An anti-interference method for vehicle-borne millimeter-wave radar echo signal based on VMD and VI-CFAR

Abstract

The present invention belongs to the technical field of millimeter wave radar signal processing, and in particular, is an anti-interference method for vehicle-mounted millimeter wave radar echo signals based on VMD and VI-CFAR, comprising the following steps: step 1: using a vehicle-mounted millimeter wave radar to transmit millimeter waves for target detection, and receiving the interfered original echo signal x(t); step 2: setting the parameter range of the decomposition layer number K and the penalty factor α of the VMD algorithm. The present invention adaptively decomposes the echo signal through VMD, adaptively decomposes the original echo signal into multiple IMFs, then performs STFT on each IMF, and screens out the relevant IMFs, then performs threshold calculation and signal reconstruction on the screened relevant IMFs through VI-CFAR to obtain new relevant IMFs, and finally superimposes all the relevant IMFs to obtain the vehicle-mounted millimeter wave radar echo signal after anti-interference, and finally realizes the anti-interference operation of the vehicle-mounted millimeter wave radar echo signal.

Description

Technical Field

[0001] The present invention relates to the technical field of millimeter wave radar signal processing, and in particular to an anti-interference method for vehicle-mounted millimeter wave radar echo signals based on VMD and VI-CFAR. Background Art

[0002] With the development of autonomous driving technology and advanced driver assistance, millimeter-wave radar has been widely used in the automotive field, and vehicle-mounted millimeter-wave radar has become an important part of the vehicle-mounted radar system. With its high accuracy in detecting the speed and distance of moving targets, its low price and its advantages of being not easily affected by bad weather, vehicle-mounted millimeter-wave radar has become the focus of vehicle-mounted radar development in recent years. However, in the process of using vehicle-mounted millimeter-wave radar, due to the existence of interference factors in the ground-air environment, the echo signal received by the vehicle-mounted millimeter-wave radar will contain a certain amount of interference signals to a certain extent, and the existence of these interference signals will have a certain impact on the accuracy of the results of target recognition and parameter estimation. Therefore, using effective anti-interference methods to perform anti-interference processing on vehicle-mounted millimeter-wave radar signals and accurately extracting useful information from the echo signals of vehicle-mounted millimeter-wave radars are the key to improving the function of vehicle-mounted millimeter-wave radars.

[0003] At present, in order to reduce the mutual interference between radars, domestic and foreign scholars have proposed many interference suppression methods. Johanna Rock et al. proposed to use convolutional neural networks to alleviate the problem of mutual interference. Although this method can retain the target signal as much as possible while resisting interference, its premise is that the number of training samples is large enough. Kim Seong Cheol and his team proposed to separate the target signal and the interference signal by wavelet denoising, but it is not possible to determine the parameters of wavelet denoising through a large amount of measured data. Eun Hee Kim et al. improved the anti-interference ability of automotive radar by designing more complex waveforms. However, complex waveforms have higher requirements for hardware, which will lead to an increase in hardware costs. E. Elsehely used wavelet transform to process the echo signal and then realize target detection, but the effect of this method in multi-target scenarios was not satisfactory. Nozawa set the interference area to zero, and setting the signal to zero will cause the continuous line of the signal to deteriorate, and may also cause the target signal in the interference signal part to be lost. Rameez used an autoregressive (AR) model to predict the target information in the interference part, but it could not restore the target information well when there were many interference signals.

[0004] Therefore, we propose an anti-interference method for vehicle-borne millimeter-wave radar echo signals based on VMD and VI-CFAR to solve the above problems. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the deficiencies in the prior art, the present invention provides an anti-interference method for vehicle-mounted millimeter-wave radar echo signals based on VMD and VI-CFAR, which solves the problems raised in the above-mentioned background technology.

[0007] (II) Technical solution

[0008] In order to achieve the above-mentioned purpose, the present invention specifically adopts the following technical solutions:

[0009] An anti-interference method for vehicle-mounted millimeter-wave radar echo signals based on VMD and VI-CFAR comprises the following steps:

[0010] Step 1: Use the vehicle-mounted millimeter-wave radar to transmit millimeter waves for target detection and receive the interfered original echo signal x(t).

[0011] Step 2: Set the parameter range of the decomposition level K and penalty factor α of the VMD algorithm.

[0012] Step 3: Use the minimum sample entropy mean of the decomposed IMF as the fitness function, and use the GAWOA algorithm to obtain the optimal parameter combination of the VMD algorithm.

[0013] Further, the step 3 comprises:

[0014] Step 3.1: Initialize the population size M, maximum number of iterations H, and value range of VMD algorithm parameters of the GAWOA algorithm, and initialize Z i , i=1,...,M, where Z i Each individual whale in the whale population represents a random and different initial parameter combination, and the initial number of iterations is set to 1.

[0015] Step 3.2: Within the range of VMD algorithm parameters, decompose the vehicle-borne millimeter-wave radar echo signal x(t). Each whale individual corresponds to a set of VMD decomposition parameter combinations.

[0016] Step 3.3: Use the parameter combination corresponding to each individual whale to perform VMD decomposition on the original signal, calculate the corresponding sample entropy value of each IMF obtained after decomposition, add the sample entropy values ​​of each IMF and take the average value, and save the minimum fitness value F and its corresponding parameter value.

[0017] Step 3.4: Update and optimize the parameters through the GAWOA algorithm, change the position of the individual whales, that is, update the value of the parameter combination, and calculate the fitness function value F corresponding to each individual whale in the hth iteration process. h, and compare it with the fitness function value corresponding to the current optimal parameter combination, and keep the parameter combination of the whale individual with the smaller fitness function value as the current optimal parameter combination.

[0018] Further, the step 3.4 includes:

[0019] Step 3.4.1: Generate a solution space with the upper and lower limits of the parameters as the limits. Assume that the current optimal candidate parameter combination is the target parameter combination or a combination close to the target parameter combination. After that, all whale individuals in the whale population try to update their positions to the current optimal candidate parameter combination. The equations for this operation are as follows:

[0020] D=|C·X * (h)-X(h)| (1)

[0021] X(h+1)=X * (h)-A·D (2)

[0022] In the above formula, h is the current iteration number, D represents the distance between the current individual and the optimal individual, A and C represent the coefficient vector, and X * (h) is the best position obtained so far. If a better solution exists, X will be updated in each iteration. * (h). The calculation formulas of coefficient vectors A and C are shown in equations (3) and (4), and the update scheme of parameter a in equation (3) is shown in equation (5):

[0023] A=2a×r1-a (3)

[0024] C=2×r2 (4)

[0025]

[0026] Where h is the current iteration number, H is the maximum iteration number; in the process of continuous iteration, the value of a gradually decreases nonlinearly from 2 to 0, r1 and r2 are random vectors in the range of [0,1];

[0027] Step 3.4.2: The position of individual whales in the optimization process of the GAWOA algorithm is updated using a spiral update strategy, which is updated through a logarithmic spiral equation. The specific expressions are as follows:

[0028] X(h+1)=D'×e bL ×cos(2πL)+X * (h) (6)

[0029] D'=|X * (h)-X(h)| (7)

[0030] In the above formula: D' represents the distance between the current search individual and the current optimal solution; b is the spiral shape parameter; L is a random number with a uniform distribution in the range of [-1, 1]. Since there are different update strategies in the process of approaching prey, the GAWOA algorithm selects different update strategies according to the probability p.

[0031] When p>0.5, the position of individual whales is updated through a mutation operation strategy similar to the genetic algorithm, that is, the parameter value is updated. The updating method is shown in formula (8), P m The value of ' is shown in formula (9):

[0032] X * (h+1)=X * (h)+P m *D*e bL *cos(2π*L) (8)

[0033]

[0034] In the above formula, X * (h+1) is the updated location of the whale, X * (h) is the current location of the whale, P m is the probability of mutation.

[0035] When p≤0.5 and |A|≥1, the position of individual whales is updated through a strategy similar to the crossover operation of the genetic algorithm. The update method is as follows:

[0036]

[0037] In the above formula, P c is the probability of crossover, X i (h) is the current location of the individual whale, X j (h) is the new position of the individual whale after the traditional whale optimization algorithm update strategy is updated, X i (h+1) and X j (h+1) are the two new positions of the whale individual after the crossover operation. The fitness values ​​of the two new positions are then compared, and the position with the smaller fitness value is determined as the new position of the whale individual after the crossover operation.

[0038] When p≤0.5 and |A|<1, the update method of formula (11) is used to update the position of individual whales:

[0039] X(h+1)=ω*X * (h)-A*D (11)

[0040] In the above formula, X(h+1) is the position of the individual whale after the position update, ω is the adaptive weight, and the adaptive weight ω is calculated according to formula (12).

[0041]

[0042] In the above formula, ω max is the maximum weight coefficient, ω min is the minimum weight coefficient, H is the maximum number of iterations, and h is the current number of iterations.

[0043] Step 3.4.3: In the process of continuous iteration and updating, the sample signal is decomposed by VMD using the parameters updated after the hth iteration, and the optimal fitness function F of the IMF decomposed after the hth iteration is obtained by calculating and comparing the fitness function values ​​of each individual whale in the whale group during the hth iteration. h , and compare it with the F corresponding to the current optimal parameter combination, leaving the parameter combination of the whale individual with a smaller fitness function F value as the current optimal parameter combination.

[0044] Step 3.5: Determine whether the GAWOA algorithm has reached the condition for iteration termination: whether the current iteration number h of the GAWOA algorithm is greater than or equal to the maximum iteration number H. When h≥H, proceed to step 3.6, otherwise h=h+1, and update the position of the whale group.

[0045] Step 3.6: When h>H, the iteration process ends and the optimal parameter combination is obtained. And the corresponding minimum fitness

[0046] Step 4: Use the optimal parameter combination of the VMD algorithm obtained in step 3 to decompose the original vehicle-borne millimeter-wave radar echo signal or the reconstructed signal to obtain multiple IMFs.

[0047] Step 5: Perform STFT on each IMF obtained by decomposition to obtain the spectrum of each IMF, and filter out the relevant IMF containing the target information based on the spectrum.

[0048] Further, the step 5 comprises:

[0049] Step 5.1: Decompose multiple IMFs according to step 4, perform STFT on each IMF, and obtain the corresponding spectrum diagram of each IMF.

[0050] Step 5.2: Based on the screening principle that the single-target vehicle-mounted millimeter-wave radar echo signal appears as a straight line on the spectrum diagram, the IMF containing the target information in each IMF is determined and defined as the relevant IMF.

[0051] Step 6: Analyze and reconstruct the selected relevant IMFs through VI-CFAR to obtain the reconstructed relevant IMFs.

[0052] Further, the step 6 comprises:

[0053] Step 6.1: Calculate the threshold value corresponding to each data point of each relevant IMF screened in step 5.2 through VI-CFAR;

[0054] Step 6.2: Compare the actual value of each point with the corresponding threshold value. If the actual value of the point is greater than or equal to the threshold value corresponding to the point, the original value of the point is retained as the value of the point; otherwise, if the value of the point is less than the threshold value corresponding to the point, the point is re-assigned, and the average value of the two points adjacent to the point before and after the point is used as the value of the point instead of the original value;

[0055] Step 6.3: After re-assignment, the reconstructed related IMF is obtained.

[0056] Step 7: Finally, all the reconstructed related IMFs are superimposed to realize the anti-interference operation of the vehicle-borne millimeter-wave radar echo signal, and finally the anti-interference vehicle-borne millimeter-wave radar echo signal is obtained.

[0057] (III) Beneficial effects

[0058] Compared with the prior art, the present invention provides an anti-interference method for vehicle-mounted millimeter-wave radar echo signals based on VMD and VI-CFAR, which has the following beneficial effects:

[0059] 1. Compared with other anti-interference methods for vehicle-mounted millimeter-wave radar echo signals, the present invention has a better anti-interference effect on vehicle-mounted millimeter-wave radar echo signals in a complex environment with multiple interference sources, and can effectively retain the useful information in the original signal without causing the target to be lost, etc., and has a good application prospect.

[0060] 2. The present invention performs corresponding processing on the echo signal, so that the anti-interference operation of the vehicle-mounted millimeter-wave radar can be completed without increasing the hardware cost of the vehicle-mounted millimeter-wave radar.

[0061] 3. The present invention improves the whale optimization algorithm through a genetic algorithm to obtain an improved whale optimization algorithm (GAWOA), which improves the update strategy of the traditional whale optimization algorithm, the update method of the convergence factor a and adds adaptive weights, thereby increasing the convergence speed of the whale optimization algorithm and, to a certain extent, avoiding the occurrence of the situation where the iterative process of the traditional whale optimization algorithm falls into the local optimum.

[0062] 4. The GAWOA algorithm is used to improve the parameter selection method of the VMD algorithm, which effectively solves the problem that the parameters need to be manually selected when the VMD algorithm is used, so that the VMD algorithm can achieve the optimal configuration of parameters when processing the echo signals of different vehicle-mounted millimeter-wave radars. Therefore, the present invention can be applied to the signal processing of vehicle-mounted millimeter-wave radars of different brands, and achieves a good anti-interference effect, and has wide applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flow chart of the anti-interference method of the vehicle-mounted millimeter-wave radar echo signal based on VMD and VI-CFAR of the present invention;

[0064] Figure 2 It is a time domain waveform diagram and a frequency domain waveform diagram of a simulated echo signal of a vehicle-mounted millimeter-wave radar interfered by multiple interference signals of the present invention;

[0065] Figure 3 This is a result diagram of the STFT of the single-target vehicle-mounted millimeter-wave radar echo signal of the present invention;

[0066] Figure 4 It is the STFT spectrum diagram of each IMF after decomposition of the present invention;

[0067] Figure 5 It is a frequency domain diagram of the signal after anti-interference of the present invention;

[0068] Figure 6 It is a time domain waveform diagram of the signal after anti-interference and the signal without interference of the present invention;

[0069] Figure 7 It is a comparison diagram of the signal-to-interference ratio (SIR) of each target signal in the non-interference situation, the interference situation and the anti-interference situation of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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.

[0071] Example

[0072] like Figure 1-7 As shown, an anti-interference method for a vehicle-mounted millimeter-wave radar echo signal based on VMD and VI-CFAR is proposed in one embodiment of the present invention, and includes the following steps.

[0073] Step 1: Use MATLAB to simulate the echo signal of the vehicle-mounted millimeter-wave radar with multiple interferences as the sample signal. The waveform of the sample signal is as follows: Figure 2 .

[0074] Step 2: Initialize the optimal value range of the decomposition level K and penalty factor α of the VMD algorithm. The value of K is [4, 15], and the value range of α is [100, 10000]. Use these two parameter ranges to generate a solution space for parameter combination optimization.

[0075] Step 3: Taking the minimum sample entropy mean as the fitness function, the optimal parameter combination is obtained through the GAWOA algorithm, which is the decomposition layer number K is 12 and the penalty factor α is 1587.7.

[0076] Furthermore, the step 3 includes.

[0077] Step 3.1: Initialize the population size of the GAWOA algorithm M = 15, the maximum number of iterations H = 50, and the mutation probability P m = 0.2 and the crossover probability P c = 0.8, initialize the position Z of each individual in the initial whale population in the solution space i (i=1,...,M), set the initial number of iterations to 1.

[0078] Step 3.2: Decompose the vehicle-borne millimeter-wave radar echo signal x(t). The spatial coordinates of each individual whale correspond to a set of VMD decomposition parameter combinations.

[0079] Step 3.3: Use the parameter combination corresponding to each individual whale to perform VMD decomposition on the original signal, calculate the corresponding sample entropy value of each IMF obtained after decomposition, add the sample entropy values ​​of each IMF and take the average value, and save the minimum fitness value F and its corresponding parameter value.

[0080] Step 3.4: Update and optimize the parameters through the GAWOA algorithm, change the position of the whale individual in the solution space, that is, update the value of the parameter combination, and calculate the fitness function value F corresponding to each whale individual in the hth iteration process h , and compare it with the fitness function value corresponding to the current optimal parameter combination, and keep the parameter combination of the whale individual with the smaller fitness function value as the current optimal parameter combination.

[0081] Furthermore, the step 3.4 includes.

[0082] Step 3.4.1: Generate a solution space with the upper and lower limits of the parameters as the limits. Assume that the current optimal candidate parameter combination is the target parameter combination or a combination close to the target parameter combination. After that, all whale individuals in the whale population try to update their positions to the current optimal candidate parameter combination. The equations for this operation are as follows:

[0083] D=|C·X * (h)-X(h)| (1)

[0084] X(h+1)=X * (h)-A·D (2)

[0085] In the above formula, h is the current iteration number, D represents the distance between the current individual and the optimal individual, A and C represent the coefficient vector, and X * (h) is the best position obtained so far. If a better solution exists, X will be updated in each iteration. * (h). The calculation formulas of coefficient vectors A and C are shown in equations (3) and (4), and the update scheme of parameter a in equation (3) is shown in equation (5):

[0086] A=2a×r1-a (3)

[0087] C=2×r2 (4)

[0088]

[0089] Among them, h is the current iteration number, H is the maximum iteration number; in the process of continuous iteration, the value of a gradually decreases nonlinearly from 2 to 0, r1 and r2 are random vectors in the range of [0,1].

[0090] Step 3.4.2: The position of individual whales in the optimization process of the GAWOA algorithm is updated using a spiral update strategy, which is updated through a logarithmic spiral equation. The specific expressions are as follows:

[0091] X(h+1)=D'×e bL ×cos(2πL)+X * (h) (6)

[0092] D'=|X * (h)-X(h)| (7)

[0093] In the above formula: D' represents the distance between the current search individual and the current optimal solution; b is the spiral shape parameter; L is a random number with a uniform distribution in the range of [-1, 1]. Since there are different update strategies in the process of approaching prey, the GAWOA algorithm selects different update strategies according to the probability p.

[0094] When p>0.5, the position of individual whales is updated through a mutation operation strategy similar to the genetic algorithm, that is, the parameter value is updated. The updating method is shown in formula (8), P m The value of ' is shown in formula (9):

[0095] X * (h+1)=X * (h)+P m *D*e bL *cos(2π*L) (8)

[0096]

[0097] In the above formula, X * (h+1) is the updated location of the whale, X * (h) is the current location of the whale, P m is the probability of mutation.

[0098] When p≤0.5 and |A|≥1, the position of individual whales is updated through a strategy similar to the crossover operation of the genetic algorithm. The update method is as follows:

[0099]

[0100] In the above formula, P c is the probability of crossover, X i (h) is the current location of the individual whale, X j (h) is the new position of the individual whale after the traditional whale optimization algorithm update strategy is updated, X i (h+1) and X j (h+1) are the two new positions of the whale individual after the crossover operation. The fitness values ​​of the two new positions are then compared, and the position with the smaller fitness value is determined as the new position of the whale individual after the crossover operation.

[0101] When p≤0.5 and |A|<1, the update method of formula (11) is used to update the position of individual whales:

[0102] X(h+1)=ω*X * (h)-A*D (11)

[0103] In the above formula, X(h+1) is the position of the individual whale after the position update, ω is the adaptive weight, and the adaptive weight ω is calculated according to formula (12).

[0104]

[0105] In the above formula, ω maxis the maximum weight coefficient, ω min is the minimum weight coefficient, H is the maximum number of iterations, and h is the current number of iterations.

[0106] Step 3.4.3: In the process of continuous iteration and updating, the sample signal is decomposed by VMD using the parameters updated after the hth iteration, and the optimal fitness function F of the IMF decomposed after the hth iteration is obtained by calculating and comparing the fitness function values ​​of each individual whale in the whale group during the hth iteration. h , and compare it with the F corresponding to the current optimal parameter combination, leaving the parameter combination of the whale individual with a smaller fitness function F value as the current optimal parameter combination.

[0107] Step 3.5: Determine whether the GAWOA algorithm has reached the condition for iteration termination: whether the current iteration number h of the GAWOA algorithm is greater than or equal to the maximum iteration number H. When h≥H, proceed to step 3.6, otherwise h=h+1, and update the position of the whale group.

[0108] Step 3.6: When h>H, the iteration process ends and the optimal parameter combination is obtained. And the corresponding minimum fitness It is 0.08958.

[0109] Step 4: Perform VMD decomposition on the sample signal using the optimal parameter combination obtained in step 3 to obtain 12 IMFs.

[0110] Step 5: Perform STFT on each IMF obtained by decomposition to obtain the spectrum of each IMF, and filter out the relevant IMF containing the target information based on the spectrum.

[0111] Furthermore, the step 5 includes.

[0112] Step 5.1: Perform STFT on the 12 IMFs obtained after decomposition. The STFT spectrum corresponding to the 12 IMFs after the sample signal is decomposed is as follows: Figure 4 shown.

[0113] Step 5.2: Figure 3 The spectrum diagram of the single-target vehicle-mounted millimeter-wave radar echo signal is used as the screening principle. The IMF containing a single straight line spectrum line is shown on the screening spectrum diagram and is defined as the relevant IMF. The relevant IMFs containing target information obtained by screening are IMF1, IMF2, IMF4 and IMF6.

[0114] Step 6: Analyze and reconstruct the selected relevant IMFs through VI-CFAR to obtain the reconstructed relevant IMFs.

[0115] Furthermore, the step 6 includes.

[0116] Step 6.1: Calculate the threshold of each point in IMF1, IMF2, IMF4 and IMF6 respectively through VI-CFAR.

[0117] Step 6.2: After calculating the threshold, compare the original value with the threshold. If the original value is greater than or equal to the threshold, the original value is retained; otherwise, the original value is replaced by the average of the values ​​corresponding to the two points before and after the point, and the point is reassigned.

[0118] Step 6.3: After completing the operations of the above steps 6.1 and 6.2, the reconstructed IMF1, IMF2, IMF4 and IMF6 are obtained.

[0119] Step 7: Superimpose the reconstructed IMF1, IMF2, IMF4 and IMF6 to obtain the anti-interference vehicle-mounted millimeter-wave two-point echo signal, and finally realize the anti-interference operation of the vehicle-mounted millimeter-wave radar.

[0120] like Figure 5 As shown in , compared with the original interference signal, the noise level of the frequency domain signal after anti-interference is significantly reduced. Figure 6 As shown in the time domain waveform comparison of the signal after anti-interference and the signal without interference, it is easy to see that the signal after anti-interference using this method eliminates a considerable part of the interference signal, reduces the overall interference level in the signal, and the degree of fit between the signal after anti-interference and the signal without interference is also good.

[0121] like Figure 7 As shown in the figure, compared with the signal in the multi-interference situation, the signal-to-interference ratio corresponding to each target signal of the signal after anti-interference using the method in this paper is greatly improved.

[0122] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An anti-interference method for vehicle-mounted millimeter-wave radar echo signals based on VMD and VI-CFAR, comprising a VMD algorithm, a whale optimization algorithm, a genetic algorithm and VI-CFAR, characterized in that: The following steps are included: Step 1: Use the vehicle-mounted millimeter-wave radar to transmit millimeter waves for target detection, and receive the interfered original echo signal x(t); Step 2: Set the parameter range of the VMD algorithm; Step 3: Using the minimum value of the sample entropy mean of the decomposed IMF as the fitness function, the optimal parameter combination of the VMD algorithm is obtained by the improved whale optimization algorithm GAWOA algorithm; it is characterized by comprising the following steps: Step 3.1: Initialize the population size M, maximum number of iterations H, and value range of VMD algorithm parameters of the GAWOA algorithm, and initialize Z i , i=1,...,M, where Z i Each individual whale in the whale population represents a random and different initial parameter combination, and the initial number of iterations is set to 1; Step 3.2: Decompose the vehicle-borne millimeter-wave radar echo signal x(t) within the range of VMD algorithm parameters. Each whale individual corresponds to a set of VMD parameter combinations. Step 3.3: Use the parameter combination corresponding to each whale individual to perform VMD decomposition on the original signal, calculate the sample entropy of each IMF after VMD decomposition, add the sample entropy values ​​of each IMF and calculate the average value, and save the fitness value F and its corresponding parameter value; Step 3.4: Update the parameters through the GAWOA algorithm to optimize the position of each whale in the population, that is, update the value of the parameter combination, and calculate the corresponding fitness function value F h , and compare it with the F corresponding to the current optimal parameter combination, and keep the parameter combination of the whale individual with the smaller fitness function value as the current optimal parameter combination; Step 3.5: Determine whether the GAWOA algorithm has reached the condition for iterative termination: whether the current iteration number h of the GAWOA algorithm is greater than or equal to the maximum iteration number H. If h≥H, proceed to step 3.6; otherwise, h=h+1 and return to step 3.

4. In step 3.4, the parameters are updated and optimized by the GAWOA algorithm, the positions of individual whales are updated, and the values ​​of the parameter combinations are updated, including the following steps: Step 3.4.1: Generate a solution space with the upper and lower limits of the parameters as the limits. Assume that the current optimal candidate parameter combination is the target parameter combination or a combination close to the target parameter combination. After that, all whale individuals in the whale population try to update their positions to the current optimal candidate parameter combination. The equations for this operation are as follows: D=|C·X * (h)-X(h)| (1) X(h+1)=X * (h)-A·D (2) In the above formula, h is the current iteration number, D represents the distance between the current individual and the optimal individual, A and C represent the coefficient vector, and X * (h) is the best position obtained so far. If a better solution exists, X will be updated in each iteration. * (h); The calculation formulas of coefficient vectors A and C are shown in equations (3) and (4), and the update scheme of parameter a in equation (3) is shown in equation (5): A=2a×r1-a (3) C＝2×r2 (4) Where h is the current iteration number, H is the maximum iteration number; in the process of continuous iteration, the value of a gradually decreases nonlinearly from 2 to 0, r1 and r2 are random vectors in the range of [0,1]; Step 3.4.2: The position of individual whales in the optimization process of the GAWOA algorithm is updated using a spiral update strategy, which is updated through a logarithmic spiral equation. The specific expressions are as follows: X(h+1)=D'×e bL ×cos(2πL)+X * (h) (6) D'=|X * (h)-X(h)| (7) In the above formula: D' represents the distance between the current search individual and the current optimal solution; b is the spiral shape parameter; L is a random number with a uniform distribution in the range of [-1, 1]; Since there are different update strategies in the process of approaching the prey, the GAWOA algorithm selects different update strategies according to the probability p; When p>0.5, the position of individual whales is updated through a mutation operation strategy similar to the genetic algorithm, that is, the parameter value is updated. The updating method is shown in formula (8), P m The value of ' is shown in formula (9): X * (h+1)=X * (h)+P m *D*e bL *cos(2π*L) (8) In the above formula, X * (h+1) is the updated location of the whale, X * (h) is the current location of the whale, P m is the probability of mutation; When p≤0.5 and |A|≥1, the position of individual whales is updated through a strategy similar to the crossover operation of the genetic algorithm. The update method is as follows: In the above formula, P c is the probability of crossover, X i (h) is the current location of the individual whale, X j (h) is the new position of the individual whale after the traditional whale optimization algorithm update strategy is updated, X i (h+1) and X j (h+1) are the two new positions of the whale individual after the crossover operation; the fitness values ​​of the two new positions are then compared, and the position with the smaller fitness value is determined as the new position of the whale individual after the crossover operation is performed; When p≤0.5 and |A|<1, the update method of formula (11) is used to update the position of individual whales: X(h+1)=ω*X * (h)-A*D (11) In the above formula, X(h+1) is the position of the individual whale after the position update, ω is the adaptive weight, and the adaptive weight ω is calculated according to formula (12); In the above formula, ω max is the maximum weight coefficient, ω min is the minimum weight coefficient, H is the maximum number of iterations, and h is the current number of iterations; Step 3.6: The iteration process ends and the optimal parameter combination is obtained And the corresponding minimum fitness ; Step 4: Use the optimal parameter combination of the VMD algorithm obtained in step 3 to decompose the original vehicle-mounted millimeter-wave radar echo signal to obtain multiple IMFs; Step 5: Perform STFT on each IMF obtained by decomposition to obtain the spectrum corresponding to each IMF, and filter out the relevant IMF containing the target information based on the spectrum; Step 6: Analyze and reconstruct the selected relevant IMFs through VI-CFAR to obtain the reconstructed relevant IMFs; Step 7: Finally, all the reconstructed related IMFs are superimposed to realize the anti-interference operation of the vehicle-borne millimeter-wave radar echo signal, and finally the anti-interference vehicle-borne millimeter-wave radar echo signal is obtained.

2. The anti-interference method of the vehicle-mounted millimeter-wave radar echo signal based on VMD and VI-CFAR according to claim 1 is characterized in that: In step 2, the VMD decomposition parameters include the number of decomposed modal components K and the quadratic penalty factor α.

3. The anti-interference method of the vehicle-mounted millimeter-wave radar echo signal based on VMD and VI-CFAR according to claim 1 is characterized in that: The step 5 comprises: Step 5.1: Decompose multiple IMFs according to step 4, perform STFT on each IMF, and obtain the corresponding spectrum of each IMF; Step 5.2: Based on the screening principle that the single-target vehicle-mounted millimeter-wave radar echo signal appears as a straight line on the spectrum diagram, the IMF containing the target information in each IMF is determined and defined as the relevant IMF.

4. The anti-interference method of the vehicle-mounted millimeter-wave radar echo signal based on VMD and VI-CFAR according to claim 1 is characterized in that: The step 6 comprises: Step 6.1: Calculate the threshold value corresponding to each data point of each relevant IMF screened in step 5.2 through VI-CFAR; Step 6.2: Compare the actual value of each point with the corresponding threshold value. If the actual value of the point is greater than or equal to the threshold value corresponding to the point, the original value of the point is retained as the value of the point; otherwise, if the value of the point is less than the threshold value corresponding to the point, the point is re-assigned, and the average value of the two points adjacent to the point before and after the point is used as the value of the point instead of the original value; Step 6.3: After reassignment, the reconstructed related IMF is obtained.

Images