CN114527434A - Convolutional neural network-based radar interference effect evaluation method - Google Patents

Convolutional neural network-based radar interference effect evaluation method Download PDF

Info

Publication number
CN114527434A
CN114527434A CN202210127378.2A CN202210127378A CN114527434A CN 114527434 A CN114527434 A CN 114527434A CN 202210127378 A CN202210127378 A CN 202210127378A CN 114527434 A CN114527434 A CN 114527434A
Authority
CN
China
Prior art keywords
radar
neural network
sample
convolutional neural
interference effect
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210127378.2A
Other languages
Chinese (zh)
Inventor
饶鲜
许凯洋
张立东
董阳阳
郭立博
孙宇峥
李梦瑶
李艳慧
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian University
Original Assignee
Xidian University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Priority to CN202210127378.2A priority Critical patent/CN114527434A/en
Publication of CN114527434A publication Critical patent/CN114527434A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/38—Jamming means, e.g. producing false echoes
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/40—Means for monitoring or calibrating
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The invention discloses a radar interference effect evaluation method based on a convolutional neural network, and aims to solve the problems that in the prior art, evaluation can be performed only when radar information is acquired from an interference party and a radar party at the same time, and evaluation index weight is artificially set. The method comprises the following implementation steps: (1) generating a training data set; (2) building a convolutional neural network; (3) training a convolutional neural network; (4) the interference effect is evaluated. The method has the advantages that the radar parameters can be evaluated only by acquiring the radar parameters from the interference party, and the evaluation can be carried out without manually setting evaluation index weight.

Description

Convolutional neural network-based radar interference effect evaluation method
Technical Field
The invention belongs to the technical field of radar communication, and further relates to a radar interference effect evaluation method based on a convolutional neural network in the technical field of electronic countermeasure. The method and the device can be used for evaluating the effect of whether the interference party effectively interferes with the enemy radar.
Background
Radar countermeasure is an important component of electronic countermeasure, and in radar countermeasure, the interference effect is an index which is very concerned and is urgently desired in the whole interference system. With the intellectualization of radar countermeasure equipment, the multifunctionalization of a radar system and the complication of a radiation source environment where the radar countermeasure is located, the evaluation of the interference effect of the radar also becomes more difficult, so that the interference effect evaluation technology is more emphasized in the field of the radar countermeasure.
Tongguangfu et al, in its published article, "evaluation of synergistic interference efficacy based on analytic hierarchy process" ("electronic information countermeasure technology" 2016,31 (04): 58-32+78), disclose a method for evaluating radar interference effect by analytic hierarchy process. The implementation scheme of the method is as follows: firstly, selecting an evaluation index: and constructing an evaluation index system according to a power criterion, a probability criterion and an efficiency criterion, wherein the evaluation index system comprises radar power, ranging precision, the number of false targets and track quality. Secondly, calculating an evaluation index: and (4) giving out an important linear matrix by integrating the expert evaluation opinions, and calculating the membership degree of each index. And thirdly, calculating the membership degree to obtain a final evaluation result. The method can perform qualitative and quantitative analysis on the evaluation result to obtain the interference effect result. However, the method still has the disadvantages that in an actual application environment, because the interfering party and the radar party are non-cooperative parties, the radar information of the interfering party can be obtained only through radar reconnaissance, and the evaluation index of the radar interference effect evaluation system established according to the method can only depend on the radar reconnaissance to obtain information of one party, but not on the evaluation index obtained under cooperation of the two parties, so that the method cannot realize that the interference effect evaluation is completed only by the interfering party.
The patent document "a radar interference effect evaluation method, device and computer equipment" applied by the university of electronic technology of west ampere "(application number: 201910229296.7; application publication number: CN110082733A) proposes a radar interference effect evaluation method. The implementation scheme of the method is as follows: the method comprises the steps of obtaining interference benefit values of radar interference factors including interference opportunity, interference frequency, interference range, interference pattern and the like through a preset method instead of radar interference evaluation indexes obtained through cooperation of two parties, obtaining interference evaluation scores input by a user, determining evaluation index weights through expert evaluation opinions, obtaining weight vectors through an analytic hierarchy process, obtaining a weighted interference benefit matrix of the interference factors through the interference benefit values and the weight vectors of the interference factors, and obtaining a radar interference effect evaluation value through a double base point method (TOPSIS). The method can evaluate the interference effect in actual conditions, but still has the defects that the evaluation result needs to determine the evaluation index weight by means of the interference evaluation score of the user and the expert evaluation opinion, and the interference cannot be objectively and reasonably evaluated due to the influence of human factors.
Disclosure of Invention
The invention aims to provide a radar interference effect evaluation method based on a convolutional neural network aiming at the defects in the prior art, so that the problem that indexes can be obtained only by cooperation of an interference party and a radar party in the prior art is solved, the radar interference evaluation can be objectively and correctly carried out, and effective interference on an enemy radar under an actual application scene is ensured.
The idea for realizing the purpose of the invention is as follows: when the training data set is generated, the radar characteristic parameter set including pulse repetition frequency, bandwidth, pulse width, amplitude and carrier frequency nuclear beam residence time is constructed firstly, and the constructed radar characteristic parameter set is obtained from an interference party only and is not obtained from a radar party, so that the problem that the index set construction can be completed only by depending on information on two sides of the radar party and the interference party in the prior art is solved. The invention constructs a convolutional neural network, and because the convolutional neural network has the capability of processing the classification problem, the interference effect evaluation essentially belongs to the classification problem, so the interference effect evaluation can be carried out by the characteristic that the convolutional neural network processes the classification problem. The convolutional neural network is trained through a training set consisting of pulse repetition frequency, bandwidth, pulse width, amplitude and carrier frequency kernel beam residence time containing radar characteristic parameters, so that the trained convolutional neural network has good nonlinear mapping capability, when the radar interference effect is evaluated, the radar characteristic parameters can be rapidly extracted from test data through the trained convolutional neural network, and corresponding interference evaluation grades can be obtained to finish the interference effect evaluation.
The specific steps for realizing the purpose of the invention are as follows:
step 1, generating a training data set:
(1a) at least 3000 samples are combined into a sample set, and the sample set comprises attributes of 6 radar characteristic parameters: dividing 6 radar characteristic parameters into three types according to interference evaluation levels, wherein each type at least comprises 1000 samples;
(1b) normalizing each sample in the sample set;
(1c) adding an interference effect evaluation grade label to each sample;
(1d) forming a training data set by all samples and the grade labels corresponding to all samples;
step 2, constructing a convolutional neural network:
a10-layer convolutional neural network is constructed, and the structure sequentially comprises the following steps: the device comprises a first convolution layer, a ReLu active layer, a first pooling layer, a second convolution layer, a ReLu active layer, a second pooling layer, a third convolution layer, a ReLu active layer, a third pooling layer and a full-connection layer; the number of convolution kernels of the first convolution layer, the second convolution layer, the third convolution layer and the fourth convolution layer is set to be 32, 16 and 8 in sequence, the sizes of the convolution kernels are all set to be 1 x 3, the first convolution layer, the second convolution layer and the third convolution layer are all in a maximum pooling mode, the sizes of the pooling kernels are all set to be 1 x 2, and the pooling step length is all set to be 1 x 2;
step 3, training a convolutional neural network:
inputting the radar characteristic parameters in the training set into a convolutional neural network, and iteratively updating parameters of each layer of the convolutional neural network by using a back propagation gradient descent method until a loss function of the convolutional neural network is converged to obtain a trained convolutional neural network;
and 4, evaluating the interference effect:
performing normalization processing on all samples of the interference effect to be evaluated by adopting the method in the step (1 b); and inputting the normalized sample into a trained convolutional neural network, and outputting an evaluation result of the interference effect.
Compared with the prior art, the invention has the following advantages:
firstly, the training data set generated by the method is generated from the perspective of an interference party, the defect that in the prior art, the evaluation can be performed only by acquiring radar information from the interference party is overcome, and the evaluation can be completed only by acquiring the radar information from the radar party, so that the method has the advantage that the evaluation of the interference effect can be performed without acquiring a large amount of radar information.
Secondly, the weight value of the convolutional neural network constructed by the invention is adaptively changed according to actual parameters, so that the defect that the evaluation can be carried out only by manually setting a weight coefficient in the prior art is overcome, and the convolutional neural network has the advantage of evaluating the radar interference effect under an objective condition.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The specific steps of the present invention will be described in further detail with reference to fig. 1 and examples.
Step 1, generating a training data set.
In the embodiment of the invention, 3000 samples are selected to form a sample set, and the sample set comprises attributes of 6 radar characteristic parameters: the pulse repetition frequency, the bandwidth, the pulse width, the amplitude, the carrier frequency and the beam dwell time of the radar are divided into three types according to the interference evaluation level, and each type at least comprises 1000 samples.
The interference evaluation level is obtained by the following formula:
Figure BDA0003501050020000041
wherein G isiRepresenting the interference evaluation grade of the ith sample in the sample set, if 0 is less than or equal to GiThe interference rating at 100 or less is "poor", if 100<GiThe interference rating at ≦ 200 is "normal", if 200 < GiWhen the interference evaluation level is equal to or less than 300, i represents the serial number of the samples in the sample set, i is 1,2iRepresenting the repetition frequency, BW, of the radar pulse of the ith sample in the set of samplesiRepresenting the bandwidth, PW, of the radar pulse of the ith sample in the set of samplesiPulse width, PA, of radar pulse representing ith sample in sample setiRepresenting the amplitude, RF, of the radar pulse at the ith sample in the setiCarrier frequency, T, of radar pulse representing ith sample in sample setiRepresenting the beam dwell time of the radar pulse for the ith sample in the sample set.
The normalization is performed for each sample in the sample set as follows.
Figure BDA0003501050020000042
Wherein x isij' represents the property of j-th radar characteristic parameter of the ith sample in the normalized sample set, j represents the serial number of the property of the radar characteristic parameter, and j equals to 1 represents sampleThe repetition frequency of the radar pulse in the sample set is represented by j 2, the bandwidth of the radar pulse in the sample set is represented by j 3, the pulse width of the radar pulse in the sample set is represented by j 4, the amplitude of the radar pulse in the sample set is represented by j 5, the carrier frequency of the radar pulse in the sample set is represented by j 6, the beam dwell time of the radar pulse in the sample set is represented by xijAn attribute, x, representing the jth radar feature parameter in the ith sample of the set of samplesjmaxRepresenting the maximum value, x, of the jth radar feature parameter attribute in the sample setjminRepresenting the minimum value of the jth radar feature parameter attribute in the sample set.
And (3) adding an interference effect evaluation grade label to each sample, setting the label of the interference effect evaluation grade 'poor' as 0, setting the label of the interference effect evaluation grade 'general' as 1, and setting the label of the interference effect evaluation grade 'good' as 2.
And (4) combining all the samples and the grade labels corresponding to each sample into a training data set.
And 2, building a convolutional neural network.
A10-layer convolutional neural network is constructed, and the structure sequentially comprises the following steps: the device comprises a first convolution layer, a ReLu active layer, a first pooling layer, a second convolution layer, a ReLu active layer, a second pooling layer, a third convolution layer, a ReLu active layer, a third pooling layer and a full-connection layer; the number of convolution kernels of the first convolution layer, the second convolution layer, the third convolution layer and the fourth convolution layer is set to be 32, 16 and 8 in sequence, the sizes of the convolution kernels are all set to be 1 x 3, the first convolution layer, the second convolution layer and the third convolution layer are all in a maximum pooling mode, the sizes of the pooling kernels are all set to be 1 x 2, and the pooling step length is all set to be 1 x 2.
And 3, training the convolutional neural network.
And inputting the radar characteristic parameters in the training set into the convolutional neural network, and iteratively updating parameters of each layer of the convolutional neural network by using a back propagation gradient descent method until the loss function of the convolutional neural network is converged to obtain the trained convolutional neural network.
The loss function calculation formula is as follows:
Figure BDA0003501050020000051
where Loss represents a Loss function of the convolutional neural network, Σ represents a summation operation, c is 0,1.. m, m represents the total number of interference effect evaluation levels, and y represents the total number of interference effect evaluation levelsicRepresenting the relation between the predicted value and the real value of the interference effect evaluation grade c of the ith sample in the training data set, and y is when the real value is equal to the predicted valueicWhen the true value is not equal to the predicted value y 1ic＝0，picAnd the predicted probability of the interference effect evaluation grade c of the ith sample in the training data set is represented.
And 4, evaluating the interference effect.
Performing normalization processing on all samples of the interference effect to be evaluated by adopting the same normalization method as the step 1; and inputting the normalized sample into a trained convolutional neural network, and outputting an evaluation result of the interference effect.
The effect of the present invention is further explained by combining the simulation experiment as follows:
1. simulation experiment conditions are as follows:
the hardware platform of the simulation experiment of the invention is as follows: the processor is an Intel i 77700 k CPU, the main frequency is 3.2GHz, and the memory is 16 GB.
The software platform of the simulation experiment of the invention is as follows: windows 10 operating system and python 3.6.
The parameters of a training set and a testing set used by the simulation experiment of the invention are set as follows: the pulse repetition frequency range of the radar is 2-20kHz, the bandwidth range is 0.2-10MHz, the pulse width range is 1-200 mus, the amplitude range is 10-60kW, the carrier frequency range is 2.9-3.1GHz, and the beam residence time range is 1-5 s.
The training set comprises 3000 samples and corresponding interference evaluation grade values, the testing set comprises 1000 samples and corresponding interference evaluation grade values, and the testing set is utilized to evaluate the interference effect evaluation result.
The convolutional neural network configuration parameters of the simulation experiment of the invention are as follows: the training batch size is set to 100, the learning rate is 0.0005, the loss rate is 0.5, the total number of training rounds is 500, and network parameters of the convolutional neural networks which are trained by using the training set in 100 rounds, 200 rounds, 300 rounds, 400 rounds and 500 rounds are recorded respectively.
2. Simulation content and result analysis thereof:
the simulation experiment of the invention is that the method of the invention is adopted to input 1000 samples in a test set into a trained convolutional neural network after normalization processing is completed, and the interference evaluation grade value of the invention is obtained.
In the simulation experiment, the adopted prior art convolutional neural network CNN classification method is a convolutional neural network SAR image classification method proposed by macchia in its published paper, "convolutional neural network-based SAR image classification" (university of electronic science and technology, major academic paper, 2021), which is referred to as convolutional neural network classification method for short.
The method of the invention is adopted to obtain the interference evaluation grade of each sample in the test set. The interference assessment levels are classified into three categories, namely "poor", "normal" and "good". Testing the test set by using the convolutional neural network which trains 100 rounds, 200 rounds, 300 rounds, 400 rounds and 500 rounds, wherein the accuracy of the interference effect evaluation grade 'poor' is the number of samples which are correctly predicted in the test set divided by the total number of samples of the interference effect evaluation grade 'poor'; the accuracy of the interference effect evaluation level 'general' is that the number of samples correctly predicted in the test set is divided by the total number of samples of the interference effect evaluation level 'general'; the accuracy of the interference effect evaluation level "good" is obtained by dividing the number of correctly predicted samples in the test set by the total number of samples of the interference effect evaluation level "good", and the interference effect evaluation level prediction result is obtained as shown in table 1.
TABLE 1 interference effect evaluation grade prediction Table (Unit: percent)
Number of training rounds 100 wheels 200 wheels 300 wheels 400 wheel 500 rounds
Interference effect evaluation rating "poor" 95.21 95.33 95.40 95.52 95.60
Interference effect evaluation level "general" 96.88 96.92 97.01 97.15 97.20
Interference effect evaluation level "very good" 95.85 95.99 96.20 96.23 96.40
As can be seen from the table 1, when the convolutional neural network is trained for 500 rounds, the accuracy of the method for predicting the poor interference evaluation grade is 95.60%, the accuracy of the method for predicting the general interference evaluation grade is 97.20%, and the accuracy of the method for predicting the good interference evaluation grade is 96.40%.

Claims (5)

1. A radar interference effect evaluation method based on a convolutional neural network is characterized in that a training set containing radar characteristic parameter attributes is generated, the convolutional neural network with a classification function is constructed by utilizing the training set to train, and the method specifically comprises the following steps:
step 1, generating a training set:
(1a) at least 3000 samples are combined into a sample set, and the sample set comprises attributes of 6 radar characteristic parameters: dividing 6 radar characteristic parameters into three types according to interference evaluation levels, wherein each type at least comprises 1000 samples;
(1b) normalizing each sample in the sample set;
(1c) adding an interference effect evaluation grade label to each sample;
(1d) forming a training set by all samples and the grade labels corresponding to all the samples;
step 2, constructing a convolutional neural network:
a10-layer convolutional neural network is constructed, and the structure sequentially comprises the following steps: the device comprises a first convolution layer, a ReLu active layer, a first pooling layer, a second convolution layer, a ReLu active layer, a second pooling layer, a third convolution layer, a ReLu active layer, a third pooling layer and a full-connection layer; the number of convolution kernels of the first convolution layer, the second convolution layer, the third convolution layer and the fourth convolution layer is set to be 32, 16 and 8 in sequence, the sizes of the convolution kernels are all set to be 1 x 3, the first convolution layer, the second convolution layer and the third convolution layer are all in a maximum pooling mode, the sizes of the pooling kernels are all set to be 1 x 2, and the pooling step length is all set to be 1 x 2;
step 3, training a convolutional neural network:
inputting the radar characteristic parameters in the training set into a convolutional neural network, and iteratively updating parameters of each layer of the convolutional neural network by using a back propagation gradient descent method until a loss function of the convolutional neural network is converged to obtain a trained convolutional neural network;
and 4, evaluating the interference effect:
performing normalization processing on all samples of the interference effect to be evaluated by adopting the same method as the step (1 b); and inputting the normalized sample into a trained convolutional neural network, and outputting an evaluation result of the interference effect.
2. The convolutional neural network based radar interference effect evaluation method as claimed in claim 1, wherein the interference evaluation level in step (1a) is obtained by the following formula:
Figure FDA0003501050010000011
wherein G isiRepresenting the interference evaluation grade of the ith sample in the sample set, if 0 is less than or equal to GiThe interference rating at 100 or less is "poor", if 100<GiThe interference rating at ≦ 200 is "normal", if 200 < GiWhen the interference evaluation level is equal to or less than 300, i represents the serial number of the samples in the sample set, i is 1,2iRepresenting the repetition frequency, BW, of the radar pulse of the ith sample in the set of samplesiRepresenting the bandwidth, PW, of the radar pulse of the ith sample in the set of samplesiPulse width, PA, of radar pulse representing ith sample in sample setiRepresenting the amplitude, RF, of the radar pulse at the ith sample in the setiCarrier frequency, T, of radar pulse representing ith sample in sample setiRepresenting the beam dwell time of the radar pulse for the ith sample in the sample set.
3. The convolutional neural network based radar interference effect evaluation method as claimed in claim 1, wherein the normalization operation in step (1b) is implemented by the following equation:
Figure FDA0003501050010000021
wherein x isij' denotes an attribute of a j-th radar feature parameter of an ith sample in the normalized sample set, j denotes a serial number of the radar feature parameter attribute, j ═ 1 denotes a repetition frequency of a radar pulse in the sample set, j ═ 2 denotes a bandwidth of the radar pulse in the sample set, j ═ 3 denotes a pulse width of the radar pulse in the sample set, j ═ 4 denotes an amplitude of the radar pulse in the sample set, j ═ 5 denotes a carrier frequency of the radar pulse in the sample set, j ═ 6 denotes a beam dwell time of the radar pulse in the sample set, and x ═ 6 denotes a beam dwell time of the radar pulse in the sample setijAn attribute, x, representing the jth radar feature parameter in the ith sample of the set of samplesjmaxRepresenting the maximum value, x, of the jth radar feature parameter attribute in the sample setjminRepresenting the minimum value of the jth radar feature parameter attribute in the sample set.
4. The convolutional neural network-based radar interference effect evaluation method as claimed in claim 1, wherein the interference effect evaluation level label is assigned to each sample in step (1c), the interference effect evaluation level "poor" label is set to 0, the interference effect evaluation level "general" label is set to 1, and the interference effect evaluation level "good" label is set to 2.
5. The convolutional neural network-based interference effect evaluation method as set forth in claim 2, wherein: the loss function of the convolutional neural network in the step (3) is as follows:
Figure FDA0003501050010000022
wherein, Loss represents a Loss function of the convolutional neural network, sigma represents a summation operation, c is 0,1icRepresenting the relation between the predicted value and the real value of the interference effect evaluation grade c of the ith sample in the training data set, and y is when the real value is equal to the predicted valueicWhen the true value is not equal to the predicted value y 1ic＝0，picAnd the predicted probability of the interference effect evaluation grade c of the ith sample in the training data set is represented.
CN202210127378.2A 2022-02-11 2022-02-11 Convolutional neural network-based radar interference effect evaluation method Pending CN114527434A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210127378.2A CN114527434A (en) 2022-02-11 2022-02-11 Convolutional neural network-based radar interference effect evaluation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210127378.2A CN114527434A (en) 2022-02-11 2022-02-11 Convolutional neural network-based radar interference effect evaluation method

Publications (1)

Publication Number Publication Date
CN114527434A true CN114527434A (en) 2022-05-24

Family

ID=81623006

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210127378.2A Pending CN114527434A (en) 2022-02-11 2022-02-11 Convolutional neural network-based radar interference effect evaluation method

Country Status (1)

Country Link
CN (1) CN114527434A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116243252A (en) * 2023-03-14 2023-06-09 电子科技大学 LSTM-based multifunctional radar working mode prediction method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116243252A (en) * 2023-03-14 2023-06-09 电子科技大学 LSTM-based multifunctional radar working mode prediction method
CN116243252B (en) * 2023-03-14 2023-09-19 电子科技大学 LSTM-based multifunctional radar working mode prediction method

Similar Documents

Publication Publication Date Title
CN108196132B (en) Method for evaluating electromagnetic compatibility in system
CN109376766B (en) Portrait prediction classification method, device and equipment
CN112070357A (en) Radar radiation source threat assessment method based on improved BP neural network
Lukić Analysis if the Efficiency of Trade in Oil Derivatives in Serbia by Applying the Fuzzy AHP-TOPSIS Method
CN112054943B (en) Traffic prediction method for mobile network base station
CN109543763A (en) A kind of Raman spectrum analysis method based on convolutional neural networks
CN112668476B (en) Data processing method and device, electronic equipment and storage medium
CN107688863A (en) The short-term wind speed high accuracy combination forecasting method that adaptive iteration is strengthened
CN114527434A (en) Convolutional neural network-based radar interference effect evaluation method
CN113822499A (en) Train spare part loss prediction method based on model fusion
Rahman et al. Data quality improvement by imputation of missing values
Baran et al. Statistical post‐processing of heat index ensemble forecasts: is there a royal road?
Costa et al. Automatic classification of computational thinking skills in elementary school math questions
Hart et al. Accuracy and power analysis of social networks built from count data
CN115616503A (en) Radar interference signal type identification method based on convolutional neural network model
CN112616160B (en) Intelligent short-wave frequency cross-frequency-band real-time prediction method and system
CN111160419B (en) Deep learning-based electronic transformer data classification prediction method and device
CN110889635B (en) Method for performing emergency drilling on food safety event processing
CN113537759A (en) User experience measurement model based on weight self-adaptation
Ivan et al. Applying Multi-Criteria Decision-Making to the Technology Investment Decision-Making Process
Hortua Constraining cosmological parameters from N-body simulations with Bayesian Neural Networks
CN114037091B (en) Expert joint evaluation-based network security information sharing system, method, electronic equipment and storage medium
Gupta et al. Input modeling using quantile statistical methods
CN113055111B (en) Channel modeling method and system based on Bayesian optimization
Thalia et al. Towards quantification of information system security

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination