阅读说明：本技术 一种核脉冲信号处理方法及系统 (Nuclear pulse signal processing method and system ) 是由 周建斌 喻杰 万文杰 于 2020-06-01 设计创作，主要内容包括：本发明公开了一种核脉冲信号处理方法及系统,首先探测器将射线转换为信号后经过模拟放大；将放大后的信号转换成数字信号；将数字信号输入FPGA中经过R-C逆变换消除信号上升沿,然后经过C-R逆变换得到原始信号,将原始信号进行积分、抽样后形成梯形模型以实时产生梯形脉冲信号。本发明采用C-R逆系统、R-C逆系统重新构建了适合于FPGA运算的数字梯形成形递推公式,未引入浮点运算,计算速度更快。并且通过R-C逆系统生成了只有2-4个采样点的高信噪比的快信号处理递推公式。因此极大的提高了测量系统的脉冲通过率。与直接采用快成形相比,需要的采样点更少,并且对于微弱信号的分辨能力更强。(The invention discloses a nuclear pulse signal processing method and a nuclear pulse signal processing system, wherein firstly, a detector converts rays into signals and then performs analog amplification; converting the amplified signal into a digital signal; the digital signal is input into the FPGA to eliminate the rising edge of the signal through R-C inverse transformation, then the original signal is obtained through C-R inverse transformation, and the original signal is integrated and sampled to form a trapezoidal model so as to generate a trapezoidal pulse signal in real time. The invention adopts the C-R inverse system and the R-C inverse system to reconstruct the digital trapezoidal forming recurrence formula suitable for FPGA operation, floating point operation is not introduced, and the calculation speed is higher. And a fast signal processing recurrence formula with high signal-to-noise ratio of only 2-4 sampling points is generated through an R-C inverse system. The pulse passing rate of the measuring system is greatly improved. Compared with the direct adoption of fast shaping, the method has the advantages that fewer sampling points are needed, and the resolving power for weak signals is stronger.)

1. A method of nuclear pulse signal processing, comprising the acts of:

the detector converts the rays into signals and then performs analog amplification;

converting the amplified signal into a digital signal;

the digital signal is input into the FPGA to eliminate the rising edge of the signal through R-C inverse transformation, then the original signal is obtained through C-R inverse transformation, and the original signal is integrated and sampled to form a trapezoidal model so as to generate a trapezoidal pulse signal in real time.

2. The method of claim 1, wherein the R-C inverse transformation comprises the following steps:

obtained by R-C conversionIf the time interval is small enough, Vin can be digitized into x (n), Vout can be digitized into y (n), and dt is 50ns, the above formula can be converted inton is 0, 1, 2 …; then carrying out inverse transformation on the obtained product through R-C to obtain the product;

x[n]＝(1+k)*y[n]-k*y[n-1]

let y [0] be 0 and get the result after integral conversion

And the sigma x [ n ] + k × y [ n ] is the digital recurrence solution of the R-C inverse transformation, and the obtained output result is the digital integral of the original signal.

3. The nuclear pulse signal of claim 1The processing method is characterized in that the specific process of the C-R inverse transformation is as follows: inverse transformation is carried out on the C-R differential forming circuit to obtain a formulaThen, C-R inverse transformation is carried out to obtain

x[n+1]-x[n]＝(1+k)*y[n+1]-y[n]

After finishing to obtain

x[n+1]-x[n]＝k*y[n+1]+(y[n+1]-y[n])

Performing integral transformation on the signal to obtain a signal with an initial value of 0

x[n+1]＝k*Σy[n+1]+y[n+1]

At the moment, the digital recurrence solution of the C-R inverse transformation is completed so as to be suitable for running in an FPGA system; in the formula: the input signal is digitalized into X (n), the output signal is digitalized into Y (n), k is dt/RC which is a floating point number, and dt is 50 ns.

4. The method of claim 3, wherein x [ n +1] ═ k ·Σin the formula

Where k dt/RC is a floating point number, y [ n +1] + y [ n +1], which is retransformed into a floating point number

m*x[n+1]＝Σy[n+1]+m*y[n+1]

Wherein m is 1/k and may be 2ⁿIn the operation, the actual output signal can be restored only by carrying out shift operation, the input signal in the formula is a standard negative exponential signal, the output signal is a step signal or a unit impulse response signal or a sawtooth signal, and then the step signal or the unit impulse response signal is converted into a trapezoidal signal.

5. The nuclear pulse signal processing method according to claim 4,

the input signal x n is a step signal whose trapezoidal shaped model is,

n_a*z(n)＝Σ(x[n]+x[n-n_a-L]-x[n-n_a]-x[n-L]) Wherein n is_aIs the rising width of the trapezoid, and L is the rising width plus the flat top width of the trapezoid.

6. The method of claim 4, wherein the input signal x1[ n ] is a sawtooth signal whose trapezoidal shaped model is,

n_a*z(n)＝(x1[n]+x1[n-n_a-L]-x1[n-n_a]-x1[n-L]) Wherein n is_aIs the rising width of the trapezoid, and L is the rising width plus the flat top width of the trapezoid.

7. The nuclear pulse signal processing method according to claim 4,

the input signal x2[ n ] is a unit impact signal whose trapezoidal shaped model is,

n_a*z(n)＝ΣΣ(x2[n]+x2[n-n_a-L]-x2[n-n_a]-x2[n-L]) Wherein n is_aIs the rising width of the trapezoid, and L is the rising width plus the flat top width of the trapezoid.

8. A nuclear pulse signal processing system, comprising:

the detector is used for acquiring a nuclear pulse signal;

an analog amplifier for amplifying the nuclear pulse signal;

the analog-to-digital converter is used for converting the amplified nuclear pulse signal into a digital signal;

and the FPGA system is used for processing digital signals, eliminating the rising edge of the signals, then obtaining original signals through C-R inverse transformation, and integrating and sampling the original signals to form a trapezoidal model so as to generate trapezoidal pulse signals in real time.

9. The system of claim 8, wherein the FPGA system comprises

The R-C inverse system is used for converting the double-exponential detector signal into a single-exponential signal and eliminating the rising edge of the pulse signal;

and the C-R inverse system is used for converting the nuclear pulse signal into an original signal.

Technical Field

The invention relates to the technical field of signal processing, in particular to nuclear signal processing, and particularly relates to a nuclear pulse signal processing method and system.

Background

The digital processing of core signals has progressed very rapidly, with the application of FPGAs playing a very important role. The data processing speed after high-speed AD sampling must be synchronous with high-speed AD, so the research of the algorithm suitable for FPGA high-speed calculation is particularly important. For example, baseline restoration (see baseline restoration techniques based on symmetric zero-area trapezoids published by Zeng, Yang Jian, Husky, etc. Nuclear instruments and methods in physical research 2017858:57-61), overlapping pulse separation (see trapezoidal pulse shaping for pulse recognition stacked in X-ray spectra published by Zhongjian, Liu Ying, hong Xue, etc. Chinese physics C, 2015, 39 (6): 110-115), pulse discrimination (see Tanglin, well-known, Zhongjian, et al, published as a new method for eliminating false peaks to obtain accurate X-ray spectra; application of radiation and isotopes 2018135:171-176), etc. are processed using FPGAs.

Jordanov and great strength et al have achieved many results by studying digital pulse processing using a deconvolution method. On the basis, a positive system and a reverse system analysis method are introduced. In signal processing, it is common to derive the output result from an existing system. However, sometimes the output signal is needed to judge the signal acquired by the detector, and in this case, the signal needs to be subjected to inverse system digital analysis, but the signal itself is not reversible in the circuit system. Known system and V_in,Solving for V_outWe define the process of (A) as a positive system, a known system and V_outSolving for V_inWe define the process of (a) as an inverse system. We have made some progress in the research of the positive system. Known as V_inAnd V_outThe process of solving the system is called inversion and is mainly applied to earthquake, geological structure and nuclear image processing.

The applicant has now studied an RC inversion system for the repair of the rising edge of a signal (see patent application No. 201910370049.9). However, in the subsequent actual development process, the inventor finds that the ray converted into the signal by the detector is amplified in an analog mode, and then converted into the digital signal to enter the FPGA for processing. The output signal after passing through the analog circuit is not a standard negative exponential signal, but a negative exponential signal with a rising edge. When the signal is directly subjected to spectrum forming processing, the difference sum and division of 2 are different in FPGAⁿThis increases the output error and reduces the processing speed of the FPGA.

Disclosure of Invention

The invention aims to overcome the defects of the prior art, and provides a nuclear pulse signal processing method and a nuclear pulse signal processing system based on further research on a repair technology of an RC inverse system for a signal rising edge. The method has a more concise digital trapezoidal forming formula which can be quickly realized in the FPGA and obtains a fast signal processing mode with higher signal-to-noise ratio.

In order to achieve the purpose, the invention adopts the technical scheme that:

a method of nuclear pulse signal processing, comprising the acts of:

the detector converts the rays into signals and then performs analog amplification;

converting the amplified signal into a digital signal;

the digital signal is input into the FPGA to eliminate the rising edge of the signal through R-C inverse transformation, then the original signal is obtained through C-R inverse transformation, and the original signal is integrated and sampled to form a trapezoidal model so as to generate a trapezoidal pulse signal in real time.

In a further preferred embodiment of the present invention, the R-C inverse transformation specifically comprises the following steps:

R-C conversion is used, and with sufficiently small time intervals, Vin can be digitized into x (n), Vout can be digitized into y (n), dt is 50ns, and the above formula can be converted into n is 0, 1, 2 …; then carrying out inverse transformation on the obtained product through R-C to obtain the product;

x[n]＝(1+k)*y[n]-k*y[n-1]

let y [0] be 0 and get the result after integral conversion

And the sigma x [ n ] + k × y [ n ] is the digital recurrence solution of the R-C inverse transformation, and the obtained output result is the digital integral of the original signal.

In a further preferred embodiment of the present invention, the specific process of the C-R inverse transform is: the C-R differential forming circuit is inversely transformed to obtain a formula, and then C-R inverse transformation is carried out to obtain

x[n+1]-x[n]＝(1+k)*y[n+1]-y[n]

After finishing to obtain

x[n+1]-x[n]＝k*y[n+1]+(y[n+1]-y[n])

Performing integral transformation on the signal to obtain a signal with an initial value of 0

x[n+1]＝k*Σy[n+1]+y[n+1]

At the moment, the digital recurrence solution of the C-R inverse transformation is completed so as to be suitable for running in an FPGA system; in the formula: the input signal is digitalized into X (n), the output signal is digitalized into Y (n), k is dt/RC which is a floating point number, and dt is 50 ns.

In a further preferred embodiment of the invention, k-dt/RC in the formula x [ n +1] ═ k ═ Σ y [ n +1] + y [ n +1] is a floating point number which is reconverted to a floating point number

m*x[n+1]＝Σy[n+1]+m*y[n+1]

In the formula, the input signal is a standard negative exponential signal, the output signal is a step signal or a unit impulse response signal or a sawtooth signal, and then the input signal is converted into a trapezoidal signal.

It is further preferred in the present invention that the input signal x n is a step signal, the trapezoidal shaped model of which is,

n_a*z(n)＝Σ(x[n]+x[n-n_a-L]-x[n-n_a]-x[n-L]) Wherein n is_aIs the rising width of the trapezoid, and L is the rising width plus the flat top width of the trapezoid.

It is a further preferred embodiment of the present invention that the input signal x1 n is a sawtooth signal, whose trapezoidal shaped pattern is,

n_a*z(n)＝(x1[n]+x1[n-n_a-L]-x1[n-n_a]-x1[n-L]) Wherein n is_aIs the rising width of the trapezoid, and L is the rising width plus the flat top width of the trapezoid.

A further preferred embodiment of the invention is that the input signal x 2n is a unit impact signal, whose trapezoidal shaped model is,

n_a*z(n)＝ΣΣ(x2[n]+x2[n-n_a-L]-x2[n-n_a]-x2[n-L]) Wherein n is_aIs the rising width of the trapezoid, and L is the rising width plus the flat top width of the trapezoid.

The invention also provides a nuclear pulse signal processing system for realizing the method, which comprises the following steps:

the detector is used for acquiring a nuclear pulse signal;

an analog amplifier for amplifying the nuclear pulse signal;

the analog-to-digital converter is used for converting the amplified nuclear pulse signal into a digital signal;

and the FPGA system is used for processing digital signals, eliminating the rising edge of the signals, then obtaining original signals through C-R inverse transformation, and integrating and sampling the original signals to form a trapezoidal model so as to generate trapezoidal pulse signals in real time.

In a further preferred embodiment of the system for processing nuclear pulse signals in the present invention, the FPGA system comprises

The R-C inverse system is used for converting the double-exponential detector signal into a single-exponential signal and eliminating the rising edge of the pulse signal;

and the C-R inverse system is used for converting the nuclear pulse signal into an original signal.

Compared with the prior art, the technical scheme of the invention has the following advantages/beneficial effects:

the invention adopts the C-R inverse system and the R-C inverse system to reconstruct the digital trapezoidal forming recurrence formula suitable for FPGA operation, floating point operation is not introduced, and the calculation speed is higher. And a fast signal processing recurrence formula with high signal-to-noise ratio of only 2-4 sampling points is generated through an R-C inverse system, so that the pulse passing rate of a measuring system is greatly improved. Compared with the direct adoption of fast shaping, the method has the advantages that fewer sampling points are needed, and the resolving power for weak signals is stronger.

The inverse system is researched in the analysis of the nuclear signal, and an algorithm suitable for FPGA processing is researched in the real-time nuclear signal processing, so that a new recursive formula for digital trapezoidal forming of the nuclear signal and a recursive formula for fast signal processing with high signal-to-noise ratio are obtained, and the signal processing speed is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

FIG. 1 is a schematic flow diagram of the method of the present invention.

Fig. 2 is a basic C-R differential forming diagram.

FIG. 3 is a schematic diagram of the FPGA implementation of equation (3-c).

Fig. 4 is a diagram illustrating the conversion of a negative exponential signal into a step signal by a CR inversion system.

Fig. 5 is a diagram showing the result of differentiation of the step signal.

Fig. 6 is a schematic diagram of the integration result of the step signal.

Fig. 7 is a schematic diagram of the FPGA implementation of the accumulation section.

Fig. 8 is a schematic diagram of the transformation of an actual detector signal into a trapezoidal signal.

Fig. 9 is a schematic diagram of an RC integration circuit.

FIG. 10 is a schematic diagram of the FPGA implementation of equation (10-b).

FIG. 11 is a graph showing simulation effect of the R-C inverse system on repairing the rising edge of the measured signal.

FIG. 12 is a schematic diagram of the simulation effect after the trapezoid forming is improved.

FIG. 13 is a graph of the effect of the R-C inverse system on generating fast pulses.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions in the embodiments of the present invention are clearly and completely described below, and it is obvious that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention. Thus, the detailed description of the embodiments of the present invention provided below is not intended to limit the scope of the invention as claimed, but is merely representative of selected embodiments of the invention.