Signal pulse amplitude analysis method and system based on parallel double channels

Abstract

The invention discloses a signal pulse amplitude analysis method and system based on parallel dual channels, and relates to the technical field of energy spectrum measurement. The method comprises the steps that pulse signals are input into a fast forming processing channel and a slow forming processing channel in parallel; bipolar pinnacle forming processing is carried out on the pulse signals in the fast forming processing channel; based on the peak value and the time interval of the bipolar pinnacle pulse signal, judging whether the pulse is accumulated or not, and generating an accumulation mark signal or an accumulation discarding mark signal; trapezoidal forming processing is carried out on the pulse signals in the slow forming processing channel, forming parameters of trapezoidal pulses are dynamically adjusted according to the accumulation mark signals so as to separate the accumulation pulses, and amplitude extraction of the corresponding accumulation pulses is abandoned according to the accumulation judgment mark signals; and extracting the amplitude of the separated or processed trapezoidal pulse, and performing energy spectrum analysis and statistics to form an energy spectrum. The problem that the energy resolution and the detection efficiency of a single-channel trapezoidal forming algorithm are difficult to consider at the same time is solved.

Description

Technical Field

[0001] This invention relates to the field of energy spectrum measurement technology, and in particular to a method and system for analyzing signal pulse amplitude based on parallel dual channels. Background Technology

[0002] Nuclear radiation energy spectrum measurement technology is fundamental to nuclear technology applications, radioactive monitoring, resource exploration, and nuclear medicine. Its core principle is that sensors such as scintillator detectors convert the energy of incident particles into weak electrical pulse signals, the amplitude of which is proportional to the particle energy. Therefore, by accurately measuring the amplitude of the pulse signal and statistically analyzing its distribution (i.e., multichannel pulse amplitude analysis, MCA), an energy spectrum reflecting the characteristics of the nuclide can be obtained. In digital multichannel pulse amplitude analysis systems, the signal output from the detector, after being preamplified, is typically a single-sided negative exponential signal with a long decay time. Directly sampling such signals and searching for peaks to obtain the amplitude is susceptible to noise interference and requires extremely precise sampling timing, resulting in large errors. Therefore, digital pulse shaping filtering technology becomes crucial. Its purpose is to suppress noise while shaping the signal into a shape that facilitates accurate amplitude extraction, such as a bell shape (Gaussian shaping) or a trapezoidal shape (trapezoidal shaping). Among these, the trapezoidal shaping algorithm is widely used due to its good balance between signal-to-noise ratio, count rate performance, and implementation complexity. However, in high count rate measurement scenarios, radioactive events occur frequently, and the pulse signals may overlap in time, resulting in pulse stacking. For traditional filtering algorithms that use fixed forming parameters (such as trapezoidal forming with a fixed flat top width), pulse stacking causes two serious problems: 1) Amplitude extraction distortion: The waveform of the stacked pulses is distorted, causing the peak value or flat top height to not accurately reflect the energy of the original particles, resulting in energy spectrum peak position shift and resolution degradation; 2) False counts and missed counts: Severe stacking may lead to the identification of multiple particle events in a single trigger, or failure to trigger effectively, thus distorting the count statistics of the energy spectrum.

[0003] Therefore, developing a pulse amplitude analysis method that can adaptively handle pulse accumulation and maintain high measurement efficiency and high energy resolution at high count rates is an urgent problem to be solved. Summary of the Invention

[0004] To solve or alleviate one or more of the above-mentioned technical problems, the present invention provides a method and system for analyzing signal pulse amplitude based on parallel dual channels.

[0005] According to one aspect of the present invention, a method for analyzing signal pulse amplitude based on parallel dual channels is proposed, the method comprising the following steps: The pulse signal is acquired and converted from analog to digital. Then the converted pulse signal is input in parallel into the fast forming processing channel and the slow forming processing channel. In the rapid prototyping channel, the analog-to-digital converted pulse signal is subjected to bipolar tip forming processing by a bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges; Based on the peak value and time interval of the bipolar spiked pulse signal, it is determined whether pulses accumulate, and an accumulation flag signal or an accumulation rejection flag signal is generated. In the slow forming processing channel, the analog-to-digital converted pulse signal is subjected to trapezoidal forming processing by a trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top; Based on the stacking flag signal sent by the rapid prototyping processing channel, the forming parameters of the trapezoidal pulse are dynamically adjusted to separate the stacking pulse; based on the stacking rejection flag signal, the amplitude extraction of the corresponding stacking pulse is abandoned. The amplitude of the separated or processed trapezoidal pulses is extracted, and energy spectrum analysis and statistics are performed to form an energy spectrum. Further, the step of determining whether pulse accumulation occurs based on the peak value and time interval of the bipolar spiked pulse signal, and generating an accumulation flag signal or an accumulation rejection flag signal, includes: Set a stacked pulse recognition threshold and a stacked pulse rejection threshold, wherein the stacked pulse recognition threshold is greater than the stacked pulse rejection threshold; When the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals is less than the stacked pulse identification threshold, the stacking flag signal is generated; When the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals is less than the stacking pulse rejection threshold, the stacking rejection flag signal is generated. Furthermore, the determination of the effective bipolar spike pulse signal is as follows: when the amplitude of the bipolar spike pulse signal is greater than or equal to the effective signal determination amplitude threshold, the bipolar spike pulse signal is determined to be an effective bipolar spike pulse signal.

[0006] Furthermore, the transfer function of the trapezoidal forming process Obtained through the Z-transform method, the expression is: ; In the formula, z represents the variable in the complex frequency domain; Indicates the attenuation factor. , The sampling period is The time constant of a one-sided negative exponential signal; , , These represent the time when the trapezoid rises to the flat top, the time when the trapezoidal flat top ends, and the number of sampling points after discretization of the total forming time, respectively. These are the normalized integer representations of the corresponding time parameters under the sampling period.

[0007] Furthermore, the step of dynamically adjusting the forming parameters of the trapezoidal pulse to separate the stacking pulse based on the stacking flag signal sent by the rapid prototyping processing channel includes: adjusting the forming parameters of the trapezoidal pulse: rise time. Peace Peak Time , making or , It represents the minimum time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals.

[0008] Further, the extraction of the amplitude of the separated or processed trapezoidal pulse includes: In the flat-top segment of the trapezoidal pulse, the amplitudes of multiple sampling points are arithmetically averaged to obtain the flat-top amplitude; the background noise is obtained as the baseline value; the effective amplitude of the pulse signal is obtained by subtracting the baseline value from the flat-top amplitude. According to another aspect of the present invention, a parallel dual-channel signal pulse amplitude analysis system is proposed to implement the above-described parallel dual-channel signal pulse amplitude analysis method; the system includes: The signal acquisition module is used to acquire pulse signals and perform analog-to-digital conversion, and then input the converted pulse signals in parallel into the fast forming processing channel and the slow forming processing channel. The rapid prototyping processing channel is connected to the signal acquisition module. It is used to perform bipolar tip forming processing on the pulse signal after analog-to-digital conversion through the bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges. Based on the peak value and time interval of the bipolar tip pulse signal, it determines whether the pulses are stacked and generates a stacking flag signal or a stacking rejection flag signal. The slow-form processing channel is connected to the signal acquisition module and the fast-form processing channel. It is used to perform trapezoidal forming processing on the pulse signal after analog-to-digital conversion using a trapezoidal forming algorithm to generate trapezoidal pulses with flat tops. Based on the stacking flag signal sent by the fast-form processing channel, the flat top width parameter of the trapezoidal pulse is dynamically adjusted to separate the stacked pulses. Based on the stacking rejection flag signal, the amplitude extraction of the corresponding stacked pulse is abandoned. The amplitude extraction module, connected to the slow-forming processing channel, is used to extract the amplitude of the separated or processed trapezoidal pulse; The energy spectrum statistics module, connected to the amplitude extraction module, is used to perform energy spectrum analysis and statistics based on the extracted amplitudes to form an energy spectrum. Furthermore, the rapid prototyping processing channel includes: The bipolar tip forming unit is configured to generate a bipolar tip pulse signal based on delay, differential, integral and convolution operations on the input negative exponential signal; The stacking identification unit is configured to identify the peak value of a valid bipolar spike pulse signal and compare the time interval between adjacent peak values ​​with a preset threshold to generate a stacking flag signal and a stacking rejection flag signal.

[0009] Furthermore, the slow forming process channel includes: The adjustable trapezoidal forming unit is configured to perform trapezoidal forming processing on the analog-to-digital converted pulse signal through a trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top. The forming parameter control unit is configured to dynamically control the trapezoidal pulse forming parameters in the trapezoidal forming unit in response to the stacking flag signal sent by the fast forming process channel. Furthermore, the system is implemented based on an FPGA, the signal acquisition module includes an ADC converter, and the fast prototyping channel, slow prototyping channel, amplitude extraction module and energy spectrum statistics module are all implemented in hardware logic in the FPGA.

[0010] The embodiments of the present invention have the following technical effects: This invention proposes a parallel dual-channel signal pulse amplitude analysis method and system. Employing digital programming technology, this invention utilizes an FPGA chip to design a pulse signal amplitude analyzer based on fast and slow forming parallel dual channels. The main components include: in the fast forming channel, a bipolar peak forming algorithm is applied to form a single-sided negative exponential pulse signal, reducing the pulse width and improving the ability to distinguish stacked pulse peaks, which is beneficial for effectively distinguishing stacked pulses under high count rate conditions; in the fast forming channel, a fixed pulse discrimination method is applied to trigger a stacked pulse flag signal. If the peak amplitude after bipolar peak forming is greater than a set amplitude threshold, it is judged as a valid trigger; if the time interval between two valid triggers is less than the set stacked pulse threshold, a stacking flag signal is triggered; if the time interval between two valid triggers is less than the set stacked pulse threshold, a stacking flag signal is triggered; if the time interval between two valid triggers is less than the set stacked pulse threshold, a stacking flag signal is triggered. If the set stacking pulse rejection threshold is reached, a stacking rejection flag signal is triggered. In the slow forming channel, the width of the top of the pulse signal is expanded by a trapezoidal forming algorithm, overcoming the difficulty of accurately extracting the amplitude of a single-sided negative exponential pulse signal with a sharp top. In the slow forming channel, after receiving the stacking flag signal sent by the fast forming channel, the forming parameters are dynamically adjusted so that the flattening time of the trapezoidal signal is less than the time interval of the stacking pulses, thus achieving the separation of the stacking signal and improving the energy resolution of the system. In the slow forming channel, after receiving the stacking rejection flag signal sent by the fast forming channel, the trapezoidal forming of the stacking pulse signal is abandoned, reducing the interference of the stacking pulse on the amplitude extraction and improving the detection (counting) efficiency of the system.

[0011] This invention can effectively identify stacked pulses, dynamically adjust forming parameters, and discard stacked pulses under high count rate conditions, solving the problem that it is difficult to balance the energy (amplitude) resolution and detection (counting) efficiency of single-channel ladder forming algorithms. Attached Figure Description

[0012] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0013] Figure 1 This is a schematic diagram of multi-channel pulse amplitude analysis; Figure 2 This is a flowchart of a signal pulse amplitude analysis method based on parallel dual channels according to an embodiment of the present invention; Figure 3 This is another flowchart of a signal pulse amplitude analysis method based on parallel dual channels as described in an embodiment of the present invention; Figure 4 This is the timing diagram for AD9235 sampling control; Figure 5 This is a simulation diagram of the trapezoidal formation of a stacked unilateral negative exponential signal; Figure 6 This is a schematic diagram of the bipolar tip forming algorithm; Figure 7 This is a simulation diagram of bipolar tip formation; Figure 8 yes Trapezoidal shape rendering; Figure 9 yes Trapezoidal shape rendering; Figure 10 yes Trapezoidal shape rendering; Figure 11 This is a pulse trigger state transition diagram; Figure 12 This is a schematic diagram of a trapezoidal shape; Figure 13 It is a simulation diagram of a trapezoidal shape; Figure 14 This is the state transition diagram of the amplitude extraction module; Figure 15 This is the state transition diagram of the energy spectrum statistics module; Figure 16 These are simulation comparison images of single-channel and dual-channel forming under pulse stacking conditions; Figure 17These are the experimental results: the response energy spectra of Co-60 and Cs-137. Figure 18 These are the experimental results: I-131 and K-40 response energy spectra. Figure 19 This is the energy spectrum response diagram of the multi-nucleon in the experiment; Figure 20 This is a comparison of the energy spectra of Cs-137 dual-channel forming and single-channel ladder forming, based on experimental results. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are part of this invention.

[0015] Each time a scintillation detector detects a radioactive particle, it outputs a light signal, which is converted into an electrical pulse signal, representing a detected radioactive event. The amplitude of the electrical pulse signal output by the detector is linearly related to the energy of the detected radioactive particle; therefore, the amplitude distribution of the detector's output pulses can represent the energy distribution of the radioactive particle. By analyzing and statistically analyzing the amplitude of the electrical pulse signals output by the detector, the energy information of the radioactive particle can be obtained. The multichannel pulse amplitude analyzer (MCA) is the core of the energy spectrum measurement system. Traditional analog multichannel pulse amplitude analyzers use discrete components to linearly amplify, filter, shape, restore the baseline, and maintain the peak value of the detector's output pulse signal, finally obtaining the signal amplitude through an ADC (analog-to-digital converter). This approach suffers from problems such as long dead time, low count rate, susceptibility of discrete components to external environmental influences, and poor system linearity and stability. Furthermore, pulse accumulation severely affects the accuracy of amplitude extraction and the system's energy resolution. Compared to analog pulse shaping, digital pulse filtering and shaping can flexibly select appropriate shaping parameters based on different input signals, reducing the impact of external factors on system stability and minimizing the impact of pulse accumulation on the system's energy resolution. The core technology of a digital energy spectrum measurement system is multichannel pulse amplitude analysis, which involves acquiring the filtered and shaped pulse amplitude, storing the quantized amplitude in the corresponding channel address, and counting the value at each channel to ultimately form the energy spectrum. For example, a 12-bit resolution ADC (analog-to-digital converter) chip outputs 0~4095, therefore the maximum number of channels in a multichannel pulse amplitude analyzer using this ADC chip is 4096. That is, for each pulse signal with amplitude A acquired, the count at channel A is incremented by 1. The statistical distribution of counts at each channel represents the energy distribution of the radioactive particles detected by the detector. In practical applications, channels can be merged according to different measurement requirements and the energy resolution characteristics of different detectors. The specific steps of multichannel pulse amplitude analysis are as follows: Figure 1 As shown.

[0016] In high-count-rate measurement scenarios, radioactive events occur frequently, and the pulse signals may overlap in time, resulting in pulse stacking. For traditional filtering algorithms that use fixed forming parameters (such as trapezoidal forming with a fixed flat top width), pulse stacking causes two serious problems: amplitude extraction distortion and spurious and missed counts.

[0017] Therefore, this invention proposes a method for analyzing signal pulse amplitude based on parallel dual-channel signals, such as... Figure 2 As shown, the method includes the following steps: S1. Acquire pulse signals and perform analog-to-digital conversion, then input the converted pulse signals in parallel into the fast forming processing channel and the slow forming processing channel; S2. In the fast prototyping channel, the analog-to-digital converted pulse signal is subjected to bipolar tip forming processing by the bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges. S3. Based on the peak value and time interval of the bipolar spike pulse signal, determine whether the pulses are stacked, and generate a stacking flag signal or a stacking rejection flag signal. S4. In the slow forming processing channel, the pulse signal after analog-to-digital conversion is subjected to trapezoidal forming processing by the trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top; S5. Based on the stacking flag signal sent by the rapid prototyping processing channel, dynamically adjust the forming parameters of the trapezoidal pulse to separate the stacking pulse; based on the stacking rejection flag signal, abandon the extraction of the amplitude of the corresponding stacking pulse; S6. Extract the amplitude of the separated or processed trapezoidal pulses and perform energy spectrum analysis and statistics to form an energy spectrum. According to embodiments of the present invention, the key to achieving high-speed pulse signal processing lies in the parallel design of fast and slow forming dual channels. The fast forming channel employs a bipolar tip forming algorithm to effectively identify accumulated pulses, while the slow forming channel employs a trapezoidal forming algorithm to effectively extract the amplitude of the pulse signal. Specifically, as... Figure 3 As shown, the FPGA provides a 20MHz clock signal to the ADC (Analog-to-Digital Converter) to control the ADC to acquire pulse signals. The ADC converts the acquired pulse signals from analog to digital and sends them in parallel to the fast and slow forming channels. In the fast forming channel, the pulse signal is formed using a bipolar tip forming algorithm. If the time interval between two trigger signals is less than a set stacking pulse recognition threshold, a stacking flag signal is triggered; if the time interval between two trigger signals is less than a set stacking pulse rejection threshold, a stacking rejection flag signal is triggered. In the slow forming channel, the amplitude and baseline of the signal are obtained using a trapezoidal forming algorithm. Based on the stacking flag signal sent from the fast forming channel, the forming parameters are dynamically adjusted to accurately extract the amplitude of the pulse signal. Based on the stacking rejection flag signal, stacking pulses that cannot be effectively separated are discarded. Subsequently, the amplitude of the signal after trapezoidal forming is analyzed and statistically analyzed to form an energy spectrum, which is then communicated with the MCU (Microcontroller Unit) via a CAN bus. The FPGA implements the acquisition, processing, and transmission of pulse signal and energy spectrum data.

[0018] The embodiments of the present invention will be described in detail below.

[0019] First, in S1, pulse signals are acquired and analog-to-digital conversion is performed. Then, the converted pulse signals are input in parallel into the fast forming processing channel and the slow forming processing channel. According to an embodiment of the present invention, the ADC (e.g., AD9235) operates at a clock frequency of 20MHz. The clock signal is provided by a clock module. When the system (i.e., a complete digital hardware platform with an FPGA as the control and data processing center, including peripherals such as the ADC, clock network, and memory; the hereinafter referred to as the system) is powered on and reset to provide a clock signal to the ADC, the ADC begins sampling on the rising edge of the clock. The sampled data is buffered internally for 7 clock cycles and sent to the FPGA through a 12-bit parallel data line. The AD9235 sampling control timing diagram is shown below. Figure 4 As shown.

[0020] As an example, the system has two clock signals: a 48MHz clock signal provided by an external active crystal oscillator for normal FPGA operation and CAN communication module access; and a 20MHz clock signal generated by the digital clock management module (DCM) IP core for ADC acquisition, subsequent shaping algorithm implementation, energy spectrum statistics and other functional modules.

[0021] Then, in S2, in the rapid prototyping channel, the analog-to-digital converted pulse signal is processed by the bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges.

[0022] According to an embodiment of the present invention, radioactive particles scintillate in a detector, generating an optical signal. This optical signal is converted into an electrical signal by a photomultiplier tube, and then processed by analog circuits such as a preamplifier and filter to output a standard single-sided negative exponential pulse signal, characterized by a short rise time and a long fall time. Because the single-sided negative exponential pulse signal has a sharp top, it is difficult for an ADC to accurately acquire the signal amplitude. It must be filtered and shaped to expand the top width before the signal amplitude can be effectively obtained. Commonly used digital shaping methods include Gaussian shaping and trapezoidal shaping. In digital multichannel pulse amplitude analysis instruments, the trapezoidal shaping algorithm is the most widely used. Therefore, this embodiment of the present invention uses the trapezoidal shaping algorithm to extract the signal amplitude. This algorithm has the advantages of high signal-to-noise ratio and high energy resolution. Especially in the separation of stacked pulses, the trapezoidal shaping algorithm is superior to the Gaussian shaping algorithm. Trapezoidal shaping shapes the pulse signal into a trapezoidal shape, and the rise time, fall time, and top width are adjustable.

[0023] At high count rates, the probability of clustering between two adjacent pulse signals after trapezoidal processing increases, interfering with accurate amplitude extraction. Simulation results of pulse signal clustering after trapezoidal processing are shown below. Figure 5As shown in the diagram. Existing technologies typically employ methods such as reducing the width of the trapezoidal top and increasing the pulse time interval to separate accumulated pulses. However, even with a high count rate, pulse signal accumulation can still occur. If the accumulated pulse signals cannot be effectively identified, the signal amplitude cannot be accurately extracted, thus reducing the system's energy resolution.

[0024] Therefore, this embodiment of the invention proposes to further reduce the pulse width of the shaped single-sided negative exponential pulse signal using a bipolar peak shaping algorithm, thereby extending the time interval between the two pulse peaks and facilitating the effective resolution of stacked pulses. A schematic diagram of the bipolar peak shaping algorithm derivation is shown below. Figure 6 As shown. The specific process is as follows.

[0025] The input signal is a discrete one-sided negative exponential signal. The impulse signal can be obtained by delaying and differentiating the input signal. : (1) In the formula, n This represents the discrete-time series index, i.e., the nth sampling time. , The attenuation factor represents the proportion of a one-sided negative exponential signal that decays between two adjacent sampling points. The sampling period; The time constant of a one-sided negative exponential signal.

[0026] Impact signal The step signal is obtained by simple integration. : (2) Step signal Two impulse signals with different delays respectively After convolution and subtraction, a symmetrical rectangular signal is obtained. : (3) In equation (3), the parameter The duration of the rising and falling edges of the bipolar spike pulse is determined, where .

[0027] For symmetrical rectangular pulse signals Perform another integration operation to obtain the trapezoidal pulse signal. : (4) Step signal A single rectangular pulse signal is obtained by convolving it with an impulse signal that has a time delay difference. : (5) trapezoidal pulse signal With single rectangular pulse signal The difference is calculated to obtain the double-hump pulse signal. : (6) Finally, the double-hump pulse signal Perform one integration to obtain the bipolar spike pulse signal. : (7) The above derivation process is then organized to obtain the bipolar spire forming algorithm: (8) right Figure 5 The corresponding stacking pulse result is processed using the bipolar tip forming algorithm corresponding to equation (8), and the result is as follows: Figure 7 As shown, from Figure 7 As can be seen, the bipolar tip forming algorithm produces a narrower forming width.

[0028] Then, in S3, based on the peak value and time interval of the bipolar spike pulse signal, it is determined whether the pulses are stacked, and a stacking flag signal or a stacking rejection flag signal is generated.

[0029] According to an embodiment of the present invention, by analyzing the time interval between two adjacent bipolar peaks, the accumulation pulse can be effectively identified, triggering an accumulation flag signal or an accumulation rejection flag signal.

[0030] First, two unilateral negative exponential pulse signals are simulated in the simulation platform, with a sampling period of... The amplitudes are 500mV and 1000mV respectively, with a pulse interval of 500mV and 1000mV. The number of sampling points in the trapezoidal forming parameters, representing the time it takes for the trapezoid to rise to the top, is discretized. Number of sampling points after discretization of the trapezoidal flat-top end time Set to respectively , The simulation results are as follows Figure 8 As shown, when At this point, the forming pulses have completely accumulated and distorted, making it impossible to extract accurate amplitude. Keeping the trapezoidal forming parameters constant, only changing the pulse interval time... ,Right now The simulation results are as follows Figure 9 As shown, at this point, the pulses after trapezoidal formation exhibit some accumulation, but amplitude information can still be extracted relatively accurately. Changing the pulse interval time... ,Right now The simulation results are as follows Figure 10 As shown. At this point, the pulses after trapezoidal formation are completely separated, and amplitude information can be accurately extracted.

[0031] Therefore, in this embodiment of the invention, firstly, S31, a stacked pulse recognition time threshold and a stacked pulse rejection time threshold are set. As an example, the stacked pulse recognition threshold is: The threshold for discarding accumulated pulses is Then, S32, when the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals... When the value is less than the stacking pulse identification threshold, a stacking flag signal is generated; then, in S33, when the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals is less than the stacking pulse identification threshold, a stacking flag signal is generated; When the value is less than the stacking pulse rejection threshold, a stacking rejection flag signal is generated. Among them, the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals The acquisition process is as follows. A pulse trigger module is set in the rapid prototyping channel. The system's background noise is tested to obtain the baseline value, which is then set as the effective signal amplitude threshold. When there is no pulse signal or the amplitude of the pulse signal after peak formation is less than the effective signal amplitude threshold (i.e., the signal is drowned out by background noise), the signal is invalid, and the pulse trigger state machine is in the IDLE state. When the amplitude after peak formation is greater than or equal to the effective signal amplitude threshold, the process jumps to the maximum value acquisition and storage state S1. In state S1, the first acquired amplitude is saved to the maximum value register. Then, the amplitudes within the region where the peak pulse polarity is positive are temporarily stored and compared one by one. If the acquired amplitude is greater than the temporarily stored amplitude, the amplitude in the register is updated until the maximum value appears. When the maximum value appears, the state machine jumps to trigger validity S2. In state S2, the trigger validity flag is set, the maximum value register is cleared, and the process returns to the IDLE state to wait for the next trigger. The time interval between two valid signals obtained based on the above process is used as the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals. The pulse trigger state transition diagram is as follows: Figure 11 As shown.

[0032] Then, in S4, in the slow forming processing channel, the pulse signal after analog-to-digital conversion is subjected to trapezoidal forming processing by the trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top.

[0033] According to an embodiment of the present invention, a trapezoidal shaping algorithm is used to perform trapezoidal shaping processing on the pulse signal after analog-to-digital conversion. The trapezoid after pulse shaping is as follows: Figure 12 As shown. By Figure 12 It can be seen that a complete trapezoid It can be represented by the superposition of four piecewise functions, as shown in the following expression: (9) (10) In the formula, The time it takes for the trapezoidal ascent to reach the flat top. The end time of the trapezoidal flat top. The total width of the trapezoid. Trapezoidal amplitude, This represents a step signal; t represents time.

[0034] The transfer function of the ladder modeling algorithm is derived using the Z-transform method. The Z-transform method discretizes the continuous time-domain signal and performs signal processing in the complex frequency domain. For the ladder modeling algorithm, the transfer function can be derived from the Z-transform expressions of the input and output signals. This method transforms the time domain to the complex frequency domain, simplifying the solution process of differential equations and making it more suitable for digital system design and implementation. Specifically, let the system sampling period be... After performing a Z-transform on equation (10), we can obtain: (11) In the formula, z represents the variable in the complex frequency domain; , , These represent the time when the trapezoid rises to the flat top, the time when the trapezoidal flat top ends, and the number of sampling points after discretization of the total forming time, respectively. These are the normalized integer representations of the corresponding time parameters under the sampling period. , , and , , The relationship is as follows: (12) Combining equations (11) and (12), the output signal By performing the Z-transform, we can obtain: (13) For one-sided negative exponential signals By performing the Z-transform, we can obtain: (14) In the formula, By combining equations (13) and (14), the transfer function for the trapezoidal shape can be obtained: (15) Performing an inverse Z-transform on equation (15) yields the output. With input The time-domain relationship expression between them: (16) In the formula, Indicates the time-domain output signal. This represents the time-domain input signal. A set of ideal negative exponential signals is simulated and shaped using the trapezoidal shaping algorithm corresponding to equation (16). The shaping effect is as follows: Figure 13 As shown.

[0035] Then, in S5, the forming parameters of the trapezoidal pulse are dynamically adjusted according to the stacking flag signal sent by the rapid forming processing channel to separate the stacking pulse; according to the stacking rejection flag signal, the amplitude extraction of the corresponding stacking pulse is abandoned.

[0036] According to embodiments of the present invention, pulse stacking can severely affect the accuracy of amplitude extraction and the energy resolution of the system. Compared with analog pulse shaping technology, digital pulse filtering shaping algorithms such as trapezoidal shaping can flexibly select appropriate shaping parameters according to different input signals, effectively separate stacked pulses, and thus accurately extract pulse amplitudes. However, for pulse signals with severe stacking and distorted shapes after trapezoidal shaping, they must be discarded to ensure the system's energy resolution.

[0037] The fixed pulse discrimination method is used to dynamically adjust the trapezoidal pulse forming parameters. The basic principle is to adjust the rise time in the trapezoidal forming parameters. Flattening time The minimum time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals The separation of the stacked pulses is achieved by adjusting the relationship between the two parameters. Specifically, this is done by adjusting the rise time of the ladder forming circuit. Peace Peak Time ,when When this is achieved, effective separation of accumulated pulses can be realized; when It can still effectively separate accumulated pulses; when If this happens, the accumulated pulses must be discarded. Minimum time interval It is the minimum value among the time intervals between the peak values ​​of multiple adjacent valid bipolar spike pulse signals.

[0038] It should be noted that in practical applications, it is necessary to combine the specific analysis and testing object and dynamically adjust the corresponding parameters according to the characteristics of the input signal.

[0039] Then, in S6, the amplitude of the separated or processed trapezoidal pulse is extracted, and energy spectrum analysis and statistics are performed to form an energy spectrum.

[0040] According to an embodiment of the present invention, amplitude extraction is a crucial part of energy spectrum statistics in a digital multichannel system. Accurate extraction of the pulse signal amplitude can improve the energy resolution of the system. An amplitude extraction module is set after the slow-forming channel, and the state transition diagram of amplitude extraction is shown below. Figure 14As shown. When a pulse triggers, the state machine transitions from the IDLE state to the trapezoidal rising state S0. In state S0, the rising edge counter starts counting. When the rising edge count is full, it jumps to the trapezoidal flat-top state S2. If the count is not full, the state remains. If the stacking flag signal arrives, it jumps to the temporary storage state S1. In state S2, the flat-top counter starts counting. When the flat-top count is full, it jumps to the trapezoidal falling state S3. If the count is not full, the state remains. If the stacking flag signal arrives, it jumps to the temporary storage state S1. In state S3, the amplitude extraction valid flag is set. The arithmetic mean of all flat-top data is calculated and the baseline value (ground noise) is subtracted before being stored in the valid amplitude register. At the same time, the falling edge counter starts counting. When the falling edge count is full, it jumps to the IDLE state. In state S1, the temporary storage counter starts counting. When the count reaches the trapezoidal width, the count is full, and it jumps to the IDLE state; otherwise, the state remains.

[0041] After extracting the effective amplitude of the trapezoidal signal, the amplitude spectrum, i.e., the energy spectrum, can be obtained by statistically analyzing the effective amplitude. The energy spectrum can effectively reflect the relevant information and radioactivity level of nuclides. An energy spectrum statistics module is set up after the amplitude extraction module. As an example, the energy spectrum statistics is implemented using a true dual-port RAM IP core. The two ports of the true dual-port RAM can independently read and write to the BRAM. The design configuration is a read / write width of 32 bits, meaning that one address can hold 2^32 amplitude statistics data points. The read / write depth is configured to 1024, corresponding to 1024 digital channels. All read / write enable pins are brought out externally for easy data read / write control. In this embodiment, a 12-bit ADC is used, with a maximum value of 4095. The design uses 1024 data addresses; therefore, after amplitude extraction, the signal is shifted right by 2 bits before addressing, and the corresponding channel address count is incremented and updated. The state transition diagram of the energy spectrum statistics module is shown below. Figure 15 As shown. When the amplitude valid signal is triggered, the energy spectrum statistics module jumps from the IDLE state to the read address data state S1; otherwise, it remains in the IDLE state. In the S1 state, the processed trapezoidal amplitude signal is used as the read signal address, and the corresponding address data is read and incremented by one. After processing, the energy spectrum statistics module transitions from the read address data state to the write data state S2. In the S2 state, the data processed in S1 is written to the address corresponding to the read data. After writing is complete, it returns to the IDLE state.

[0042] Simulation comparison of single-channel and dual-channel shaping under pulse stacking conditions is shown below. Figure 16 High-radioactivity applications (count rate > 10) 5 In CPS (Constant Pulse Sensing), the probability of pulse signal accumulation in radiation detection systems is extremely high, such as... Figure 16 As shown in (a). Figure 16Figure (a) shows the original pulse signal acquired by the system, where two accumulation points occur: one with three pulses and the other with two pulses. If the traditional unipolar shaping method is used, the accumulated pulses cannot be accurately identified, resulting in a loss of count rate. The output signal is as follows: Figure 16 As shown in Figure (b). The output signal of the forming method proposed in this invention is as follows. Figure 16 As shown in Figure (c), the system can accurately identify and effectively separate the accumulated pulses, thereby improving the system's count rate and energy resolution.

[0043] The technical effects of the present invention were further verified through experiments.

[0044] The response energy spectra of four nuclides—Co-60, I-131, Cs-137, and K-40—in water obtained using the method of this invention are as follows: Figure 17 and Figure 18 As shown. Figure 17 The left-middle figure shows the energy spectrum response of Co-60, where the characteristic peaks of 1.17 MeV and 1.33 MeV produced by its decay can be observed. There is also a backscattering peak with an energy of 214 keV on the left. Figure 17 The right-middle figure shows the energy spectrum response of Cs-137, where a photoelectric peak of 662 keV generated by its decay can be seen, and the left-middle figure shows a backscattering peak with an energy of about 198 keV. Figure 18 The left-middle figure shows the energy spectrum response of I-131, where the photoelectric peaks of 364 keV and 636 keV generated by its decay can be seen. Figure 18 The right-middle figure shows the energy spectrum response of K-40, revealing the characteristic peak of 1.46 MeV produced by its decay. The Compton plateau is evident in the energy spectrum responses of the four nuclides, and significant scattering occurs in the low-energy region. For example... Figure 19 As shown, the characteristic peaks of the four nuclides Co-60, I-131, Cs-137, and K-40 obtained using the method of the present invention are clearly visible.

[0045] The linearity of a system is an important indicator for evaluating an energy spectrum measurement system, reflecting the correlation between the amplitude of the input pulse signal and the energy spectrum channel address. The linear regression equation is as follows: ,in The amplitude of the input signal. The center channel address of the corresponding energy spectrum; fitting coefficients =0.9999. The experiment used a signal generator to simulate negative exponential pulse signals ranging from -100mV to -3000mV at 100ms intervals. The center channel address of the energy spectrum under different input amplitude pulse signals was statistically analyzed using this invention. The experimental results verify that the linearity of this invention is good and meets the design requirements.

[0046] System stability is an evaluation of the ability of a system's key parameters to remain constant over time. System stability can be represented by the relative deviation Sy of the measured parameter (or related parameters or related intermediate parameters) over time; the smaller the Sy value, the better the system stability. The formula for calculating the relative deviation Sy is as follows: (17) In the formula, The deviation is relative and expressed as a percentage (%). The maximum value from multiple measurements. This is the minimum value obtained from multiple measurements.

[0047] To test the stability of the system under long-term measurement conditions, the detector was placed in a lead chamber at room temperature, and the 137-Cs radioactive source was measured for 10 minutes every 2 hours for 16 consecutive hours. The changes in the center location of the full-energy peak and the area of ​​the full-energy peak are shown in Table 1.

[0048] Table 1. Results of pulse counting and full-energy peak stability tests As shown in Table 1 and Equation 17, the average count rate of 137-Cs is 2452, with a relative deviation of 0.122%; the total peak address is stable at around 384 channels, with a maximum deviation of no more than 2 channels, and a relative deviation of 0.174%; the average area of ​​the total peak is 476706, with a relative deviation of 0.131%, indicating good stability of the system count rate and the total peak address.

[0049] Under the conditions of a highly radioactive Cs-137 source, comparative experiments were conducted on the energy spectrum of Cs-137 using both dual-channel shaping and single-channel ladder shaping algorithms. The measurement duration was 30 seconds. This verified the applicability of the fast-slow dual-channel shaping algorithm, which combines ladder shaping and bipolar tip shaping, under conditions of high count rate and severe pulse accumulation. The test results are as follows: Figure 20 As shown, under the dual-channel shaping algorithm, the total system count reaches 2,495,231, the CPS is 83,174, and the Cs-137 full-energy peak energy resolution is 8.57%. Under the single-channel trapezoidal shaping algorithm, the total system count is 2,438,763, the CPS is 81,292, and the Cs-137 full-energy peak energy resolution is 10.89%. Compared to the single-channel trapezoidal shaping algorithm, the dual-channel shaping algorithm can provide higher energy resolution.

[0050] This invention also proposes a parallel dual-channel signal pulse amplitude analysis system, which is used to implement the parallel dual-channel signal pulse amplitude analysis method described in the above embodiments; the system includes: The signal acquisition module is used to acquire pulse signals and perform analog-to-digital conversion, and then input the converted pulse signals in parallel into the fast forming processing channel and the slow forming processing channel. The rapid prototyping processing channel is connected to the signal acquisition module. It is used to perform bipolar tip forming processing on the pulse signal after analog-to-digital conversion through the bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges. Based on the peak value and time interval of the bipolar tip pulse signal, it determines whether the pulses are stacked and generates a stacking flag signal or a stacking rejection flag signal. The slow forming processing channel is connected to the signal acquisition module and the fast forming processing channel. It is used to perform trapezoidal forming processing on the pulse signal after analog-to-digital conversion through the trapezoidal forming algorithm to generate trapezoidal pulses with flat tops; according to the stacking flag signal sent by the fast forming processing channel, the forming parameters of the trapezoidal pulses are dynamically adjusted to separate the stacked pulses; according to the stacking rejection flag signal, the amplitude extraction of the corresponding stacked pulses is abandoned. The amplitude extraction module, connected to the slow-forming processing channel, is used to extract the amplitude of the separated or processed trapezoidal pulse; The energy spectrum statistics module, connected to the amplitude extraction module, is used to perform energy spectrum analysis and statistics based on the extracted amplitudes to form an energy spectrum. In this embodiment, optionally, the rapid prototyping processing channel includes: The bipolar tip forming unit is configured to generate a bipolar tip pulse signal based on delay, differential, integral and convolution operations on the input negative exponential signal; The stacking identification unit is configured to identify the peak value of a valid bipolar spike pulse signal and compare the time interval between adjacent peak values ​​with a preset threshold to generate a stacking flag signal and a stacking rejection flag signal.

[0051] In this embodiment, optionally, the slow forming process channel includes: The adjustable trapezoidal forming unit is configured to perform trapezoidal forming processing on the analog-to-digital converted pulse signal through a trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top. The forming parameter control unit is configured to dynamically control the trapezoidal pulse forming parameters in the trapezoidal forming unit in response to the stacking flag signal sent by the fast forming process channel. In this embodiment, optionally, the system is implemented based on an FPGA, the signal acquisition module includes an ADC converter, and the fast prototyping channel, slow prototyping channel, amplitude extraction module and energy spectrum statistics module are all implemented in hardware logic in the FPGA.

[0052] The function of the signal pulse amplitude analysis system based on parallel dual channels described in this embodiment of the invention can be explained by the aforementioned signal pulse amplitude analysis method based on parallel dual channels. Therefore, for the parts not described in detail in the system embodiment, please refer to the above method embodiment, and they will not be repeated here.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing signal pulse amplitude based on parallel dual channels, characterized in that, Includes the following steps: The pulse signal is acquired and converted from analog to digital. Then the converted pulse signal is input in parallel into the fast forming processing channel and the slow forming processing channel. In the rapid prototyping channel, the analog-to-digital converted pulse signal is subjected to bipolar tip forming processing by a bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges; Based on the peak value and time interval of the bipolar spiked pulse signal, it is determined whether pulses accumulate, and an accumulation flag signal or an accumulation rejection flag signal is generated. In the slow forming processing channel, the analog-to-digital converted pulse signal is subjected to trapezoidal forming processing by a trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top; Based on the stacking flag signal sent by the rapid prototyping channel, the forming parameters of the trapezoidal pulse are dynamically adjusted to separate the stacking pulses; Based on the stacking rejection flag signal, the amplitude extraction of the corresponding stacking pulse is abandoned; The amplitude of the separated or processed trapezoidal pulses is extracted, and energy spectrum analysis and statistics are performed to form an energy spectrum.

2. The method for analyzing signal pulse amplitude based on parallel dual channels according to claim 1, characterized in that, The step of determining whether pulse accumulation occurs based on the peak value and time interval of the bipolar spiked pulse signal, and generating an accumulation flag signal or an accumulation rejection flag signal, includes: Set a stacked pulse recognition threshold and a stacked pulse rejection threshold, wherein the stacked pulse recognition threshold is greater than the stacked pulse rejection threshold; When the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals is less than the stacked pulse identification threshold, the stacking flag signal is generated; When the time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals is less than the stacking pulse rejection threshold, the stacking rejection flag signal is generated.

3. The method for analyzing signal pulse amplitude based on parallel dual channels according to claim 2, characterized in that, The determination of a valid bipolar spike pulse signal is as follows: when the amplitude of the bipolar spike pulse signal is greater than or equal to the valid signal determination amplitude threshold, the bipolar spike pulse signal is determined to be a valid bipolar spike pulse signal.

4. The signal pulse amplitude analysis method based on parallel dual channels according to claim 1, characterized in that, The transfer function of the trapezoidal forming process Obtained through the Z-transform method, the expression is: ； In the formula, z represents the variable in the complex frequency domain; Indicates the attenuation factor. , The sampling period is The time constant of a one-sided negative exponential signal; , , These represent the time when the trapezoid rises to the flat top, the time when the trapezoidal flat top ends, and the number of sampling points after discretization of the total forming time, respectively. These are the normalized integer representations of the corresponding time parameters under the sampling period.

5. The method for analyzing signal pulse amplitude based on parallel dual channels according to claim 4, characterized in that, The step of dynamically adjusting the forming parameters of the trapezoidal pulse to separate the stacking pulse based on the stacking flag signal sent by the rapid prototyping channel includes: adjusting the forming parameters of the trapezoidal pulse: rise time Peace Peak Time , making , It represents the minimum time interval between the peak values ​​of two adjacent valid bipolar spike pulse signals.

6. The signal pulse amplitude analysis method based on parallel dual channels according to claim 1, characterized in that, The extraction of the amplitude of the separated or processed trapezoidal pulse includes: In the flat-top segment of the trapezoidal pulse, the amplitudes of multiple sampling points are arithmetically averaged to obtain the flat-top amplitude; the background noise is obtained as the baseline value; the effective amplitude of the pulse signal is obtained by subtracting the baseline value from the flat-top amplitude.

7. A signal pulse amplitude analysis system based on parallel dual channels, characterized in that, A system for implementing the parallel dual-channel signal pulse amplitude analysis method according to any one of claims 1-6; the system comprises: The signal acquisition module is used to acquire pulse signals and perform analog-to-digital conversion, and then input the converted pulse signals in parallel into the fast forming processing channel and the slow forming processing channel. The rapid prototyping processing channel is connected to the signal acquisition module. It is used to perform bipolar tip forming processing on the pulse signal after analog-to-digital conversion through the bipolar tip forming algorithm to generate a bipolar tip pulse signal with steep leading and trailing edges. Based on the peak value and time interval of the bipolar tip pulse signal, it determines whether the pulses are stacked and generates a stacking flag signal or a stacking rejection flag signal. The slow-form processing channel is connected to the signal acquisition module and the fast-form processing channel. It is used to perform trapezoidal forming processing on the pulse signal after analog-to-digital conversion using a trapezoidal forming algorithm to generate trapezoidal pulses with flat tops. Based on the stacking flag signal sent by the fast-form processing channel, the flat top width parameter of the trapezoidal pulse is dynamically adjusted to separate the stacked pulses. Based on the stacking rejection flag signal, the amplitude extraction of the corresponding stacked pulse is abandoned. The amplitude extraction module, connected to the slow-forming processing channel, is used to extract the amplitude of the separated or processed trapezoidal pulse; The energy spectrum statistics module, connected to the amplitude extraction module, is used to perform energy spectrum analysis and statistics based on the extracted amplitudes to form an energy spectrum.

8. A signal pulse amplitude analysis system based on parallel dual channels according to claim 7, characterized in that, The rapid prototyping process channel includes: The bipolar tip forming unit is configured to generate a bipolar tip pulse signal based on delay, differential, integral and convolution operations on the input negative exponential signal; The stacking identification unit is configured to identify the peak value of a valid bipolar spike pulse signal and compare the time interval between adjacent peak values ​​with a preset threshold to generate a stacking flag signal and a stacking rejection flag signal.

9. A signal pulse amplitude analysis system based on parallel dual channels according to claim 8, characterized in that, The slow forming process channel includes: The adjustable trapezoidal forming unit is configured to perform trapezoidal forming processing on the analog-to-digital converted pulse signal through a trapezoidal forming algorithm to generate a trapezoidal pulse with a flat top. The forming parameter control unit is configured to dynamically control the trapezoidal pulse forming parameters in the trapezoidal forming unit in response to the stacking flag signal sent by the fast forming process channel.

10. A signal pulse amplitude analysis system based on parallel dual channels according to claim 9, characterized in that, The system is implemented based on FPGA. The signal acquisition module includes an ADC converter. The fast prototyping channel, slow prototyping channel, amplitude extraction module and energy spectrum statistics module are all implemented in hardware logic in the FPGA.

Images