CN117250594B - Radar target classification and identification method - Google Patents
Radar target classification and identification method Download PDFInfo
- Publication number
- CN117250594B CN117250594B CN202311534557.9A CN202311534557A CN117250594B CN 117250594 B CN117250594 B CN 117250594B CN 202311534557 A CN202311534557 A CN 202311534557A CN 117250594 B CN117250594 B CN 117250594B
- Authority
- CN
- China
- Prior art keywords
- pulse
- target
- radar
- doppler
- classification
- 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 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 10
- 238000012706 support-vector machine Methods 0.000 claims abstract description 8
- 238000001228 spectrum Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 5
- 238000002592 echocardiography Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function 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
- G01S7/41—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
- G01S7/411—Identification of targets based on measurements of radar reflectivity
-
- 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
- G01S7/417—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 involving the use of neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computer Networks & Wireless Communication (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a radar target classifying and identifying method, which belongs to the technical field of radar, and comprises the steps of extracting micro Doppler characteristics of targets under radar spread and alternate pulse group time sequence through phase compensation, classifying by utilizing an artificial neural network, a support vector machine, an artificial judging criterion and the like, solving the technical problems of radar multidimensional parameter fusion target identification of spread and alternate pulse group time sequence and quick classifying and identifying targets.
Description
Technical Field
The invention belongs to the technical field of radars, and relates to a radar target classification and identification method.
Background
In the existing radar narrowband target recognition technology, a recognition method based on micro Doppler features is more commonly used, a rotor wing of a rotor craft target such as an unmanned plane and a helicopter has obvious micro Doppler features, but the radar is required to irradiate the target with equal period pulses for a long time to extract the micro Doppler features of the target, and the requirement on radar time sequence is high. And after receiving the complete echo of the target, extracting the micro Doppler characteristic of the target through Fourier transform of a slow time dimension. However, in practical applications, the main task of radar equipment is to detect and track targets, so as to meet the detection requirements of different targets and avoid speed/distance ambiguity of the targets, a longer equal-period pulse time sequence is not designed specifically for target identification, but is designed in a staggered and alternating manner, which makes extracting micro-doppler features of the targets difficult. Therefore, in radar equipment mainly having an object detection function, how to extract micro doppler features of an object based on a staggered, alternating timing is a problem to be solved.
The micro Doppler characteristic of the radar target can reflect the characteristic of the target to a certain extent, but the characteristic is greatly influenced by the fluctuation of the environment and the target, clutter, shielding and noise can influence the accuracy of the micro Doppler characteristic, so that the single micro Doppler characteristic is taken as the basis of classification recognition, and the high accuracy is difficult to achieve, which is why the conventional micro Doppler characteristic recognition is difficult to be applied to actual radar equipment. In practical radar equipment, signal processing and data processing can complete extraction of other features of radar, such as RCS, speed, altitude, distance, etc., and how to fuse micro-doppler features with other features of a target is also a problem to be solved.
Disclosure of Invention
The invention aims to provide a radar target classification and identification method, which solves the technical problems of radar multidimensional parameter fusion target identification of staggered and alternate pulse group time sequences and rapid classification and identification of targets.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a radar target classification and identification method comprises the following steps:
step 1: constructing a pulse Doppler radar system, wherein the pulse Doppler radar system comprises a radar, a signal receiving module, a signal processing module and a learning classification module, and the signal receiving module sends echo signals to the signal processing module after receiving the echo signals of radar signals;
step 2: the signal processing module performs signal processing and data processing on the echo signals to obtain stable tracks, acquires the echo signals in the continuous time of the target according to a distance unit where the target is located, performs Fourier transformation on each section of equal-period pulse group signals in the echo signals, extracts a maximum value, and calculates the radial speed of the target according to the maximum value;
step 3: the signal processing module sequentially performs phase compensation on all pulse group signals in a beam width according to the radial speed of the target and the interval between pulse groups in the time sequence and a phase compensation algorithm to obtain signals after phase compensation;
step 4: the signal processing module splices all pulse groups in one beam width in a slow time dimension, performs non-uniform Fourier transform in the slow time dimension according to the period ratio among the pulse groups to obtain Doppler frequency spectrum, stores Doppler data after the module is obtained, and establishes a Doppler data set;
step 5: the learning classification module trains an artificial neural network classifier by using the Doppler data set and is used for classifying, identifying and judging whether the target has micro Doppler characteristics or not;
step 6: the learning classification module takes the micro Doppler characteristic as a main characteristic, combines the main characteristic with the target RCS, speed and height information obtained by resolving, and trains a plurality of two-classification support vector machine SVM classifiers;
step 7: and the learning classification module classifies the targets by cascading a plurality of SVM classifiers to obtain a final classification result.
Preferably, the pulse doppler radar system includes a pulse doppler radar, and the pulse doppler radar operates in a pulse train mode, i.e. a pulse group mode, wherein pulse periods in each pulse group are equal, and pulse periods of different pulse groups are different.
Preferably, when the step 2 is executed, the signal processing module performs signal processing and data processing on the echo signal to obtain a target RCS, a height and a speed;
forming a stable track according to the results of signal processing and data processing, intercepting echoes of a distance unit where a target is located and a plurality of units near the distance unit from IQ data of echo signals according to the azimuth and the distance in track information, and extracting data of all the pair of air pulse groups;
after target echo data are obtained, N-point FFT fast Fourier transform is respectively carried out on 6 pulse groups of each distance unit in a slow time dimension, the size of each distance unit is selected firstly after the FFT fast Fourier transform result is obtained in a modulo mode, then the size of each distance unit is selected in a frequency dimension, and the target radial speed is calculated according to the position where the maximum value is located.
Preferably, in executing step 3, the phase compensation is performed on the signal according to the target radial velocity and the fixed interval between pulse groups, the phase compensation algorithm specifically uses the phase of the first pulse group as a reference, and the 2 nd to 6 th pulse groups compensate the sea pulse period time according to the interval between the second pulse group and the first pulse group, and the specific formula is as follows:
；
wherein,original IQ signal representing all pulse groups on the distance cell where the target is located, < >>Representing the phase compensated signal, n=1, 2,3, … n, k=0, 1,2, …, fd represents the estimated target doppler frequency, ts is the period of the spaced pair of sea pulses.
Preferably, in executing step 4, performing non-uniform DFT discrete fourier transform of n×6 points on n×6 pulses after compensation according to pulse period ratio of 6 pulse groups, and modulo data after DFT discrete fourier transform to obtain a doppler spectrum of a target, which specifically includes the steps of:
step 4-1: calculating deviation of the repetition frequency by taking the repetition frequency 1/T1 of the first pulse as a reference frequency, wherein the specific formula is as follows:
；
where i denotes the pulse count, for a total of N x 6 pulses, i.e. i=1, 2,3 … N x 6,for the repetition frequency deviation of the ith pulse from the first pulse, < >>For the repetition frequency of the ith pulse, +.>A repetition frequency for the first pulse;
step 4-2: the sampling interval in the Fourier transform rotation factor is calculated, and the specific formula is as follows:
；
wherein t is n A time interval representing an nth twiddle factor;
step 4-3: the sampling interval is carried into a DFT formula for calculation, and the formula is as follows:
；
wherein,the Doppler resolution of the target is represented and is the reciprocal of the total observation time length of the target; m represents the number of points of DFT, taking an integer number greater than n×6 and 2.
Preferably, in the step 7, the method specifically includes cascading trained SVM classifiers, sequentially classifying features of the target, and finally obtaining a type of the target.
The radar target classification recognition method solves the technical problem of rapid classification recognition of targets by fusing radar multidimensional parameters with staggered and alternate pulse group time sequences, can extract micro Doppler features of the targets under the radar staggered and alternate pulse group time sequences through phase compensation, can meet the requirements of target detection and recognition at the same time, comprehensively utilizes micro Doppler features, signal level features, motion features and map information of the targets to recognize the targets, avoids the defect of single feature, comprehensively utilizes artificial neural network, support vector machine, artificial discrimination criteria and the like, realizes higher recognition accuracy by using rules of statistical knowledge to assist manual summary, and has the advantages of simple realization, low calculation amount, parallel processing and good real-time performance.
Drawings
FIG. 1 is a main flow chart of the present invention;
FIG. 2 is a flowchart of step 7 in the present embodiment;
FIG. 3 is a block diagram of a neural network of the present invention;
FIG. 4 is a schematic diagram of a confusion matrix of test data set recognition results according to the present invention;
FIG. 5 is a diagram of an original IQ of the present invention;
FIG. 6 is a schematic diagram of a 16-pulse FFT according to the invention;
fig. 7 is a schematic diagram of the phase compensation according to the present invention.
Detailed Description
The radar target classification and identification method as described in fig. 1-7 comprises the following steps:
step 1: constructing a pulse Doppler radar system, wherein the pulse Doppler radar system comprises a radar, a signal receiving module, a signal processing module and a learning classification module, and the signal receiving module sends echo signals to the signal processing module after receiving the echo signals of radar signals;
the pulse Doppler radar system comprises a pulse Doppler radar, wherein the pulse Doppler radar works in a pulse train mode, namely a pulse group mode, pulse periods in each pulse group are equal, and pulse periods of different pulse groups are different.
In this embodiment, the radar system is a pulse doppler radar, and works in a pulse train (i.e. pulse group) mode, where pulse periods in each pulse group are equal, pulse periods in different pulse groups are unequal (staggered time sequence, so as to meet the requirement of distance and speed ambiguity resolution), and pulses with a certain length can be inserted between pulse groups (alternate time sequence, so as to meet the detection requirements of different targets). This timing is one commonly used in current radar equipment.
The radar sequentially transmits 3 pulse groups with staggered periods to the air detection pulse groups, each pulse group comprises N pulses, the pulse periods in the 3 pulse groups are respectively T1, T2 and T3, each air detection pulse group transmits a sea pulse with a period of Ts, under the normal working rotating speed of the radar, the radar comprises 6 air detection pulse groups and 5 sea detection pulses in one azimuth beam width, and the total detection time of a single target is as follows:
。
step 2: the signal processing module performs signal processing and data processing on the echo signals to obtain stable tracks, acquires the echo signals in the continuous time of the target according to a distance unit where the target is located, performs Fourier transformation on each section of equal-period pulse group signals in the echo signals, extracts a maximum value, and calculates the radial speed of the target according to the maximum value;
the signal processing module performs signal processing and data processing on the echo signals to obtain target RCS, height and speed;
forming a stable track according to the results of signal processing and data processing, intercepting echoes of a distance unit where a target is located and a plurality of units near the distance unit from IQ data of echo signals according to the azimuth and the distance in track information, and extracting data of all the pair of air pulse groups;
IQ data are orthogonal complex data.
After target echo data are obtained, N-point FFT fast Fourier transform is respectively carried out on 6 pulse groups of each distance unit in a slow time dimension, the size of each distance unit is selected firstly after the FFT fast Fourier transform result is obtained in a modulo mode, then the size of each distance unit is selected in a frequency dimension, and the target radial speed is calculated according to the position where the maximum value is located.
In this embodiment, the calculation formula of the target radial velocity is as follows:
；
where v is the target radial velocity.
Step 3: the radial speed of the signal processing module and the interval between pulse groups in the time sequence sequentially carry out phase compensation on all pulse group signals in one beam width according to a phase compensation algorithm to obtain signals after phase compensation;
according to the radial speed of the target and the fixed interval between pulse groups, the phase compensation is carried out on the signal, the phase compensation algorithm is specifically based on the phase of the first pulse group, the 2 nd to 6 th pulse groups compensate the sea pulse period time according to the interval between the 2 nd to 6 th pulse groups and the first pulse group, and the specific formula is as follows:
；
wherein,original IQ signal representing all pulse groups on the distance cell where the target is located, < >>Representing the phase compensated signal, n=1, 2,3, … n, k=0, 1,2, …, fd represents the estimated target doppler frequency, ts is the period of the spaced pair of sea pulses.
Step 4: the signal processing module splices all pulse groups in one beam width in a slow time dimension, performs non-uniform Fourier transform in the slow time dimension according to the period ratio among the pulse groups to obtain Doppler frequency spectrum, stores Doppler data after the module is obtained, and establishes a Doppler data set;
the invention is corresponding to each section of equal period pulse group, a plurality of equal period pulse groups are spliced together, and all pulse groups in one beam width are spliced in a slow time dimension.
And carrying out non-uniform DFT discrete Fourier transform on N6 pulses after compensation according to the pulse period ratio of 6 pulse groups, and carrying out modulo on the data after the DFT discrete Fourier transform to obtain a target Doppler frequency spectrum, wherein the specific steps are as follows:
step 4-1: calculating deviation of the repetition frequency by taking the repetition frequency 1/T1 of the first pulse as a reference frequency, wherein the specific formula is as follows:
；
where i denotes the pulse count, for a total of N x 6 pulses, i.e. i=1, 2,3 … N x 6,for the repetition frequency deviation of the ith pulse from the first pulse, < >>For the repetition frequency (unit Hz) of the ith pulse,>a repetition frequency reference repetition frequency in Hz for the first pulse;
step 4-2: the sampling interval in the Fourier transform rotation factor is calculated, and the specific formula is as follows:
；
wherein t is n A time interval representing an nth twiddle factor;
step 4-3: the sampling interval is carried into a DFT formula for calculation, and the formula is as follows:
；
wherein,the Doppler resolution of the target is represented and is the reciprocal of the total observation time length of the target; m represents the number of points of DFT, taking an integer number greater than n×6 and 2.
Step 5: the learning classification module trains an artificial neural network classifier by using the Doppler data set and is used for classifying, identifying and judging whether the target has micro Doppler characteristics or not;
in this embodiment, a neural network classifier with M-dimensional input and 1-dimensional output is trained in advance by using doppler spectrum as a data set, and the hidden layer number of the network is 2.
Step 6: the learning classification module takes the micro Doppler characteristic as a main characteristic, combines the main characteristic with the target RCS, speed and height information obtained by resolving, and trains a plurality of two-classification support vector machine SVM classifiers;
step 7: the learning classification module classifies the targets by cascading a plurality of SVM classifiers to obtain a final classification result;
the method specifically comprises the steps of cascading trained SVM classifiers, sequentially classifying the characteristics of the target, and finally obtaining the type of the target.
In this embodiment, the trained network model is used to classify the new doppler spectrum, the micro doppler feature is marked as 1, the non-micro doppler feature is marked as 0, and fig. 5 is a confusion matrix obtained by classifying the neural network classifier on the test set data, and it can be seen that the accuracy rate on the test set exceeds 95%.
In the embodiment, four-dimensional information of target RCS, height and speed estimated by signal processing and data processing is used as characteristics, and a plurality of two-class Support Vector Machines (SVMs) are trained in advance by using micro Doppler classification results output by a neural network classifier.
In this embodiment, the trained SVM classifier is used for cascading, and the characteristics of the target are classified in sequence, so as to finally obtain the type (human, vehicle, unmanned aerial vehicle, ship) of the target, and the steps of the matrix are as follows:
step 7-1: the feature data enter a first SVM classifier, whether a target is an unmanned aerial vehicle is judged, if yes, the type of the output target is the unmanned aerial vehicle, and if not, the next SVM classifier is entered;
step 7-2: the characteristic data enter a second classifier, whether the target is a ship or not is judged, if yes, the type of the target is the ship, and if not, the next SVM classifier is entered;
step 7-3: the characteristic data enter a third classifier, whether the target is a car or not is judged, if yes, the type of the target is a ship, and if not, the next SVM classifier is entered;
step 7-4: and the last classifier judges whether the target is a person, if so, the output target type is a person, and if not, the output target type is unknown.
The radar target classification recognition method solves the technical problem of rapid classification recognition of targets by fusing radar multidimensional parameters with staggered and alternate pulse group time sequences, can extract micro Doppler features of the targets under the radar staggered and alternate pulse group time sequences through phase compensation, can meet the requirements of target detection and recognition at the same time, comprehensively utilizes micro Doppler features, signal level features, motion features and map information of the targets to recognize the targets, avoids the defect of single feature, comprehensively utilizes artificial neural network, support vector machine, artificial discrimination criteria and the like, realizes higher recognition accuracy by using rules of statistical knowledge to assist manual summary, and has the advantages of simple realization, low calculation amount, parallel processing and good real-time performance.
Claims (6)
1. A radar target classification and identification method is characterized in that: the method comprises the following steps:
step 1: constructing a pulse Doppler radar system, wherein the pulse Doppler radar system comprises a radar, a signal receiving module, a signal processing module and a learning classification module, and the signal receiving module sends echo signals to the signal processing module after receiving the echo signals of radar signals;
step 2: the signal processing module performs signal processing and data processing on the echo signals to obtain stable tracks, acquires the echo signals in the continuous time of the target according to a distance unit where the target is located, performs Fourier transformation on each section of equal-period pulse group signals in the echo signals, extracts a maximum value, and calculates the radial speed of the target according to the maximum value;
step 3: the signal processing module sequentially performs phase compensation on all pulse group signals in a beam width according to a phase compensation algorithm according to the target radial speed and the interval between pulse groups in the time sequence to obtain signals after phase compensation;
step 4: the signal processing module splices all pulse groups in one beam width in a slow time dimension, performs non-uniform Fourier transform in the slow time dimension according to the period ratio among the pulse groups to obtain Doppler frequency spectrum, stores the Doppler data after the modulo calculation and establishes a Doppler data set;
step 5: the learning classification module trains an artificial neural network classifier by using the Doppler data set and is used for classifying, identifying and judging whether the target has micro Doppler characteristics or not;
step 6: the learning classification module takes the micro Doppler characteristic as a main characteristic, combines the main characteristic with the target RCS, speed and height information obtained by resolving, and trains a plurality of two-classification support vector machine SVM classifiers;
step 7: and the learning classification module classifies the targets by cascading a plurality of SVM classifiers to obtain a final classification result.
2. A radar target classification and identification method as claimed in claim 1, wherein: the pulse Doppler radar system comprises a pulse Doppler radar, wherein the pulse Doppler radar works in a pulse train mode, namely a pulse group mode, pulse periods in each pulse group are equal, and pulse periods of different pulse groups are different.
3. A radar target classification and identification method as claimed in claim 1, wherein: when the step 2 is executed, the signal processing module performs signal processing and data processing on the echo signals to obtain target RCS, height and speed;
forming a stable track according to the results of signal processing and data processing, intercepting echoes of a distance unit where a target is located and a plurality of units near the distance unit from IQ data of echo signals according to the azimuth and the distance in track information, and extracting data of all the pair of air pulse groups;
after target echo data are obtained, N-point FFT fast Fourier transform is respectively carried out on 6 pulse groups of each distance unit in a slow time dimension, the size of each distance unit is selected firstly after the FFT fast Fourier transform result is obtained in a modulo mode, then the size of each distance unit is selected in a frequency dimension, and the target radial speed is calculated according to the position where the maximum value is located.
4. A radar target classification and identification method as claimed in claim 3, wherein: when executing step 3, according to the radial speed of the target and the fixed interval between pulse groups, the phase compensation is carried out on the signal, the phase compensation algorithm specifically uses the phase of the first pulse group as a reference, and the 2 nd to 6 th pulse groups compensate the sea pulse period time according to the interval between the pulse groups, and the specific formula is as follows:
；
wherein,original IQ signal representing all pulse groups on the distance cell where the target is located, < >>The phase compensated signal is represented by i for pulse count, N x 6 pulses total, i.e., i=1, 2,3 … N x 6, n=1, 2,3, … N, k=0, 1,2, …, fd for estimated target doppler frequency, ts for the interval of the marine pulse period.
5. A radar target classification and identification method as claimed in claim 1, wherein: when executing step 4, for N.6pulses after compensation, according to the pulse period ratio of 6 pulse groups, non-uniform DFT discrete Fourier transform of N.6points is performed, and data after DFT discrete Fourier transform is modulo to obtain Doppler frequency spectrum of a target, the specific steps are as follows:
step 4-1: calculating deviation of the repetition frequency by taking the repetition frequency 1/T1 of the first pulse as a reference frequency, wherein the specific formula is as follows:
；
where i denotes the pulse count, for a total of N x 6 pulses, i.e. i=1, 2,3 … N x 6,for the repetition frequency deviation of the ith pulse from the first pulse, < >>For the repetition frequency of the ith pulse, +.>A repetition frequency for the first pulse;
step 4-2: the sampling interval in the Fourier transform rotation factor is calculated, and the specific formula is as follows:
；
wherein t is n A time interval representing an nth twiddle factor;
step 4-3: the sampling interval is carried into a DFT formula for calculation, and the formula is as follows:
；
wherein,the Doppler resolution of the target is represented and is the reciprocal of the total observation time length of the target; m represents the number of points of DFT, taken an integer number greater than n×6 and 2, pi represents pi.
6. A radar target classification and identification method as claimed in claim 3, wherein: and (3) when the step (7) is executed, the method specifically comprises the steps of cascading the trained SVM classifiers, sequentially classifying the characteristics of the target, and finally obtaining the type of the target.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311534557.9A CN117250594B (en) | 2023-11-17 | 2023-11-17 | Radar target classification and identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311534557.9A CN117250594B (en) | 2023-11-17 | 2023-11-17 | Radar target classification and identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117250594A CN117250594A (en) | 2023-12-19 |
| CN117250594B true CN117250594B (en) | 2024-01-30 |
Family
ID=89135393
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311534557.9A Active CN117250594B (en) | 2023-11-17 | 2023-11-17 | Radar target classification and identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117250594B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020052441A1 (en) * | 2018-09-14 | 2020-03-19 | 深圳市道通智能航空技术有限公司 | Target classification method and related device |
| CN112580486A (en) * | 2020-12-15 | 2021-03-30 | 北京理工大学 | Human behavior classification method based on radar micro-Doppler signal separation |
| CN114594428A (en) * | 2022-03-02 | 2022-06-07 | 哈尔滨工业大学 | Intermittent sampling interference suppression method based on linear frequency modulation in pulse-to-pulse frequency coding |
| CN116660840A (en) * | 2023-06-06 | 2023-08-29 | 青岛九维华盾科技研究院有限公司 | True and false identification method based on fast-slow time domain joint characteristics under false target sample lack condition |
| CN116824629A (en) * | 2023-06-02 | 2023-09-29 | 大连理工大学 | High-robustness gesture recognition method based on millimeter wave radar |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3646051A4 (en) * | 2017-06-29 | 2021-02-24 | Sensing Management Pty Limited | A system and method of detecting objects |
-
2023
- 2023-11-17 CN CN202311534557.9A patent/CN117250594B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020052441A1 (en) * | 2018-09-14 | 2020-03-19 | 深圳市道通智能航空技术有限公司 | Target classification method and related device |
| CN112580486A (en) * | 2020-12-15 | 2021-03-30 | 北京理工大学 | Human behavior classification method based on radar micro-Doppler signal separation |
| CN114594428A (en) * | 2022-03-02 | 2022-06-07 | 哈尔滨工业大学 | Intermittent sampling interference suppression method based on linear frequency modulation in pulse-to-pulse frequency coding |
| CN116824629A (en) * | 2023-06-02 | 2023-09-29 | 大连理工大学 | High-robustness gesture recognition method based on millimeter wave radar |
| CN116660840A (en) * | 2023-06-06 | 2023-08-29 | 青岛九维华盾科技研究院有限公司 | True and false identification method based on fast-slow time domain joint characteristics under false target sample lack condition |
Non-Patent Citations (4)
| Title |
|---|
| 基于二维映射的LFMUWB雷达动目标速度估计;杨利民;苏卫民;顾红;薄超;;雷达学报(第03期);全文 * |
| 基于双雷达微动特征融合的无人机分类识别;章鹏飞;李刚;霍超颖;殷红成;;雷达学报(第05期);全文 * |
| 杨利民 ; 苏卫民 ; 顾红 ; 薄超 ; .基于二维映射的LFMUWB雷达动目标速度估计.雷达学报.2012,(第03期),全文. * |
| 章鹏飞 ; 李刚 ; 霍超颖 ; 殷红成 ; .基于双雷达微动特征融合的无人机分类识别.雷达学报.2018,(第05期),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117250594A (en) | 2023-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Deep learning-based UAV detection in pulse-Doppler radar | |
| De Wit et al. | Radar micro-Doppler feature extraction using the singular value decomposition | |
| CN107728142B (en) | Radar high-resolution range profile target identification method based on two-dimensional convolutional network | |
| CN110018453B (en) | Intelligent model identification method based on aircraft track characteristics | |
| CN105093215A (en) | Doppler information based method for tracking low-altitude low-speed small target through radar | |
| CN104077601A (en) | Method for carrying out synthetic target recognition through information of different types | |
| CN103176187B (en) | A kind of airborne early warning radar ground surface high speed highway goal filtering method | |
| CN109117776B (en) | Aircraft and meteorological clutter classification and identification method based on flight path information | |
| CN109409225A (en) | Unmanned plane classification method and device based on the fusion of radar multipath signal time-frequency characteristics | |
| CN115061113B (en) | Target detection model training method and device for radar and storage medium | |
| CN107390193A (en) | Frequency modulated continuous wave radar Aircraft Targets sorting technique based on the fusion of more range cells | |
| Gao et al. | Hybrid SVM-CNN classification technique for moving targets in automotive FMCW radar system | |
| Yang et al. | Echoformer: Transformer architecture based on radar echo characteristics for UAV detection | |
| CN115657005A (en) | Unmanned aerial vehicle identification method based on Doppler radar multi-information fusion | |
| Karlsson et al. | Model-aided drone classification using convolutional neural networks | |
| Ren et al. | Estimating physical parameters from multi-rotor drone spectrograms | |
| Zhu et al. | Radar HRRP group-target recognition based on combined methods in the backgroud of sea clutter | |
| CN117250594B (en) | Radar target classification and identification method | |
| CN111796247B (en) | Echo simulation method for group unmanned aerial vehicle based on micro Doppler effect | |
| Kaya et al. | Neural network based drone recognition techniques with non-coherent s-band radar | |
| CN107192993A (en) | Frequency modulated continuous wave radar Aircraft Targets sorting technique based on image entropy feature | |
| Xiao et al. | Multi-target ISAR imaging based on image segmentation and short-time Fourier transform | |
| Brooks et al. | Drone recognition by micro-doppler and kinematic | |
| Sayed et al. | UAV Classification Using Neural Networks and CAD-Generated Radar Datasets | |
| Jiang et al. | A High-resolution radar automatic target recognition method for small UAVs based on multi-feature fusion |
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 |