power at each wavelength step point is obtained by multiplying the net photon count at that step point by the energy of a single photon corresponding to that wavelength, and then dividing by the corresponding sampling integration time. That is, the net photon count reflects the total number of photons actually detected at that wavelength step point, and the photon energy can be calculated from Planck's constant, the speed of light, and the wavelength.

[0097] The human-computer interaction display module also includes a manual calibration function, which allows users to adjust system parameters (such as light source power, detector gain, sampling integration time, etc.) by providing an adjustable parameter interface, and displays the adjusted wavelength-power curve in real time.

[0098] The preset local uncertainty index threshold is set by staff. By collecting different local uncertainty indices, the average value of multiple local uncertainty indices is taken as the preset local uncertainty index threshold.

[0099] In this embodiment, by calculating the photon count change rate at adjacent wavelength step points and dynamically adjusting the step size, the scanning step size can adaptively adjust according to spectral changes. This allows for increased sampling density in regions of rapid spectral change and decreased sampling density in regions of gradual spectral change, thereby capturing spectral features more accurately. Dynamic step size adjustment ensures high resolution in key spectral regions while avoiding wasting timing resources in flat regions, achieving optimized allocation of timing resources and balancing resolution and efficiency. Recording photon counts and calculating the change rate in real time at each wavelength step point allows for sensitive response to spectral changes, reducing local information loss and improving the integrity of the spectral signal and the accuracy of subsequent data processing. It can adaptively adjust the scanning strategy for spectral signals from different light sources and with different intensity distributions, enhancing the system's applicability and flexibility.

[0100] By modeling with Poisson distribution, correcting for incident photon flux and detection efficiency, and correcting for dead time, more accurate true photon counts can be obtained, thereby improving the reliability and accuracy of single-spectral optical signals. Combining the Fisher information maximization principle and local event-triggered scheduling, the sampling integration time for each wavelength step can be adaptively allocated, increasing the sampling density at key wavelengths and saving time measurement resources in flat regions, thus balancing spectral resolution and measurement efficiency. Incident photon flux and detection efficiency can be corrected in real time during measurement, and the sampling integration time can be dynamically adjusted. The system can adapt to light source fluctuations, environmental changes, and differences in spectral characteristics, achieving more stable and reliable spectral measurements. Initializing the sampling integration time using an approximate allocation criterion and using it as a local scheduling baseline provides a simple and easy-to-implement engineering method while ensuring that the measurement results are close to the optimal allocation of Fisher information.

[0101] Example 2: Please refer to Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A method for implementing a single-photon spectrometer is provided, which specifically includes the following steps:

[0102] S1. Acquire broadband optical signals and perform wavelength division processing to calculate the photon count change rate between adjacent wavelengths; dynamically adjust the wavelength step according to the photon count change rate to achieve adaptive scanning of spectral changes and output monospectral optical signals.

[0103] S2. Convert the single-spectral optical signal into photon events through a preset array channel; within a preset time window, calculate the variance and Shannon entropy of each photon event to construct a local uncertainty index; perform spectral self-focusing and resampling based on the local uncertainty index to obtain photon counting data.

[0104] S3. Based on the photon counting data, establish a Poisson distribution statistical model for photon counting, and adaptively allocate timing resources according to the criterion of maximizing Fisher information. Calculate the optimal sampling time of the array channel, and simultaneously perform dead time and nonlinear correction to obtain the true photon count value.

[0105] S4. Based on the real photon count value, identify the dark count and background noise of the array channel, acquire the dark spectral template through no light input, and perform adaptive subtraction compensation in the measurement. The adaptive subtraction compensation is used to adaptively subtract photon events caused by dark count and background noise, realize joint noise reduction in the wavelength domain and time domain, and generate a net spectral count matrix.

[0106] S5. Based on the net spectral count matrix, plot the wavelength-power curve of the broadband optical signal and allow users to manually calibrate it.

[0107] Since the electronic device described in this embodiment is the electronic device used to implement the single-photon spectrometer implementation apparatus and method of this application embodiment, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the single-photon spectrometer implementation apparatus and method described in this application embodiment. Therefore, how the electronic device implements the method in this application embodiment will not be described in detail here. Any electronic device used by those skilled in the art to implement the single-photon spectrometer implementation apparatus and method of this application embodiment falls within the protection scope of this application.

[0108] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0109] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A device for implementing a single-photon spectrometer, characterized in that, include: The spectral wavelength scanning module acquires broadband optical signals, performs wavelength division processing, calculates the photon count change rate between adjacent wavelengths, dynamically adjusts the wavelength step according to the photon count change rate, realizes adaptive scanning for spectral changes, and outputs a monospectral optical signal. The self-focusing detection module converts single-spectral light signals into photon events through a preset array channel; within a preset time window, it calculates the variance and Shannon entropy of each photon event to construct a local uncertainty index; based on the local uncertainty index, it performs spectral domain self-focusing and resampling to obtain photon counting data. The main control analysis and calibration module establishes a Poisson distribution statistical model of photon counting based on photon counting data, and performs adaptive allocation of timing resources based on the maximization of Fisher information, calculates the optimal sampling time of the array channel, and performs dead time correction to obtain the true photon count value. The dark count compensation module identifies the dark count and background noise of the array channel based on the real photon count value. It acquires the dark spectral template through no light input and performs adaptive subtraction compensation during measurement. The adaptive subtraction compensation is used to adaptively subtract photon events caused by dark count and background noise, realizing joint noise reduction in the wavelength domain and time domain, and generating a net spectral count matrix. The human-computer interaction display module plots the wavelength-power curve of the broadband optical signal based on the net spectral counting matrix and allows users to manually calibrate it.

2. The apparatus for implementing a single-photon spectrometer according to claim 1, characterized in that, The method for acquiring broadband optical signals and performing wavelength division processing includes: The broadband optical signal emitted by the light source under test is input to the spectral wavelength scanning module via fiber optic coupling. After being modulated by a pre-optical filter, it is input to the beam splitter. The beam splitter includes a tunable optical filter, a Fabry-Perot resonator, or a grating, which is used to perform spectral separation on the input broadband optical signal and separate the light energy of different wavelength components to form a wavelength-splitting beam.

3. The apparatus for implementing a single-photon spectrometer according to claim 2, characterized in that, The method for obtaining the rate of change of photon count includes: The initial wavelength step is preset, driving the tunable optical filter to gradually change within the preset wavelength range, thereby outputting single-spectral optical signals with different center wavelengths in the time series; the photon count is recorded synchronously at each wavelength step point, and the rate of change of photon count between adjacent wavelength step points is calculated.

4. The apparatus for implementing a single-photon spectrometer according to claim 3, characterized in that, The method for outputting a single-spectral optical signal includes: Based on the photon count change rate, the next wavelength step is dynamically adjusted. The dynamically adjusted wavelength step is used to drive the tunable optical filter to scan to the next center wavelength. At the same time, the photon count value is recorded. The photon count change rate is repeatedly calculated and the wavelength step is adjusted until the scanning of the entire wavelength range is completed, and a single-spectrum optical signal is output.

5. The apparatus for implementing a single-photon spectrometer according to claim 4, characterized in that, The method for constructing the local uncertainty index includes: The single-spectral optical signal is converted into photon events through a preset array channel. Each photon event records the photon arrival time and the corresponding array channel number. Within a preset time window, the photon events of each array channel are counted to obtain the photon event count within the preset time window. The statistical variance of photon events counted by each array channel within a preset time window is calculated. The distribution of photon events within the preset time window is regarded as a probability distribution. Shannon entropy is calculated, and the variance of each array channel is weighted and fused with Shannon entropy to construct a local uncertainty index.

6. The apparatus for implementing a single-photon spectrometer according to claim 5, characterized in that, The method for acquiring the photon counting data includes: Based on the local uncertainty index calculated by each array channel within a preset time window, the single-spectrum optical signal is subjected to self-focusing processing in the spectral domain. The self-focusing processing includes a preset local uncertainty index threshold and adjusting the sampling density according to the preset local uncertainty index threshold. For wavelength regions where the local uncertainty index is greater than the preset local uncertainty index threshold, the sampling density is increased; for wavelength regions where the local uncertainty index is less than or equal to the preset local uncertainty index threshold, the sampling density is decreased. Based on the self-focusing processing results, the spectral signal is resampled. The resampling is achieved by adjusting the number of photon count samplings at each wavelength step point, and finally, the photon count data is obtained.

7. The apparatus for implementing a single-photon spectrometer according to claim 6, characterized in that, The method for obtaining the actual photon count value includes: Based on photon counting data, a Poisson distribution statistical model for photon counting is established. The Poisson distribution statistical model predetermines the expected photon count value at each wavelength step by the incident photon flux density, detection efficiency, and sampling integration time. The photon counting data follows a Poisson distribution. By maximizing Fisher information, the optimal sampling integration time for each wavelength step is determined under the preset total measurement time budget constraint. The sampling integration time is initialized by using an approximate allocation criterion and used as the baseline for event-triggered local scheduling. At the same time, array channel dead time correction is performed to obtain the true photon count value.

8. The apparatus for implementing a single-photon spectrometer according to claim 7, characterized in that, The method for generating the net spectral count matrix includes: Under the condition of not receiving broadband optical signals, dark photon events are collected through each array channel of the single photon detection linear array, and a dark spectral template is obtained through cumulative statistics. The real photon count value of each array channel is compared and analyzed with the dark spectral template, and the photon events caused by dark count and background noise are adaptively subtracted to obtain the noise-corrected net photon count value. Adaptive subtraction compensation is performed jointly in the wavelength and time domains. The net photon count values ​​after adaptive subtraction compensation for each array channel at each wavelength step point are organized into a matrix form to form a net spectral count matrix.

9. The apparatus for implementing a single-photon spectrometer according to claim 8, characterized in that, The method for plotting the wavelength-power curve of a broadband optical signal and allowing users to manually calibrate it includes: Based on the net spectral count matrix, a wavelength-power curve of the broadband optical signal is plotted with wavelength as the abscissa and power as the ordinate. The power is obtained by converting the net photon count value at each wavelength step point, combined with the photon energy and sampling integration time of the corresponding wavelength. The human-computer interaction display module also includes a manual calibration function, which allows users to adjust system parameters by providing an adjustable parameter interface and display the adjusted wavelength-power curve in real time.

10. A method for implementing a single-photon spectrometer, implemented using the apparatus for implementing a single-photon spectrometer as described in any one of claims 1 to 9, characterized in that, include: S1. Acquire broadband optical signals and perform wavelength division processing to calculate the photon count change rate between adjacent wavelengths; dynamically adjust the wavelength step according to the photon count change rate to achieve adaptive scanning of spectral changes and output monospectral optical signals. S2. Convert the single-spectral optical signal into photon events through a preset array channel; within a preset time window, calculate the variance and Shannon entropy of each photon event to construct a local uncertainty index; perform spectral self-focusing and resampling based on the local uncertainty index to obtain photon counting data. S3. Based on the photon counting data, establish a Poisson distribution statistical model for photon counting, and adaptively allocate timing resources according to the criterion of maximizing Fisher information. Calculate the optimal sampling time of the array channel, and simultaneously perform dead time and nonlinear correction to obtain the true photon count value. S4. Based on the real photon count value, identify the dark count and background noise of the array channel, acquire the dark spectral template through no light input, and perform adaptive subtraction compensation in the measurement. The adaptive subtraction compensation is used to adaptively subtract photon events caused by dark count and background noise, realize joint noise reduction in the wavelength domain and time domain, and generate a net spectral count matrix. S5. Based on the net spectral count matrix, plot the wavelength-power curve of the broadband optical signal and allow users to manually calibrate it.

Images