CN113156382B - Signal identification method and device for vehicle-mounted range radar - Google Patents
Signal identification method and device for vehicle-mounted range radar Download PDFInfo
- Publication number
- CN113156382B CN113156382B CN202110397704.7A CN202110397704A CN113156382B CN 113156382 B CN113156382 B CN 113156382B CN 202110397704 A CN202110397704 A CN 202110397704A CN 113156382 B CN113156382 B CN 113156382B
- Authority
- CN
- China
- Prior art keywords
- signal
- signals
- sampling
- amplitude
- frequency
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 239000013598 vector Substances 0.000 claims abstract description 71
- 238000005070 sampling Methods 0.000 claims abstract description 65
- 230000004044 response Effects 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 16
- 230000009466 transformation Effects 0.000 claims abstract description 4
- 230000003595 spectral effect Effects 0.000 claims description 12
- 238000002156 mixing Methods 0.000 claims description 5
- 230000002452 interceptive effect Effects 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 2
- 230000010355 oscillation Effects 0.000 claims 2
- 230000006870 function Effects 0.000 description 16
- 238000012360 testing method Methods 0.000 description 10
- 239000000523 sample Substances 0.000 description 9
- 238000012706 support-vector machine Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000010200 validation analysis Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000010998 test method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G01S7/36—Means for anti-jamming, e.g. ECCM, i.e. electronic counter-counter measures
-
- 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
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a signal identification method and a device for a vehicle-mounted range radar, wherein the method comprises the following steps: acquiring a plurality of sampling signals, wherein each sampling signal is a target echo signal or an interference signal; processing each sampling signal to obtain a plurality of processed signals; performing Fourier transformation on each processed signal to obtain amplitude-frequency response information corresponding to each sampling signal; determining a feature vector corresponding to each sampling signal according to the amplitude-frequency response information; training a classification vector machine by utilizing the feature vector; and identifying the received signals by using the trained two-classification vector machine so as to determine whether the received signals belong to the target echo signals or the interference signals.
Description
Technical Field
The invention relates to the technical field of vehicle-mounted radars, in particular to the field of signal identification methods and devices for vehicle-mounted range radars.
Background
In practical application, the vehicle-mounted range radars are mounted on a large number of automobiles and work simultaneously due to complex and changeable road traffic conditions, and when the respective working frequencies are very close, mutual interference is likely, so that false alarms are likely to occur. In addition, other objects on the road surface can generate certain noise to the vehicle-mounted radar, and as the total output of the noise is a random variable, the vehicle-mounted radar can judge whether targets exist or not according to whether the signal amplitude exceeds a threshold, so that false targets possibly exist, the false targets do not exist in reality, and the false targets appear as real targets to the radar probe. If the interference signal falls into the bandwidth of the radar receiver, a false alarm phenomenon occurs, and in practice the interference information should be detected and suppressed during signal processing. The waveforms, timing, bandwidth, antenna pattern and signal processing used by each manufacturer are typically slightly different, which is an advantage in terms of interference signal rejection, but results in different radar responses to interference. Thus, testing different kinds of noise will more fully reflect these interference problems.
The different types of noise have a plurality of characteristic parameters, and the traditional threshold value determining method can not determine the value between a target signal and an interference signal, so that a classifier of a Support Vector Machine (SVM) can well process the problem of irregular data through nonlinear mapping.
Disclosure of Invention
First, the technical problem to be solved
The invention discloses a signal identification method and device for a vehicle-mounted range radar, which aim to at least partially solve the technical problems.
(II) technical scheme
To achieve the above object, an embodiment of the present invention provides a signal recognition method for a vehicle-mounted range radar, including: acquiring a plurality of sampling signals, wherein each sampling signal is a target echo signal or an interference signal; processing each sampling signal to obtain a plurality of processed signals; performing Fourier transformation on each processed signal to obtain amplitude-frequency response information corresponding to each sampling signal; determining a feature vector corresponding to each sampling signal according to the amplitude-frequency response information; training a classification vector machine by utilizing the feature vector; and identifying the received signals by using the trained two-classification vector machine so as to determine whether the received signals belong to the target echo signals or the interference signals.
According to an embodiment of the invention, the interference signal comprises at least one of a noise amplitude modulated interference signal and a sinusoidal amplitude modulated interference signal.
According to an embodiment of the present invention, processing each of the sampled signals to obtain a plurality of processed signals includes: calculating the average value of the time domain amplitude of each sampling signal; subtracting the average value of the time domain amplitude of each sampling signal from the time domain amplitude of each sampling signal to obtain the processed signal.
According to an embodiment of the invention, the feature vector of each of the sampled signals comprises at least one of a mean value, a standard deviation, a kurtosis, a difference between maximum and minimum values, and an average power spectral density of amplitude-frequency response information of each of the sampled signals.
According to an embodiment of the invention, the kurtosis is calculated as follows:
wherein K is kurtosis, s (f) i Representing the amplitude-frequency response information of each of the sampled signals, sigma representing the standard deviation of the amplitude-frequency response of each of the sampled signals,and n is the number of sampling points of each sampling signal, and i is any positive integer from 1 to n.
According to an embodiment of the invention, the difference between the maximum and minimum values is calculated as follows: pk=max-min, where pk is the difference between the maximum value and the minimum value, max is the maximum value of the amplitude-frequency response information of each sampling signal, and min is the minimum value of the amplitude-frequency response information of each sampling signal.
According to an embodiment of the invention, the average power spectral density is calculated by:
wherein E is average power spectral density, f is frequency of each sampling signal, s (f) is amplitude-frequency response information of each sampling signal, and N is number of sampling points of each sampling signal.
According to an embodiment of the present invention, a gaussian radial basis function is preset in the bi-classification vector machine, and training the bi-classification vector machine using the feature vector includes: obtaining a feature vector sample; carrying out normalization processing on the feature vector samples to obtain target feature vector samples, wherein the maximum value and the minimum value of each type of feature vector in the target feature vector samples are respectively kept consistent; and inputting the target feature vector sample into a two-class vector machine, and outputting a class decision function containing target parameters, wherein the target parameters comprise a penalty parameter C and a kernel function parameter gamma.
According to an embodiment of the present invention, identifying the received signal by using the trained binary vector machine to determine whether the received signal belongs to the target echo signal or the interference signal includes: identifying the received signals by using the classification decision function to obtain interface information for distinguishing the target echo signals or the interference signals; and determining the target echo signal or the interference signal according to the interface information.
The invention further provides a signal identification device for the vehicle-mounted range radar, which is used for realizing the method.
(III) beneficial effects
According to the method and the device, signal identification can be performed according to different characteristics of amplitude-frequency response of the target signal and the interference signal, so that the ranging capability of the vehicle-mounted range radar in a complex road environment is improved.
Drawings
Fig. 1 schematically shows a flow chart of a signal recognition method for a vehicle-mounted range radar according to an embodiment of the invention;
FIG. 2 schematically illustrates a resulting graph of amplitude-frequency response information of a target echo signal in accordance with an embodiment of the invention;
FIG. 3 schematically illustrates a resulting graph of amplitude-frequency response information of a noise amplitude modulated interfering signal in accordance with an embodiment of the invention;
FIG. 4 schematically illustrates a resulting graph of amplitude-frequency response information of a sinusoidal amplitude modulated jamming signal in accordance with an embodiment of the present invention;
fig. 5 schematically shows a flow chart of the acquisition of feature vectors according to an embodiment of the invention.
Detailed Description
The present invention will be further described in detail below with reference to specific embodiments and with reference to the accompanying drawings, in order to make the objects, technical solutions and advantages of the present invention more apparent.
An embodiment of the present invention provides a signal identification method for a vehicle-mounted range radar, including: acquiring a plurality of sampling signals, wherein each sampling signal is a target echo signal or an interference signal; processing each sampling signal to obtain a plurality of processed signals; performing Fourier transform on each processed signal to obtain amplitude-frequency response information corresponding to each sampling signal; determining a feature vector corresponding to each sampling signal according to the amplitude-frequency response information; training a classification vector machine by using the feature vector; and identifying the received signals by using the trained binary vector machine to determine whether the received signals belong to target echo signals or interference signals.
Fig. 1 schematically shows a flow chart of a signal recognition method for a vehicle-mounted range radar according to an embodiment of the invention.
As shown in fig. 1, the flow includes operations S101 to S105.
In operation S101, a target echo signal and an interference signal are acquired.
According to an embodiment of the present invention, the target echo signal and the interference signal constitute, for example, the plurality of sampling signals, which are, for example, sampling signals in the form of doppler frequencies. Taking a specific embodiment as an example, the sampling process of each target echo signal or interference signal may, for example, include: the method comprises the steps of transmitting a sinusoidal signal by a modulating signal generator, forming a signal with a center frequency of 1GHz (which is only an example and can be randomly adjusted according to practical application scenes, the same applies below) through an oscillator, enabling frequency change to be a triangular wave signal, namely a local oscillator, enabling offset of the local oscillator to be 25MHz/V (which is only an example), generating a local oscillator by the modulating signal generator and the oscillator, generating a return signal through a circulator, mixing the local oscillator and the return signal, generating a difference frequency signal and a sum frequency signal, enabling the difference frequency signal to pass through a band-pass filter, obtaining a required harmonic frequency, enabling the distance to be fixed, enabling the harmonic signal to be subjected to secondary mixing, and obtaining Doppler frequency after filtering, namely the target echo signal.
According to an embodiment of the present invention, the interference signal may include at least one of a noise amplitude-modulated interference signal and a sinusoidal amplitude-modulated interference signal, for example. After the target echo signal is acquired in the foregoing manner, for example, the original circulator in the process for acquiring the target echo signal may be replaced with a sinusoidal signal emitter and a noise emitter, where the center frequency is aligned to the center frequency of the target echo signal, the frequency of the interference signal is, for example, 50Hz (only for example), and the step size (the time taken for each sweep is several times, and the coverage of all the required sweep rates is completed after sweeping) is 0.001s, for 500 steps. By changing the interference frequency of the interference signal and increasing the distance, the double-pass distance of the echo signal is changed into a single-pass distance, and for example, the acquisition of the noise amplitude modulation interference signal and the sine amplitude modulation interference signal can be realized.
According to the embodiment of the invention, for example, 100 target echo signals under the action of no interference, 100 noise amplitude modulation interference signals and 100 sine amplitude modulation interference signals can be collected as sampling signals, the sampling frequency is for example 100kHz, the sampling time is for example 0.019s, and it is required to ensure that all amplitude-frequency response information is contained in each time period when the sampling time is determined.
In operation S102, a fast fourier transform is performed.
According to an embodiment of the present invention, the operation may further include, for example, processing each of the sampled signals to obtain a plurality of processed signals, and performing fourier transform on each of the processed signals to eliminate a zero frequency component, i.e., a direct current component, and obtain amplitude-frequency response information corresponding to each of the sampled signals.
Fig. 2 schematically shows a result diagram of amplitude-frequency response information of a target echo signal according to an embodiment of the invention.
Fig. 3 schematically shows a resulting graph of amplitude-frequency response information of a noise amplitude modulated interfering signal according to an embodiment of the invention.
Fig. 4 schematically shows a resulting graph of amplitude-frequency response information of a sinusoidal amplitude modulated interfering signal according to an embodiment of the invention.
As shown in fig. 2 to 4, it can be seen that the amplitude-frequency response peak value of the target echo signal is prominent and the amplitude-frequency response peak value of the noise signal is small.
According to an embodiment of the present invention, the above-mentioned manner of processing each sampling signal may include, for example: calculating the average value of the time domain amplitude of each sampling signal; subtracting the average value of the time domain amplitude of each sampling signal from the time domain amplitude of each sampling signal to obtain the processed signal.
In operation S103, a feature vector is extracted.
According to an embodiment of the invention, the operation may for example comprise determining a feature vector corresponding to each sampled signal from the amplitude-frequency response information described above, wherein the feature vector of each sampled signal may for example comprise at least one of a mean, a standard deviation, a kurtosis, a difference between maximum and minimum values and an average power spectral density of the amplitude-frequency response information of each sampled signal.
Fig. 5 schematically shows a flow chart of the acquisition of feature vectors according to an embodiment of the invention.
As shown in fig. 5, the process includes calculating a mean value of a frequency spectrum, a standard deviation, a kurtosis of the frequency spectrum, a difference between maximum and minimum values, an average power spectral density, and determining a feature vector for amplitude-frequency response information obtained after fourier transformation.
According to the embodiment of the invention, the amplitude-frequency response mean value, standard deviation, kurtosis and the difference between the maximum value and the minimum value of the processed signal and the average power spectral density can be obtained through a known function.
According to an embodiment of the present disclosure, the kurtosis may also be calculated, for example, by the following formula:
wherein s (f) i Amplitude-frequency response information representing each of the sampled signalsSigma represents the standard deviation of the amplitude-frequency response of each of said sampled signals,and n is the number of sampling points of each sampling signal, and i is any positive integer from 1 to n.
According to an embodiment of the present disclosure, the difference between the maximum and minimum values may also be calculated, for example, by:
pk＝max-min
wherein pk is the difference between the maximum value and the minimum value, max is the maximum value of the amplitude-frequency response information of each sampling signal, and min is the minimum value of the amplitude-frequency response information of each sampling signal.
According to embodiments of the present disclosure, the average power spectral density may also be calculated, for example, by:
wherein E is average power spectral density, f is frequency of each sampling signal, s (f) is amplitude-frequency response information of each sampling signal, and N is number of sampling points of each sampling signal.
According to an embodiment of the present disclosure, referring to fig. 5, in determining a feature vector, for example, the kurtosis, the difference between the maximum and minimum values, and the average power spectral density calculated above may be used as the feature vector, and the feature vector may be in the form of d= [ K, pk, E ] as an input supporting a two-class vector machine.
In order to better know whether the feature vector is valid or not, according to the embodiment of the invention, kruskal-Wallis is used for checking whether the distribution of a plurality of populations has significant difference analysis, which is originally assumed to be H 0 The independent samples are from the same population or the multiple populations generating the independent samples obey the same distribution Kruskal-Wallis to carry out non-parametric analysis on the data and then return the p-value of the test result, wherein the p-value is an important parameter of the hypothesis testThe smaller the value, the more pronounced the result, i.e. rejection of H 0 The more adequate the reason for (1) or the more confident we have to reject H 0 While receiving alternative hypothesis H 1 There is a significant difference in the distribution of the populations and the results of the test are considered reliable for p-values less than 0.01.
In operation S104, a bi-class vector machine is trained.
According to the embodiment of the invention, since the weight vector of the support vector machine must be mapped from low dimension to high dimension, a kernel function is used as an indispensable condition, which makes the sample linearly separable, and based on this, the two kinds of classification vector machines are preset with, for example, a gaussian radial basis function. Since some sample points may fall between the hyperplane and the boundary, a relaxation variable needs to be introduced, and for each relaxation variable, a payment cost needs to be paid, a penalty parameter C is a penalty on the relaxation variable, the larger the C value is, the larger the misclassification penalty is, the smaller the C value is, the smaller the misclassification penalty is, so that the interval is as large as possible and the misclassification point is as small as possible, for example, C may be introduced into the binary vector machine to make a blending coefficient.
According to an embodiment of the present invention, the above operation S401 may be represented as training a classification vector machine using the feature vector, and specifically, in the case of an interference signal training sample, the operation may include: obtaining a feature vector sample; carrying out normalization processing on the feature vector samples to obtain target feature vector samples, wherein the maximum value and the minimum value of each type of feature vector in the target feature vector samples are respectively kept consistent; and inputting the target feature vector sample into a two-class vector machine, and outputting a class decision function containing target parameters, wherein the target parameters comprise a penalty parameter C and a kernel function parameter gamma.
According to an embodiment of the present invention, before the obtaining the feature vector samples, for example, the method may further include: and a two-class SVM (vector machine) is adopted, an interface is established between the characteristic parameter values of the interference signal and the target echo signal, and the interference signal and the target echo signal are respectively positioned at two sides of the interface. By using three parameters (such as the feature vector d) as the input of the SVM through a standard C-SVM, the kernel function selects a Gaussian radial basis kernel function, and four support vectors and corresponding classification decision functions and interfaces of the interference signals and the target signals can be obtained by solving the programming problem under the conditions of parameters C=i0 and gamma=0.01.
According to the embodiment of the invention, the target parameters can be determined by using a 10-fold cross validation method, for example, by randomly dividing the original training set into 10 parts, selecting one part as a validation set, using the rest 9 parts as training sets for model training, obtaining a model after training on the training set, testing on the validation set by using the model, and repeating the second step for 10 times (ensuring that each subset has a chance to be the validation set) by preserving the evaluation index of the model.
According to an embodiment of the present invention, in order to further optimize the target parameter, the feature vector may be divided into a training set and a test set by using, for example, a 10-fold cross test method, specifically, for example, the normalized feature vector sample data may be divided into 10 samples, one of the samples is used as the test set, and the remaining samples are used as the training set, and in order to further determine the parameter, for example, the obtained parameter value may be divided into grid patterns, and each grid may calculate the verification classification accuracy by using the 10-fold cross test method to obtain the optimal target parameter value. After the optimal target parameter value is obtained, 3/4 samples (namely 75 groups of target echo signals and 150 groups of interference signals) are randomly taken out from the target signal samples and the interference signal samples respectively to serve as training samples, and learning training of the SVM classifier is carried out, for example, the classification decision function containing the target parameter can be obtained.
According to the embodiment of the invention, after the classification decision function is obtained, for example, the remaining 1/4 sampling samples (namely, 25 groups of target echo signals and 50 groups of interference signals) can be used as test samples, and the test samples are input into a training SVM classifier to obtain the target detection rate and the interference detection rate of the experiment. By performing this random packet test process 200 times, for example, and finally averaging the target detection rate and the interference detection rate obtained each time, for example, the final target detection rate and interference detection rate can be obtained, and the average value of the classification accuracy can be used as the performance index of the machine algorithm.
According to the embodiment of the invention, by continuing to train the SVM model and performing simulation test, for example, a visual implementation of the classification decision function containing the target parameters can be finally obtained.
In operation S105, classification recognition capability is obtained.
According to an embodiment of the present invention, this operation may be expressed, for example, in actual implementation: and identifying the received signals by using the trained binary vector machine to determine whether the received signals belong to target echo signals or interference signals. Specifically, the implementation procedure may include, for example: identifying the received signals by using a classification decision function to obtain interface information for distinguishing a target echo signal or the interference signal; and determining a target echo signal or an interference signal according to the interface information.
Another embodiment of the present invention provides a signal recognition apparatus for a vehicle-mounted range radar, which can implement the method as described above.
By the method and the device disclosed by the invention, signal identification can be performed according to different characteristics of amplitude-frequency response of the target signal and the interference signal, so that the ranging capability of the vehicle-mounted range radar in a complex road environment is improved.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the invention thereto, but to limit the invention thereto, and any modifications, equivalents, improvements and equivalents thereof may be made without departing from the spirit and principles of the invention.
Claims (8)
1. A signal recognition method for a vehicle-mounted range radar, comprising:
acquiring a plurality of sampling signals, wherein each sampling signal is a target echo signal or an interference signal, and the interference signal comprises at least one of a noise amplitude modulation interference signal and a sine amplitude modulation interference signal;
processing each sampling signal to obtain a plurality of processed signals;
performing Fourier transformation on each processed signal to obtain amplitude-frequency response information corresponding to each sampling signal;
determining a feature vector corresponding to each sampling signal according to the amplitude-frequency response information;
training a classification vector machine by utilizing the feature vector;
identifying the received signals by using the trained two-classification vector machine to determine whether the received signals belong to the target echo signals or the interference signals;
wherein the acquiring a plurality of sampling signals comprises: transmitting a sinusoidal signal by a modulating signal generator; processing the sinusoidal signal by using an oscillator to obtain a local oscillation signal with a frequency variation of a triangular wave; generating a return signal by using a circulator; mixing the local oscillation signal and the return signal to generate a difference frequency signal and a sum frequency signal; processing the difference frequency signal by using a band-pass filter to obtain harmonic frequency; spacing according to the harmonic frequency; carrying out secondary mixing on harmonic signals in the difference frequency signals to obtain secondary mixed difference frequency signals; filtering the secondary mixed difference frequency signal to obtain the target echo signal;
wherein after obtaining the target echo signal, the method further comprises at least one of: the circulator is replaced by a sinusoidal signal transmitter so as to obtain a sinusoidal amplitude modulation interference signal with the same central frequency as the target echo signal; and obtaining a noise amplitude modulation interference signal with the center frequency identical to the target echo signal by replacing the circulator with a noise transmitter;
processing each sampling signal to obtain a plurality of processed signals, wherein the processing of each sampling signal comprises:
calculating the average value of the time domain amplitude of each sampling signal;
subtracting the average value of the time domain amplitude of each sampling signal from the time domain amplitude of each sampling signal to obtain the processed signal.
2. The method of claim 1, wherein the feature vector of each of the sampled signals comprises at least one of a mean, a standard deviation, a kurtosis, a difference between maximum minima, and an average power spectral density of amplitude-frequency response information of each of the sampled signals.
3. The method of claim 2, wherein the kurtosis is calculated by:
wherein K is kurtosis, s (f) i Representing the amplitude-frequency response information of each of the sampled signals, sigma representing the standard deviation of the amplitude-frequency response of each of the sampled signals,and n is the number of sampling points of each sampling signal, and i is any positive integer from 1 to n.
4. The method of claim 2, wherein the difference between the maximum and minimum values is calculated by:
pk＝max-min，
wherein pk is the difference between the maximum value and the minimum value, max is the maximum value of the amplitude-frequency response information of each sampling signal, and min is the minimum value of the amplitude-frequency response information of each sampling signal.
5. The method of claim 2, wherein the average power spectral density is calculated by:
wherein E is average power spectral density, f is frequency of each sampling signal, s (f) is amplitude-frequency response information of each sampling signal, and N is number of sampling points of each sampling signal.
6. The method of claim 1, wherein the bi-classification vector machine is pre-configured with a gaussian radial basis function, and training the bi-classification vector machine using the feature vector comprises:
obtaining a feature vector sample;
carrying out normalization processing on the feature vector samples to obtain target feature vector samples, wherein the maximum value and the minimum value of each type of feature vector in the target feature vector samples are respectively kept consistent;
and inputting the target feature vector sample into a two-class vector machine, and outputting a class decision function containing target parameters, wherein the target parameters comprise a penalty parameter C and a kernel function parameter gamma.
7. The method of claim 6, wherein identifying the received signal with the trained bi-classification vector machine to determine whether the received signal belongs to the target echo signal or the interfering signal comprises:
identifying the received signals by using the classification decision function to obtain interface information for distinguishing the target echo signals or the interference signals;
and determining the target echo signal or the interference signal according to the interface information.
8. A signal recognition device for an in-vehicle range radar, wherein the device is for implementing the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110397704.7A CN113156382B (en) | 2021-04-13 | 2021-04-13 | Signal identification method and device for vehicle-mounted range radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110397704.7A CN113156382B (en) | 2021-04-13 | 2021-04-13 | Signal identification method and device for vehicle-mounted range radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113156382A CN113156382A (en) | 2021-07-23 |
| CN113156382B true CN113156382B (en) | 2023-12-01 |
Family
ID=76890344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110397704.7A Active CN113156382B (en) | 2021-04-13 | 2021-04-13 | Signal identification method and device for vehicle-mounted range radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113156382B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107817532A (en) * | 2017-10-10 | 2018-03-20 | 成都菲斯洛克电子技术有限公司 | A kind of millimeter wave sensor and Smart Detector |
| CN108256488A (en) * | 2018-01-19 | 2018-07-06 | 中国人民解放军陆军装甲兵学院 | A kind of radar target identification method based on micro-Doppler feature extraction and deep learning |
| CN108267724A (en) * | 2018-01-19 | 2018-07-10 | 中国人民解放军陆军装甲兵学院 | A kind of unknown object recognition methods of radar target recognition |
| CN109164429A (en) * | 2018-10-17 | 2019-01-08 | 赣南师范大学 | Multi-fractal features Aircraft Targets classification method based on principal component analysis |
| CN110109080A (en) * | 2019-05-29 | 2019-08-09 | 南京信息工程大学 | Method for detecting weak signals based on IA-SVM model |
| CN110443138A (en) * | 2019-07-05 | 2019-11-12 | 东南大学 | Combine SVM and CNN Multi-Target Classification Method based on vehicle-mounted millimeter wave radar |
| CN111160176A (en) * | 2019-12-19 | 2020-05-15 | 南京理工大学 | Fusion feature-based ground radar target classification method for one-dimensional convolutional neural network |
| CN111175718A (en) * | 2019-12-18 | 2020-05-19 | 南京理工大学 | Time-frequency domain combined ground radar automatic target identification method and system |
-
2021
- 2021-04-13 CN CN202110397704.7A patent/CN113156382B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107817532A (en) * | 2017-10-10 | 2018-03-20 | 成都菲斯洛克电子技术有限公司 | A kind of millimeter wave sensor and Smart Detector |
| CN108256488A (en) * | 2018-01-19 | 2018-07-06 | 中国人民解放军陆军装甲兵学院 | A kind of radar target identification method based on micro-Doppler feature extraction and deep learning |
| CN108267724A (en) * | 2018-01-19 | 2018-07-10 | 中国人民解放军陆军装甲兵学院 | A kind of unknown object recognition methods of radar target recognition |
| CN109164429A (en) * | 2018-10-17 | 2019-01-08 | 赣南师范大学 | Multi-fractal features Aircraft Targets classification method based on principal component analysis |
| CN110109080A (en) * | 2019-05-29 | 2019-08-09 | 南京信息工程大学 | Method for detecting weak signals based on IA-SVM model |
| CN110443138A (en) * | 2019-07-05 | 2019-11-12 | 东南大学 | Combine SVM and CNN Multi-Target Classification Method based on vehicle-mounted millimeter wave radar |
| CN111175718A (en) * | 2019-12-18 | 2020-05-19 | 南京理工大学 | Time-frequency domain combined ground radar automatic target identification method and system |
| CN111160176A (en) * | 2019-12-19 | 2020-05-15 | 南京理工大学 | Fusion feature-based ground radar target classification method for one-dimensional convolutional neural network |
Non-Patent Citations (2)
| Title |
|---|
| 噪声调制类有源干扰信号识别技术研究;罗彬珅 等;雷达科学与技术;第17卷(第6期);597-602 * |
| 姚德臣.轨道交通故障诊断技术.中国铁道出版社,2017,23-24. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113156382A (en) | 2021-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110244271A (en) | Radar emitter sorting recognition methods and device based on multiple simultaneous compressed transform | |
| CN108919203B (en) | Radar active interference identification method and system | |
| CN112859014B (en) | Radar signal sorting-based radar interference suppression method, device and medium | |
| CN112965037B (en) | Radar signal waveform uncertainty test system | |
| CN114881093B (en) | Signal classification and identification method | |
| CN110647951A (en) | Wireless radio frequency equipment identity recognition method and system based on machine learning algorithm | |
| CN108490281B (en) | Method for predicting random noise electromagnetic radiation effect of frequency equipment and terminal equipment | |
| CN112307969A (en) | Pulse signal classification identification method and device and computer equipment | |
| CN112083393A (en) | Intermittent sampling forwarding interference identification method based on spectrogram average time characteristic | |
| Akyon et al. | Classification of intra-pulse modulation of radar signals by feature fusion based convolutional neural networks | |
| CN111401185B (en) | Electromagnetic signal and equipment association analysis method and system based on radio frequency fingerprint SEV | |
| CN113156382B (en) | Signal identification method and device for vehicle-mounted range radar | |
| CN112213697A (en) | Feature fusion method for radar deception jamming recognition based on Bayesian decision theory | |
| Xu et al. | Individual radio transmitter identification based on spurious modulation characteristics of signal envelop | |
| CN104596369B (en) | A kind of continuous wave Doppler Radio Fuze Anti sweep-frequency Békésy audiometer interference method based on SVM | |
| CN104125050A (en) | Ultrahigh-frequency RFID (radio frequency identification) reader protocol conformance testing method | |
| CN115166453B (en) | Partial discharge continuous monitoring method and device based on edge real-time radio frequency pulse classification | |
| Martone et al. | Characterization of RF devices using two-tone probe signals | |
| Dutt et al. | Discrete wavelet transform based unsupervised underdetermined blind source separation methodology for radar pulse deinterleaving using antenna scan pattern | |
| CN117420514A (en) | Electronic interference effect evaluation method for extracting radar parameter variation | |
| Sun et al. | Unintentional modulation microstructure enlargement | |
| Alammar et al. | Automatic extraction of signal areas from radio spectrograms based on the Hough transform | |
| Shahi et al. | Distinguishing electronic devices using harmonic radar based on a linear model | |
| CN115345216A (en) | FMCW radar interference elimination method fusing prior information | |
| KR20230026776A (en) | A method for determining a single sampling frequency of the classification target signal in order to predict a modulation scheme of the classification target signal, and a method and apparatus for predicting the modulation scheme using the classification target signal sampled with the single sampling frequency |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |