Particle detector signal pulse shape analysis method based on hann filter and differential algorithm

Abstract

This invention relates to the field of particle detection technology and proposes a method for analyzing the pulse shape of particle detector signals based on Hanning filtering and differential algorithms. This method addresses the problem of detector signals being susceptible to high-frequency noise and low-frequency oscillation interference. It employs a Hanning filter function to eliminate high-frequency noise and utilizes a differential algorithm to cancel baseline low-frequency oscillations, significantly improving the signal-to-noise ratio, enhancing time resolution, and reducing amplitude measurement errors. Simultaneously, a triple screening mechanism—setting a positive-negative threshold crossover order, an upper limit limit on the threshold interval, and a positive-negative extreme value ratio threshold—accurately distinguishes between real particle signals and noise pulses, significantly reducing the false positive rate. Furthermore, this invention has low computational complexity, is applicable to different types of detectors and signal pulses ranging from nanoseconds to microseconds, meets the requirements of high count rate experiments, and improves the accuracy of experimental research results.

Description

Technical Field

[0001] This invention relates to the field of particle detection technology, specifically to a method for analyzing the pulse shape of particle detector signals based on Hanning filtering and differential algorithms. Background Technology

[0002] In the field of particle detection technology, the signal characteristics (such as time and amplitude) output by the detector are the core data of experimental research, and their accuracy directly affects the reliability of the physical analysis results. However, the detector signal is susceptible to high-frequency noise interference (such as electronic noise and environmental electromagnetic interference) and low-frequency baseline oscillations during the acquisition process, which leads to signal distortion and reduces the accuracy of characteristic parameter extraction.

[0003] In existing technologies, detector signal pulse shape analysis mainly relies on the summation-averaging method to suppress noise and uses simple peak-finding algorithms to determine signal characteristics. For example, traditional methods reduce random noise by averaging multiple samples and then use threshold triggering or derivative extrema to locate signal peaks. However, such methods have significant drawbacks:

[0004] Insufficient high-frequency noise suppression: The summation and averaging method has limited effectiveness in eliminating periodic or sudden high-frequency noise, and residual noise will still cause errors in signal time and amplitude measurement;

[0005] Low-frequency oscillation interference: Baseline drift or low-frequency interference can cause peak finding algorithms to misjudge the starting point or peak position of the signal;

[0006] The selection criteria are too simplistic: the existing threshold judgment lacks in-depth analysis of the signal morphology, which can easily lead to noise pulses being misjudged as valid signals, thus reducing the reliability of the data.

[0007] In practical applications, such as fission ionization chambers or silicon detectors, signal pulse widths may be only a few nanoseconds to microseconds, making high-frequency noise and baseline fluctuations easily mask the true characteristics. If traditional methods are used, the time resolution may decrease by more than 30%, and the amplitude measurement error may exceed 15%, severely restricting the needs of high-precision physical experiments.

[0008] Therefore, there is an urgent need for an analysis method that can simultaneously eliminate high-frequency noise and low-frequency oscillations and accurately select effective pulses based on signal morphology, so as to improve the accuracy of detector signal feature extraction. Summary of the Invention

[0009] To address the aforementioned issues, this invention proposes a pulse shape analysis method based on Hanning filtering and differential algorithms. This method is applicable to the output signal processing of particle detection equipment such as fission ionization chambers (gas detectors), silicon detectors (semiconductor detectors), and cesium iodide detectors (scintillator detectors). The aim is to filter noise and accurately extract signal feature information (such as time and amplitude) through digital signal processing technology, thereby improving the accuracy of experimental research results.

[0010] The technical solution adopted in this invention is: a method for analyzing the pulse shape of particle detector signals based on Hanning filtering and differential algorithms, comprising the following steps:

[0011] S1. The output signal of the detector is completely acquired and recorded through the analog-to-digital conversion data acquisition card, and the analog signal is digitized. The sampling rate of the data acquisition card is 1 GHz and the precision is 12 bits.

[0012] S2. Design an algorithm to filter the high-frequency noise in the detector signal and then differentiate it to eliminate the low-frequency oscillation of the detector baseline while preserving the complete information of the original detector signal.

[0013] S3. Judge and filter the filtered and differentiated signal to select the true detector signal; S4. Extract the feature values ​​from the filtered and differentiated signal to obtain the corresponding time and amplitude information in the original detector signal.

[0014] The design algorithm steps specifically include:

[0015] First, assume that the function expression of the detector signal is s(t), which is a function of time t;

[0016] Then, the low-pass filter function to eliminate high-frequency noise is selected as f(t), whose Fourier transform in the frequency domain is a low-pass filter function;

[0017] Finally, the original signal s(t) is convolved with the filter function f(t) and then differentiated to obtain the filtered and differentiated signal h(t).

[0018] A method for analyzing the pulse shape of particle detector signals based on Hanning filtering and differential algorithms includes the following:

[0019] First, assume the detector signal is expressed as s(t), which is a function of time t; then, the low-pass filter function for eliminating high-frequency noise is f(t), whose Fourier transform in the frequency domain is also a low-pass filter function; convolving the original signal with the filter function and then differentiating it yields h(t):

[0020] h(t)=(f(t)*s(t)), (1)

[0021] Based on the properties of convolution, we can obtain:

[0022] h(t)=f'(t)*s(t), (2)

[0023] According to formulas (2) and (3), the filtered and differentiated signal h(t) can be obtained as follows:

[0024]

[0025] The steps for judging and filtering the filtered and differentiated signal specifically include: for a unipolar negative signal, setting two positive and negative thresholds with equal absolute values; only when the h(t) signal passes through the two thresholds in the order of negative threshold-negative threshold-positive threshold-positive threshold, the original signal is determined to be a valid signal.

[0026] The steps of judging and filtering the filtered and differentiated signal further include: setting an upper threshold for the time interval between the second crossing of the negative threshold and the first crossing of the positive threshold, and setting an upper threshold for the ratio of the absolute values ​​of the positive and negative extreme values, for further filtering.

[0027] The steps for extracting feature values ​​from the filtered and differentiated signal specifically include: selecting the time corresponding to the negative extreme value of the differentiated signal h(t) as the time information of the original signal; and selecting the sum of the absolute values ​​of the positive and negative extreme values ​​of h(t) as the amplitude information of the original signal.

[0028] The method is applicable to the shape analysis of the output signals of fission ionization chambers (gas detectors), silicon detectors (semiconductor detectors), and cesium iodide detectors (scintillator detectors).

[0029] The method is applicable to fission ionization chambers as gas detectors, silicon detectors as semiconductor detectors, and cesium iodide detectors as scintillator detectors.

[0030] The signal filtering and differentiation techniques are based on Fourier transform, using Hanning filter function and combined with differentiation algorithm to eliminate high-frequency noise and low-frequency oscillations in the detector signal, while completely preserving the original information of the signal.

[0031] The signal judgment and screening conditions based on the filtering differential algorithm include detector signal screening criteria such as threshold judgment, threshold crossing order analysis, signal width screening, and comparison of the absolute values ​​of positive and negative extreme values.

[0032] The particle detector signal pulse shape analysis method proposed in this invention, based on Hanning filtering and differential algorithms, significantly improves the accuracy and reliability of detector signal feature extraction through innovative signal processing techniques. Specific beneficial effects are as follows:

[0033] First, this invention employs the Hanning filter function combined with a differential algorithm to precisely design low-pass filter characteristics in the frequency domain. This effectively eliminates high-frequency noise (such as electronic noise and electromagnetic interference) in the detector signal, while simultaneously offsetting low-frequency baseline oscillations (such as baseline drift in fission ionization chamber signals) through differential operations. Experiments show that, compared to the traditional summation and averaging method, this invention can improve the signal-to-noise ratio by more than 40%, reduce the time resolution error to within 2%, and control the amplitude measurement error to within 5%, significantly improving signal fidelity.

[0034] Secondly, this invention employs a triple screening mechanism—combining a positive / negative threshold crossover order, an upper limit limit on the threshold interval, and a positive / negative extreme value ratio threshold—to accurately distinguish between real particle signals and noise pulses. For example, in silicon detector applications, this method reduces the false positive rate from 15% of traditional methods to below 2%, while ensuring a 100% effective signal capture rate, significantly improving the reliability of experimental data.

[0035] Meanwhile, this invention avoids the dependence of traditional peak-finding algorithms on signal morphology by selecting the time corresponding to the negative extreme value of the differential signal as the original signal time and the sum of the absolute values ​​of the positive and negative extreme values ​​as the amplitude. This method is applicable to different types of detectors such as fission ionization chambers (gas detectors), silicon detectors (semiconductor detectors), and cesium iodide detectors (scintillator detectors), and can work stably in the range of signal pulse widths from nanoseconds to microseconds.

[0036] Finally, this invention, based on the convolution operation of Fourier transform and combined with the rapid decay characteristic of the Hanning window function, significantly reduces computational complexity. It meets the analytical requirements of high-count-rate experiments (such as nuclear physics and high-energy physics experiments) under conditions of a 1GHz sampling rate and 12-bit precision. Attached Figure Description

[0037] Figure 1 This is a flowchart of the signal analysis and processing in this invention;

[0038] Figure 2 This is the original signal diagram in this invention;

[0039] Figure 3 This is a signal diagram after filtering and differentiation of the original signal in this invention. Detailed Implementation

[0040] The particle detector signal pulse shape analysis method based on Hanning filtering and differential algorithm of the present invention mainly includes signal acquisition, filtering and differential processing, signal judgment and screening, and feature value extraction. The specific implementation methods of each step are described in detail below:

[0041] S1. Signal Acquisition: A high-sampling-rate, high-precision analog-to-digital converter (ADC) data acquisition card (1 GHz sampling rate, 12-bit precision) is used to completely acquire and record the detector's output signal, digitizing the analog signal. This step ensures the integrity and high precision of the original signal, providing a reliable data foundation for subsequent processing.

[0042] S2 Filtering and Differentiation Processing: The algorithm is designed by first assuming the detector signal has a functional expression s(t), which is a function of time t. The Hanning filter function is chosen as the low-pass filter function f(t), whose Fourier transform in the frequency domain is a low-pass filter function, effectively eliminating high-frequency noise. The original signal s(t) is convolved with the filter function f(t), followed by differentiation to obtain the filtered and differentiated signal h(t). This step, through Hanning filtering and differentiation, simultaneously eliminates high-frequency noise and low-frequency baseline oscillations, while preserving the complete information of the original detector signal.

[0043] S3 Signal Judgment and Filtering: For unipolar negative signals, two positive and negative thresholds with equal absolute values ​​are set. The original signal is considered valid only when the h(t) signal crosses both thresholds in the order of negative threshold-negative threshold-positive threshold-positive threshold. An upper threshold is set for the time interval between the second crossing of the negative threshold and the first crossing of the positive threshold, and an upper threshold is set for the ratio of the absolute values ​​of the positive and negative extremes, to further filter the signal and ensure its accuracy and reliability.

[0044] S4 Feature Extraction: The time corresponding to the negative extreme value of the differential signal h(t) is selected as the time information of the original signal. The sum of the absolute values ​​of the positive and negative extreme values ​​of h(t) is selected as the amplitude information of the original signal. This step, by extracting key feature values, obtains the corresponding time and amplitude information in the detector's original signal.

[0045] Example: Pulse Shape Analysis of Fission Ionization Chamber Signals Based on Hanning Filter and Differential Algorithm

[0046] like Figure 1-3 As shown, where Figure 1 This is a flowchart of the signal analysis and processing in this invention, illustrating the complete process from signal acquisition to feature extraction, including key steps such as filtering and differentiation, signal judgment and screening, providing clear process guidance for implementing this invention. In nuclear physics experiments, the output signal of the fission ionization chamber, used as a gas detector, is susceptible to high-frequency noise and low-frequency baseline oscillations. The particle detector signal pulse shape analysis method based on Hanning filtering and differentiation algorithms of this invention is specifically implemented as follows:

[0047] First, signal acquisition is performed: using an analog-to-digital converter data acquisition card with a 1GHz sampling rate and 12-bit precision, the output signal of the fission ionization chamber is fully acquired and recorded, and the analog signal is digitized.

[0048] Next, filtering and differentiation are performed: Assume the functional expression of the fission ionization chamber signal is s(t). The Hanning filter function f(t) is selected as the low-pass filter function, and its Fourier transform is designed to have low-pass characteristics in the frequency domain. The original signal s(t) is convolved with the filter function f(t), and then differentiated to obtain the filtered and differentiated signal h(t). At this point, high-frequency noise and baseline low-frequency oscillations are effectively eliminated, and the signal fidelity is significantly improved.

[0049] More specifically, by convolving the original signal with the filter function and then differentiating it, we can obtain h(t):

[0050] h(t)=(f(t)*s(t)), (1)

[0051] Based on the properties of convolution, we can obtain:

[0052] h(t)=f'(t)*s(t), (2)

[0053] The Hanning filter function is selected:

[0054]

[0055] Where ρ represents the normalization coefficient, and w is the full width at half maximum (FWHM) of the signal. According to formulas (2) and (3), the filtered and differentiated signal h(t) can be obtained as follows:

[0056]

[0057] Then, signal judgment and filtering are performed: For the unipolar negative signal output from the fission ionization chamber, two positive and negative thresholds with equal absolute values ​​are set. When the h(t) signal crosses the threshold in the order of negative threshold-negative threshold-positive threshold-positive threshold, it is determined to be a valid signal. An upper threshold is set for the time interval between the second crossing of the negative threshold and the first crossing of the positive threshold, and the ratio of the absolute values ​​of the positive and negative extremes is compared for further filtering to ensure the accuracy and reliability of the signal.

[0058] Finally, feature value extraction is performed: the time corresponding to the negative extreme value of h(t) is selected as the original signal time information. The sum of the absolute values ​​of the positive and negative extreme values ​​of h(t) is selected as the original signal amplitude information.

[0059] like Figure 2 and 3 As shown, where Figure 2This is the original signal diagram in this invention, showing the original signal waveform output from the fission ionization chamber, which includes high-frequency noise and baseline low-frequency oscillations, providing basic data for subsequent processing. Figure 3 This is a waveform of the original signal after filtering and differentiation in this invention. The waveform shows the signal after Hanning filtering and differentiation. Compared to the original signal, the filtered signal is smoother, high-frequency noise and low-frequency baseline oscillations are effectively eliminated, and signal characteristics are more pronounced, providing favorable conditions for subsequent signal judgment and screening, and feature value extraction.

Claims

1. A method for analyzing the pulse shape of a signal from a particle detector based on the Hann filter and the differential algorithm, characterized in that, Includes the following steps: S1. Signal Acquisition: Using an analog-to-digital converter data acquisition card with a sampling rate of 1GHz and a precision of 12 bits, the output signal of the particle detector is fully acquired and recorded, and the analog signal is digitized. S2. Filtering and Differentiation Processing: Assuming the function expression of the detector signal is s(t), the Hanning filter function is selected as the low-pass filter function f(t), whose Fourier transform in the frequency domain exhibits low-pass filtering characteristics. The original signal s(t) and the Hanning filter function f(t) are first convolved, and then differentiated to obtain the filtered and differentiated signal h(t), thereby simultaneously eliminating high-frequency noise and low-frequency baseline oscillations in the signal while preserving the complete information of the original signal. S3. Signal Judgment and Filtering: For unipolar negative signals, set two positive and negative thresholds with equal absolute values; only when the h(t) signal meets the negative threshold... negative threshold Positive threshold When the positive threshold passes through two thresholds in sequence, it is determined to be a valid particle detector signal; at the same time, an upper threshold is set for the time interval between the second crossing of the negative threshold and the first crossing of the positive threshold, and an upper threshold is set for the ratio of the absolute values ​​of the positive and negative extreme values, for secondary screening; S4. Feature extraction: Select the time corresponding to the negative extreme value of the differential signal h(t) as the time information of the original signal; select the sum of the absolute values ​​of the positive and negative extreme values ​​of h(t) as the amplitude information of the original signal.

2. The particle detector signal pulse shape analysis method based on Hanning filtering and differential algorithm according to claim 1, characterized in that, In the filtering and differentiation steps, the filtered and differentiated signal h(t) is obtained according to the following formula: ， Based on the properties of convolution, it is transformed as follows: ， The Hanning filter function f(t) is: , ρ is the normalization coefficient, and w is the full width at half maximum (FWHM) of the signal; The final result is: 。 3. The particle detector signal pulse shape analysis method based on Hanning filtering and differential algorithm according to claim 1, characterized in that, The particle detector is a fission ionization chamber, a silicon detector, or a cesium iodide detector.

4. The particle detector signal pulse shape analysis method based on Hanning filtering and differential algorithm according to claim 1, characterized in that, The method is applicable to particle detector signal analysis with pulse widths ranging from nanoseconds to microseconds.

Images