A randomness removal method for backscatter echo in Monte Carlo simulation of smoke, dust and fog environment

Abstract

The present invention discloses a method for removing randomness from backscatter echoes in Monte Carlo simulations of smoke and fog environments. Using a laser backscatter echo model for smoke and fog environments, several backscatter echoes are simulated under the same parameter conditions to form a total echo dataset under these parameter conditions. The most representative characteristic parameters of the backscatter echo waveforms are extracted and organized into an array. The backscatter echoes are screened using a multi-characteristic box plot method, and the quality evaluation coefficient e of individual echoes that meet the screening criteria is calculated. Finally, the echo with the smallest e value is selected as the optimal backscatter echo output under these parameter conditions. The present invention improves the stability of the Monte Carlo simulation, effectively solves the problem of large random errors in simulated echoes, and enhances the accuracy of backscatter echo simulation in smoke and fog environments.

Description

Technical Field

[0001] The invention belongs to the field of laser detection echo simulation, and in particular relates to a method for removing randomness of backscattered echoes in Monte Carlo simulation of a smoke, dust, cloud and fog environment. Background Art

[0002] Laser detection has the advantages of strong directionality, strong resistance to electromagnetic interference, and high ranging accuracy. Therefore, it is widely used in the detection systems of various ammunition. The accuracy of the laser detection system directly affects the weapon system's effectiveness in damaging the target. However, in practice, laser detection systems are often interfered with by various environmental factors, such as smoke, dust, and fog. These environmental particles can cause the laser to be absorbed or scattered, thereby affecting the accuracy of the laser detection system. Therefore, it is of great significance to study the transmission of detection laser in smoke, dust, and fog environments and the characteristics of its backscattered echo.

[0003] Monte Carlo simulation is commonly used to study the transmission of detection lasers in smoke and fog environments. This method transforms the transmission process into a series of photon collisions with ambient particles in smoke and fog environments. The characteristic states of randomly wandering photons after collision are simulated and calculated. The number of photons received within different time series is counted to obtain the laser echo signal. The general process of this method includes the generation and migration of photons, the collision and absorption of photons, and the reception and disappearance of photons.

[0004] The Monte Carlo method is essentially a numerical calculation method based on random sampling. Its core idea is to estimate the value of a target quantity through a large number of repeated experiments simulating random events. The more repeated experiments, the closer the result is to the actual situation. However, in the laser transmission process used in actual detection, the number of photons in a single transmission is huge. Considering time and hardware, it is difficult to simulate a number of photons close to the actual situation. Therefore, when simulating laser transmission and calculating the simulated echo, the number of photons is often reduced. However, reducing the number of photons can cause greater randomness in the Monte Carlo simulation echo data. Under the same parameters, the simulated echo may have large errors, which will seriously affect the simulation accuracy. The lack of simulation stability will reduce the credibility of the simulation results. Summary of the Invention

[0005] The present invention proposes a method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment, so as to improve the simulation stability of Monte Carlo waveform of laser detection in smoke, dust and fog environment.

[0006] The technical solution to achieve this invention is: a method for removing randomness from backscatter echoes in Monte Carlo simulation of smoke, dust, and fog environments, comprising the following steps:

[0007] Step 1: Using the laser backscatter echo model in a smoke and fog environment, simulate several backscatter echoes under the same parameter conditions. Each backscatter echo is described by a time-amplitude point on the coordinate axis. The time-amplitude point descriptions of all echoes constitute the overall echo data set under the parameter conditions, and then proceed to step 2.

[0008] Step 2: For the backscattered echo waveform, extract the four most representative characteristic parameters of the backscattered echo, including the peak value of the first peak, the peak value of the second peak, the time corresponding to the first peak, and the time corresponding to the second peak; solve the four characteristic parameters of each backscattered echo in the overall echo data set, and classify the echo characteristic parameters into an array, corresponding to four characteristic parameter data sets, and proceed to step 3.

[0009] Step 3. Combine the representative characteristic parameters extracted in step 2 and the obtained data set, and use the multi-characteristic parameter box plot method to screen the backscattered echo; process the four characteristic parameter data sets obtained in step 2 respectively to obtain five statistics that can describe each characteristic parameter data set, including the upper edge, upper quartile, median, lower quartile and lower edge; determine whether the four characteristic parameters of each echo fall within the interval formed by the lower edge and upper edge of the characteristic parameter data set. If so, go to step 4; if not, discard the echo.

[0010] Step 4: Calculate the quality evaluation coefficient e of a single backscatter echo, and finally select the echo with the smallest e value as the optimal backscatter echo output under the parameter condition.

[0011] Compared with the prior art, the present invention has the following significant advantages:

[0012] (1) The present invention takes into account the randomness of the Monte Carlo method and selects the most appropriate sample under the parameter from a large number of samples as the output echo data. Compared with the single simulation result of the original simulation model, the simulation waveform is more accurate and more stable without increasing the number of simulated photons and excessive simulation time.

[0013] (2) When extracting waveform characteristic parameters, the present invention takes into account the particularity of the backscattered echo shape and specifically extracts four features that can better describe the waveform conditions, and provides solution methods for these four features under normal conditions and waveform distortion. Based on these waveform characteristics, a multi-feature parameter box plot method is proposed to extract the echo data set into multiple feature sets, which not only retains the original echo waveform characteristics but also avoids the use of complex image processing methods, simplifies the data processing method while ensuring the accuracy and representativeness of the final output waveform.

[0014] (3) The present invention proposes for the first time a criterion and coefficient formula for judging the quality of backscattered echoes, which can be used to quantitatively describe whether a certain backscattered echo is representative of a large number of sample echoes operating under the same parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a principle block diagram of the randomness removal method of the Monte Carlo simulation backscatter echo in a smoke, cloud and fog environment according to the present invention.

[0016] Figure 2 The figure shows the results of 100 Monte Carlo simulations under the same parameter conditions.

[0017] Figure 3 Figure 3 shows the three main situations of backscattered echo waveforms, where (a) is a schematic diagram showing one peak, (b) is a schematic diagram showing two peaks and the target echo peak is lower than the interference echo peak, and (c) is a schematic diagram showing two peaks and the target echo peak is higher than the interference echo peak.

[0018] Figure 4 These are the results of 100 Monte Carlo simulations under the same parameters and the echo simulation result output diagram obtained by the method of the present invention (first group), where (a) represents the overall diagram of 100 echoes under the same parameter conditions and the corresponding optimal output echoes, and (b) represents the local enlarged diagram of 100 echoes under the same parameter conditions and the corresponding optimal output echoes.

[0019] Figure 5 These are the results of 100 Monte Carlo simulations under the same parameters and the echo simulation result output diagram obtained by the method of the present invention (second group). (a) shows the overall diagram of 100 echoes under the same parameter conditions and the corresponding optimal output echoes, and (b) shows the local enlarged diagram of 100 echoes under the same parameter conditions and the corresponding optimal output echoes. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In the present invention, unless otherwise specified or limited, the terms "connection" and "fixation" should be understood in a broad sense. For example, "fixation" can refer to fixed connection, detachable connection, or integration; "connection" can refer to mechanical connection or electrical connection. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0022] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in this field can implement it. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0023] The following will further introduce the specific implementation methods, as well as the technical difficulties and inventive points of this invention in combination with this design example.

[0024] Combine Figure 1 A method for removing randomness from backscatter echoes in Monte Carlo simulation of smoke, dust and fog environments is proposed. The specific implementation steps are as follows:

[0025] Step 1: Using the laser backscatter echo model in a smoke and fog environment, several backscatter echoes are simulated under the same parameter conditions. Each backscatter echo is described by a time-amplitude point on the coordinate axis. The time-amplitude data of all echoes constitute the overall echo data set under the parameter conditions.

[0026] Under the same parameter conditions, B backscattered echoes are simulated. Considering the sufficiency of the sample size and the time consumed in running, B is set to 100-200. Let the horizontal coordinate of the i-th scattered point on the backscattered echo be time t i , the vertical axis is the amplitude v i . Combined Figure 2 , the simulation obtains 100 groups of echoes under the same parameters, the horizontal coordinate of the echo signal is time t, and the vertical coordinate is amplitude v. The scattered coordinates (t, v) of these 100 echo signals are stored in sequence.

[0027] Step 2: Extract the four most representative characteristic parameters of the backscattered echo waveform, including the peak value of the first peak, the peak value of the second peak, the time corresponding to the first peak, and the time corresponding to the second peak. Calculate the four echo characteristic parameters for each backscattered echo in the overall echo dataset and organize them into arrays based on the echo characteristic parameters, resulting in four corresponding characteristic parameter datasets.

[0028] Combine Figure 3 ,Generally speaking, the simulated backscattered echo waveform mainly has the following ,situations: 1) there is only one peak, 2) there are two peaks, and the target ,echo peak is lower than the interference echo peak, 3) there are two peaks, and the ,target echo peak is higher than the interference echo peak.

[0029] S2.1. Since the location of the peak is also the location of the maximum point, the location of the backscattered signal peak can be determined by finding the maximum point of the echo, and the location of the maximum point can be determined based on the slope.

[0030] The slope k of the i-th scatter point i is calculated as follows:

[0031] k i =(v i+1 -v i ) / (t i+1 -t i )

[0032] (t i+1 , v i+1 ) represents the coordinates of the i+1th scattered point on the backscattered echo.

[0033] If the slope k i If the following relationship is satisfied, the i+1th point of the echo signal is the maximum point:

[0034]

[0035] If the simulated echo has only one or two peaks, proceed to step S2.3. However, due to uncertainties in the simulation, in some cases the number of echo peaks may be greater than two, indicating that the echo is distorted. In this case, the following method is used, assuming that the number of peaks at this time is s, and s>2:

[0036] 1) Take v_max=max(v), if v n <λ·v_max, then the point is removed from the peak point set, s=s-1.

[0037] v n represents the nth peak value of the echo, λ is the proportional coefficient, which can be set according to the simulation signal-to-noise ratio in actual simulation, and can be set to λ = 0.01 to 0.1. v_max represents the maximum point of the echo, and max() represents the maximum value function.

[0038] 2) Take t_z=1 / f, if in the interval [t n -0.5t_z,t n If there is an extreme point other than the point's location within [t_z], then the point is removed from the peak point set, s = s-1. f is the noise frequency obtained from simulation experience, and t_z represents the theoretical time span of the noise during the simulation process.

[0039] If the final value of s is 1 or 2, go to step S2.3; if s is greater than 2 or 0, take the maximum value point as the peak point and go to step S2.3.

[0040] S2.3. Array the first peak value, second peak value, first peak corresponding time, and second peak corresponding time of the B backscattered echoes in order and by echo parameter classification. These are stored in vectors V1, V2, T1, and T2 as four feature parameter data sets. If a backscattered echo has only one peak (the target echo and backscattered echo are located close together), then the first peak value is the same as the second peak value, and the first peak corresponding time is the same as the second peak corresponding time.

[0041] Step 3: Combine the representative characteristic parameters extracted in step 2 with the resulting data set and use the multi-characteristic parameter box plot method to screen the backscattered echoes. Each of the four characteristic parameter data sets obtained in step 2 is processed separately to derive five statistics describing each characteristic parameter data set: the upper edge, upper quartile, median, lower quartile, and lower edge. Determine whether the four characteristic parameters of each echo fall within the interval formed by the lower and upper edges of the characteristic parameter data set. If so, proceed to step 4; otherwise, discard the echo.

[0042] The box plot method is a common method in data processing, accurately and stably depicting the discrete distribution and outliers of data. However, it is difficult to directly use a box plot to process the simulated backscatter echo waveform set. Therefore, it is possible to extract multiple characteristic parameter data sets that can describe the backscatter echoes and perform multi-characteristic box plot processing. This avoids the use of complex image processing methods while ensuring that the selected backscatter echo characteristic parameters are free of obvious anomalies.

[0043] S3.1. Calculate the median M, lower quartile Q1, and upper quartile Q2 of the four feature parameter data sets. The median is the value in the middle of the data after it is arranged in ascending order. If the sequence is even, it is the average of the two middle values. The lower quartile is the value at the 25th percentile of the data sequence; the upper quartile is the value at the 75th percentile of the data sequence.

[0044] The medians M of the four characteristic parameters are: M_V1, M_V2, M_T1, and M_T2.

[0045] The lower quartiles Q1 of the four characteristic parameters are: Q1_V1, Q1_V2, Q1_T1, and Q1_T2.

[0046] The upper quartiles Q2 of the four characteristic parameters are: Q2_V1, Q2_V2, Q2_T1, and Q2_T2. S3.2. Calculate the interquartile range IQR of the characteristic parameter data set using the following formula:

[0047] IQR=Q2-Q1

[0048] The interquartile ranges (IQRs) of the four characteristic parameters are: IQR_V1, IQR_V2, IQR_T1, and IQR_T2:

[0049] S3.3. Calculate the lower edge U1 and upper edge U2 of the characteristic parameter data set using the following two formulas:

[0050] U1=Q1-1.5*IOR

[0051] U2=Q2+1.5*IQR

[0052] The lower edges U1 of the four characteristic parameters are obtained as follows: U1_V1, U1_V2, U1_T1, and U1_T2.

[0053] The upper edges U2 of the four characteristic parameters are obtained as follows: U2_V1, U2_V2, U2_T1, and U2_T2.

[0054] S3.4. Determine whether the characteristic parameters V1(m), V2(m), T1(m), and T2(m) of the mth echo meet the requirements of the following formula. If so, it means that the characteristic values ​​of this waveform are within a reasonable range and proceed to step 4. If not, discard the echo and make a judgment on the m+1th echo.

[0055]

[0056] In the above formula, V1(m) represents the peak value of the first wave peak of the m-th echo, V2(m) represents the peak value of the second wave peak of the m-th echo, T1(m) represents the time corresponding to the first wave peak of the m-th echo, and T2(m) represents the time corresponding to the second wave peak of the m-th echo.

[0057] Step 4: Calculate the quality evaluation coefficient e of a single echo according to the definition of the backscatter echo quality evaluation coefficient e, and finally select the echo with the smallest e value as the optimal backscatter echo output under the parameter condition.

[0058] When the value of a characteristic parameter of an echo is close to the median of the characteristic parameter data set, it means that the characteristic parameter value of the echo is representative of the characteristic parameter of the B echoes. The single waveform quality evaluation coefficient e obtained by simulation is thus defined as:

[0059]

[0060] Where e(m) represents the evaluation coefficient of the mth echo.

[0061] The calculation formula for the evaluation coefficient e is divided into four parts, representing the relative distance between the four characteristic parameters and their respective medians. Because amplitude and time have different dimensions, each of the four components is divided by the corresponding upper edge U2 to normalize and eliminate the dimension, facilitating the summation process.

[0062] The echo with the smallest e value is selected as the simulated echo in this case.

[0063] Example

[0064] The backscatter echo randomness removal method of the present invention is used in a Monte Carlo echo simulation example. The visibility is set to 10m, the target distance is 4.5m, and the laser emission pulse width is 10ns. The simulation results are Figure 4 and Figure 5 100 groups of original backscatter echoes and the optimal echo as the output result after being processed by the backscatter echo simulation randomness removal method based on multi-feature parameter box plot. Figure 4 and Figure 5 The relative errors of the four characteristic parameters of the first peak peak value, the second peak peak value, the first peak corresponding time and the second peak corresponding time of the optimal echo obtained twice are all less than 10%. It can be seen that the present invention can better suppress the randomness of the backscattered echo in the Monte Carlo simulation of smoke and fog environment.

Claims

1. A method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment, characterized by: Here are the steps: Step 1: Using the laser backscatter echo model for a smoke and fog environment, simulate several backscatter echoes under the same parameter conditions. Each backscatter echo is described by a time-amplitude point on the coordinate axis. The time-amplitude point descriptions of all echoes constitute the overall echo data set under the parameter conditions, and then proceed to Step 2. Step 2: Extract the four most representative characteristic parameters of the backscattered echo waveform, including the peak value of the first peak, the peak value of the second peak, the time corresponding to the first peak, and the time corresponding to the second peak. Calculate the four characteristic parameters of each backscattered echo in the overall echo data set, and classify them into arrays according to the echo characteristic parameters. Four corresponding characteristic parameter data sets are obtained, and then proceed to step 3. Step 3: Combine the representative characteristic parameters extracted in step 2 and the obtained data set, and use the multi-characteristic parameter box plot method to screen the backscattered echo; process the four characteristic parameter data sets obtained in step 2 respectively, and obtain five statistics that can describe each characteristic parameter data set, including the upper edge, upper quartile, median, lower quartile and lower edge; determine whether the four characteristic parameters of each echo fall within the interval formed by the lower edge and upper edge of the characteristic parameter data set respectively. If so, proceed to step 4; if not, discard the echo; Step 4: Calculate the quality evaluation coefficient e of a single backscatter echo, and finally select the echo with the smallest e value as the optimal backscatter echo output under the parameter condition.

2. The method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment according to claim 1 is characterized by: In step 1, the laser backscatter echo model in smoke and fog environment is used to simulate B backscatter echoes under the same parameter conditions, B = 100 to 200, and the horizontal coordinate of the i-th scattered point on the backscatter echo is set to time t i , the vertical axis is the amplitude v i .

3. The method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment according to claim 2 is characterized by: The waveform of the backscattered echo has the following conditions: 1) there is only one peak, 2) there are two peaks, and the peak of the target echo is lower than the peak of the interference echo, 3) there are two peaks, and the peak of the target echo is higher than the peak of the interference echo.

4. The method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment according to claim 3 is characterized in that: In step 2, the four most representative characteristic parameters of the backscattered echo waveform are extracted, including the peak value of the first peak, the peak value of the second peak, the time corresponding to the first peak, and the time corresponding to the second peak. The echo characteristic parameters are classified into arrays to obtain four corresponding characteristic parameter data sets. The steps are as follows: S2.

1. Since the peak position is also the maximum point position, the peak position of the backscattered signal can be determined by finding the maximum point of the echo, and the slope at the maximum point is 0; The slope k of the i-th scatter point i is calculated as follows: k i =(v i+1 -v i ) / (t i+1 -t i ) (t i+1 , v i+1 ) represents the coordinates of the i+1th scattered point on the backscattered echo; If the slope k i If the following relationship is satisfied, the i+1th scattered point on the backscatter echo is the maximum point: If the simulated echo has only one or two peaks, proceed to step S2.

3. However, due to uncertainties in the simulation, in some cases the number of echo peaks may be greater than two, indicating that the echo is distorted. In this case, the following method is used, assuming that the number of peaks at this time is s, where s>2: 1) Take v_max=max(v), if v n <λ·v_max, then the point is removed from the peak point set, s=s-1; v n Indicates the nth peak value of the echo, λ is the proportional coefficient, which is set according to the simulation signal-to-noise ratio in actual simulation, and is set to λ = 0.01 to 0.1; v_max indicates the maximum value point of the echo, and max() indicates the maximum value function; 2) Take t_z=1 / f, if in the interval [t n -0.5t_z,t n +0.5t_z], if there is an extreme point other than the location of this point, then the peak is removed from the peak point set, s = s-1; f is the noise frequency obtained from simulation experience, and t_z represents the theoretical time span of the noise in the simulation process; If the final value of s is 1 or 2, proceed to S2.3; if s is greater than 2 or 0, take the maximum value point as the peak point and proceed to S2.3; S2.

3. The first peak value, the second peak value, the time corresponding to the first peak, and the time corresponding to the second peak of the B backscattered echoes are sorted into arrays in order and classified according to echo parameters, and stored in vectors V1, V2, T1, and T2 as four feature parameter data sets; If a backscattered echo has only one peak point, the peak value of the first peak is the same as the peak value of the second peak, and the corresponding time of the first peak is the same as the corresponding time of the second peak.

5. The method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment according to claim 4 is characterized in that: In step 3, the backscatter echoes are screened using the multi-feature parameter box plot method. The four feature parameter data sets obtained in step 2 are processed separately to obtain five statistics that can describe each feature parameter data set, including the upper edge, upper quartile, median, lower quartile, and lower edge. It is determined whether the four feature parameters of each echo fall within the interval formed by the lower edge and upper edge of the feature parameter data set. The steps are as follows: S3.

1. Calculate the median M, lower quartile Q1, and upper quartile Q2 for each of the four feature parameter data sets. The median is the value in the middle of the data after it is arranged in ascending order. If the sequence is an even number of numbers, it is the average of the two middle numbers. The lower quartile is the number at the 25th percentile of the data sequence. The upper quartile is the number at the 75th percentile of the data sequence. S3.

2. Calculate the interquartile range (IQR) of the feature parameter data set: IQR=Q2-Q1 S3.

3. Calculate the lower edge U1 and upper edge U2 of the feature parameter data set: U1=Q1-1.5*IOR U2=Q2+1.5*IOR The median M, lower quartile Q1, upper quartile Q2, lower edge U1 and upper edge U2 of the four feature parameter data sets V1, V2, T1 and T2 are obtained as follows: M_TZ, Q1_TZ, Q2_TZ, U1_TZ, U2_TZ, and the feature variable TZ = V1, V2, T1, T2; S3.

4. Determine whether the characteristic parameters V1(m), V2(m), T1(m), and T2(m) of the mth echo satisfy the following requirements: If satisfied, it means that the characteristic parameter values ​​of the backscattered echo waveform are within a reasonable range and proceed to step 4; If not, the backscatter echo is discarded; In the above formula, V1(m) represents the peak value of the first wave peak of the m-th echo, V2(m) represents the peak value of the second wave peak of the m-th echo, T1(m) represents the time corresponding to the first wave peak of the m-th echo, and T2(m) represents the time corresponding to the second wave peak of the m-th echo.

6. The method for removing randomness of backscatter echoes in Monte Carlo simulation of smoke, dust and fog environment according to claim 5, characterized in that: In step 4, the quality evaluation coefficient e of a single backscatter echo is calculated, and finally the echo with the smallest e value is selected as the optimal backscatter echo output under the parameter conditions, as follows: When a characteristic parameter value of a backscattered echo is close to the median of the characteristic parameter data set, it means that the characteristic parameter value of the echo is representative of the characteristic parameter of the B echoes. Based on this, the single waveform quality evaluation coefficient e obtained by simulation is defined, and the e value of the echo that meets the requirements is calculated according to the expression of e: e(m) represents the quality evaluation coefficient value of the mth backscatter echo; the echo with the smallest e value is selected as the optimal backscatter echo under the parameter condition.

Images