CN103076600A - Radar target recognition method based on multi-azimuth pulse-E technology - Google Patents
Radar target recognition method based on multi-azimuth pulse-E technology Download PDFInfo
- Publication number
- CN103076600A CN103076600A CN2012105742935A CN201210574293A CN103076600A CN 103076600 A CN103076600 A CN 103076600A CN 2012105742935 A CN2012105742935 A CN 2012105742935A CN 201210574293 A CN201210574293 A CN 201210574293A CN 103076600 A CN103076600 A CN 103076600A
- Authority
- CN
- China
- Prior art keywords
- target
- pulse
- echo
- angle
- radar
- 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.)
- Granted
Links
Images
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a radar target recognition method based on a multi-azimuth pulse-E technology. A radar target library is provided with M targets, echo data of each target in multiple azimuth angles is acquired through actual measurement or simulation, a radar target echo database is established, and the echo data in the database is used as a training sample; the echo data of each target on any azimuth angle in a first-step radar target library is acquired through actual measurement or simulation to be used as a test sample; the multi-azimuth pulse-E training algorithm and training sample are used for establishing multiple pulses E for each target in the radar target library; and the test sample is used for testing a recognition effect of the multi-azimuth pulse-E technology. Due to the adoption of the method, the recognition probability and the noise resistance of the pulse-E technology for the radar target can be improved.
Description
Technical field
The invention belongs to the Technology of Radar Target Identification field, particularly a kind of radar target identification method based on multi-faceted E pulse work.
Background technology
Rise the nineties in last century, the pole characteristics of radar target just is widely used in the identification of short-pulse radar target, limit is the multiple natural resonance frequency of target, only by decisions such as the characteristic of target itself such as shape, size, materials, irrelevant with targeted attitude and polarization radar mode, therefore utilize pole characteristics to carry out target and identify features such as can overcoming scattering center with the shortcoming of target carriage change.At first the target identification method based on pole characteristics is to extract in advance the limit of known target and set up database, behind the time domain echo of receiving radar target to be identified, thereby behind echo the time Extraction of culmination and with database the limit identification target of comparing.Because signal to noise ratio (S/N ratio) is lower in the time of behind the actual radar echo signal that receives, be difficult to therefrom accurately Extraction of culmination during practical operation, so this method practicality is not strong.The beginning of the nineties in last century, there is the scholar to propose the E pulse work, this technology allows to extract in advance the limit of radar target in larger signal to noise ratio (S/N ratio) environment, then these limits are used to construct the E pulse, the E pulse good each target formation stores, behind the echo that receives target to be identified, the E pulse that utilizes each known target respectively with echo after the time convolution, if convolution results is 0, then target to be identified is identified as corresponding known target.Obviously, this method does not need to extract Target Pole from the echoed signal that receives in real time, has reduced the requirement to the echoed signal signal to noise ratio (S/N ratio).
Yet, it is pointed out that traditional E pulse work uses the limit of the reflection pickup target in the single orientation of target usually, then only is E pulse of each target formation, and think that the E pulse of structure and the echo convolution of respective objects arbitrary orientation all are 0, thereby reach the purpose of identification target.The theoretical foundation of this understanding is Target Pole and incident wave orientation-independent, corresponding residual is with the orientation relevant but it has ignored limit, residual represented corresponding limit to echo after the time contribution, may be very little in residual corresponding to some limit of some angles, representing corresponding limit can not be encouraged well, utilize limit extraction algorithm such as Matrix Pencil, can only extract the larger limit of residual, the limit that the different azimuth residual is larger is not necessarily identical, and this limit that has just caused different azimuth to be extracted is not necessarily identical.In this case, if the limit structure E pulse that only utilizes certain angle extraction to arrive, this E pulse is to the not effect of echo in other orientation, and discrimination is on the low side in the time of finally can causing traditional actual use of E pulse work.
Summary of the invention
The object of the present invention is to provide a kind of radar target identification method based on multi-faceted E pulse work, the method is to a plurality of E pulses of each target different angles areal structure, in the different angles echo of whole target all covered, can significantly improve the E pulse work to identification probability and the noise resisting ability of radar target, can be radar target identification method important reference is provided.
The technical scheme that realizes the object of the invention is:
The first step, establishing has M target in the radar target storehouse, utilize actual measurement or emulation to obtain a plurality of azimuthal echo datas of each target, sets up the radar target database, and the echo data in the storehouse is all as training sample;
Second step utilizes actual measurement or emulation to obtain the arbitrarily individual azimuthal echo data of each target in the first step radar target storehouse, as test sample book;
In the 3rd step, utilize the training sample in multi-faceted E pulse training algorithm and the first step, for each target in the radar target storehouse is trained a plurality of E pulses;
In the 4th step, utilize the test sample book in the second step to test the recognition effect of multi-faceted E pulse work.Even a plurality of E pulse convolution of the test sample book in the second step and each target, target corresponding to the minimum value in all convolution results is recognition result.
Echo data in the first step is based on the echo data that the short-pulse radar system gets access to, and the rule that a plurality of position angles angle is chosen is to choosing an angle every 5 ~ 10 degree in the azimuth range of target.
Azimuthal angular range of supposing a plurality of echo datas of each target in the first step is φ
i MinTo φ
i Max, angle intervals is △ φ
i, the 3rd step concrete steps of multi-faceted E pulse training algorithm are as follows:
The first step, initialization i=1, j=1, wherein i represents target designation, j represents E pulse numbering, makes n
i=0, n
iRepresent the E pulse number that i target constructed, for each target in the 2nd to M the target, in its azimuthal angle variation range, evenly choose 5 ~ 8 angles, from the radar target database, obtain the echo of corresponding angle, recycling E pulse work is according to E pulse of echo structure of each angle, thus a plurality of interim E pulses that have been the 2nd to M each target formation in the target;
Second step, the whole echo angular regions (φ of i target of mark
i Min, φ
i Max) be inactive area, show that the current E of not having pulse is effective to this area echo;
In the 3rd step, the center angle of the invalid angular regions of i target is designated as φ
j, utilize the echo of this angle by E pulse of E pulse work structure, and be designated as E
i j
The 4th step, calculate the convolution of the E pulse of certain echo of inactive area of i target and all targets, if the E pulse that E pulse of convolution results minimum is i target, show that then i target is correctly validated, the angle that this echo is corresponding is put into the effective coverage, shows that to have the E pulse of i target effective to the echo of this angle; When all echoes in the inactive area of i target being executed this operation in step, storage E
i j, n
i=n
i+ 1;
In the 5th step, if all angles of i target all have been put into the effective coverage, then record n
iBe the E pulse sum of i target, forwarded for the 6th step to; Otherwise j=j+1 forwarded for the 3rd step to;
The 6th step, i=i+1; If i＜M then removes all interim E pulses of i target, and makes j=1, forward second step to; If i〉M, then finish to carry out.
The present invention compared with prior art, its remarkable advantage is: (1) has improved correct recognition rata.To a plurality of E pulses of each target different angles areal structure, thereby make the E pulse of structure can cover the echo in each orientation of target, significantly improve identification probability; (2) opposing noise ability strengthens.After in echo, adding noise, utilize stronger than classic method based on the radar target identification method recognition capability of multi-faceted E pulse work.
Below in conjunction with accompanying drawing the present invention is described in further detail.
Description of drawings
Fig. 1 is the illustraton of model of F22, F35, three kinds of radar targets of VFY218, (a) F22 model (b) F35 model (c) VFY218 model.
Fig. 2 is the discrimination of multi-faceted E pulse work under different signal to noise ratio (S/N ratio)s that the present invention proposes.
Fig. 3 is the discrimination of traditional E pulse technology under different signal to noise ratio (S/N ratio)s, (a) utilizes the discrimination (b) of the E pulse of 10 degree echo structures to utilize the discrimination (c) of the E pulse of 90 degree echo structures to utilize the discrimination of the E pulse of 170 degree echo structures.
Embodiment
The present invention is the radar target identification method based on multi-faceted E pulse work, at first obtain a plurality of azimuthal echo datas of each target as training data by actual measurement or emulation, obtain the echo data of each target arbitrary orientation as test data by actual measurement or emulation again, be that each target is trained a plurality of E pulses in the radar target storehouse by multi-faceted E pulse training algorithm, at last test data identified.The method can improve correct recognition rata, and has stronger noise resisting ability.
Based on the radar target identification method of multi-faceted E pulse work, step is as follows among the present invention:
The first step, if M target arranged in the radar target storehouse, directly utilize short-pulse radar to obtain a plurality of azimuthal echoes of target, or set up the geometric model of target, utilize the simulation algorithm in the computational electromagnetics to obtain a plurality of azimuthal echoes of target, these echoes are used for consisting of the radar target database, and the echo data in the storehouse is all as training sample.Notice that the general rule that a plurality of position angles angle is chosen is to choosing an angle every 5 ~ 10 degree in the azimuth range of target.For example for this rotational symmetry target of aircraft, suppose that along continuous straight runs one week of machine of being diversion is 360 degree, corresponding 0 and 360 degree of head, corresponding 180 degree of tail, we only need choose 0 ~ 180 degree this half and calculate the echo of a plurality of angles so, and angle intervals generally is chosen for 10 degree;
Second step utilizes actual measurement or emulation to obtain the arbitrarily individual azimuthal echo data of each target in the first step radar target storehouse, and this moment, the position angle can be chosen arbitrarily, and the echo that gets access to is as test sample book;
In the 3rd step, utilize the training sample in multi-faceted E pulse training algorithm and the first step, for each target in the radar target storehouse is trained a plurality of E pulses.The angular range of supposing a plurality of azimuthal echo datas of each target in the first step is φ
i MinTo φ
i Max, angle intervals is △ φ
i, the concrete steps of multi-faceted E pulse training algorithm are as follows:
One, initialization i=1, j=1, wherein i represents target designation, j represents E pulse numbering, makes n
i=0, n
iRepresent the E pulse number that i target constructed, for each target in the 2nd to M the target, in its azimuthal angle variation range, evenly choose 5 ~ 8 angles, from the radar target database, obtain the echo of corresponding angle, recycling E pulse work is according to E pulse of echo structure of each angle, thus a plurality of interim E pulses that have been the 2nd to M each target formation in the target.For Aircraft Targets, can in 0 ~ 180 degree scope, choose the echoes of 0 degree, 45 degree, 90 degree, 135 degree, five angles of 180 degree, utilize E pulse of E pulse work structure for the echo of each angle, thereby obtain five E pulses;
Two, the whole echo angular regions (φ of i target of mark
i Min, φ
i Max) be inactive area, show that the current E of not having pulse is effective to this area echo;
Three, the center angle with the invalid angular regions of i target is designated as φ
j, utilize the echo of this angle by E pulse of E pulse work structure, and be designated as E
i j
Four, the convolution of the E pulse of certain echo of the inactive area of i target of calculating and all targets, if the E pulse that E pulse of convolution results minimum is i target, show that then i target is correctly validated, the angle that this echo is corresponding is put into the effective coverage, shows that to have the E pulse of i target effective to the echo of this angle.When all echoes in the inactive area of i target being executed this operation in step, storage E
i j, n
i=n
i+ 1;
If all angles of five an i target all have been put into the effective coverage, then recording ni is the E pulse sum of i target, forwards step 6 to.Otherwise j=j+1 forwards step 3 to;
Six, i=i+1.If i＜M then removes all interim E pulses of i target, and makes j=1, forward step 2 to; If i〉M, then finish to carry out.
The E pulse work of mentioning among the present invention in " echo to certain angle utilizes E pulse of E pulse work structure " is a kind of proven technique, all introduced its implementation step in document [1], [2], [3], at the second section of document [1], the 3rd joint of document [2], second section to the five joints of document [3] detailed introduction is arranged all, document is as follows: [1] Xiao Shunping, Guo Guirong, literary composition is encouraged in the village, " based on the target method of identification of waveform synthesis ", electronic countermeasure, the 2nd phase in 1993.[2]Rothwell, E.; Nyquist, D.; Kun-Mu Chen; Drachman, B.; , "Radar target discrimination using the extinction-pulse technique," Antennas and Propagation, IEEE Transactions on , vol.33, no.9, pp. 929- 937, Sep 1985[3]Rothwell, E.; Kun-Mu Chen; Nyquist, D.; Weimin Sun; , "Frequency domain E-pulse synthesis and target discrimination," Antennas and Propagation, IEEE Transactions on , vol.35, no.4, pp. 426- 434, Apr 1987 。
In the 4th step, utilize the test sample book in the second step to test the recognition effect of multi-faceted E pulse work.Each target that is operating as in the radar target storehouse in the 3rd step has been constructed a plurality of E pulses, the method of test recognition effect is to calculate respectively the convolution value of certain test sample book and a plurality of E pulses of each target, target corresponding to the minimum value in all convolution results is recognition result, all test sample books are all carried out this process, the number of times of statistical correction identification can calculate correct recognition rata.
In order to verify the validity of the inventive method, in conjunction with different target, the multi-faceted E pulse work of utilizing traditional E pulse work and the present invention to propose has been carried out emulation experiment relatively.At first use business software ANSYS respectively F22, F35, three kinds of radar targets of VFY218 to be carried out modeling, as shown in Figure 1.Then utilize method of moment to calculate single station scattered field of target from 3MHz to 600MHz, frequency sweep is spaced apart 3MHz, then the frequency sweep data are added Gaussian window and carry out inverse Fourier transform, obtain target in certain orientation with the modulation Gauss pulse as the excitation the time domain echo, calculate in this way the VV polarization time domain echo of every kind of target, the angle of pitch is 5 °, the position angle changes to 180 ° from 0 °, be spaced apart 5 °, 0 ° of corresponding nose cone direction, obviously, each target can obtain the echo of 37 different angles.In following experiment, we utilize the odd number echo as training data, and the even number echo is as test data.
Fig. 2 has shown that the present invention is to F22, F35, discrimination and the average recognition rate of three kinds of targets of VFY218 under different signal to noise ratio (S/N ratio) environment.Fig. 3 has shown that the traditional E pulse technology is to F22, F35, discrimination and the average recognition rate of three kinds of targets of VFY218 under different signal to noise ratio (S/N ratio) environment, Fig. 3 (a) is discrimination and the average recognition rate that three targets are all utilized 10 degree echo structure E arteries and veins tests, Fig. 3 (b) is discrimination and the average recognition rate that three targets are all utilized 90 degree echo structure E pulse tests, and Fig. 3 (c) is discrimination and the average recognition rate that three targets are all utilized 170 degree echo structure E arteries and veins tests.Can find by observing, when signal to noise ratio (S/N ratio) is 20db, average recognition rate of the present invention can reach 95%, and only utilizing the average recognition rate of the E pulse of 10 degree or 90 degree or 170 degree echo structures to be respectively 47%, 60%, 41%, this has proved that the present invention can significantly improve discrimination and noise resisting ability.
Claims (4)
1. radar target identification method based on multi-faceted E pulse work is characterized in that step is as follows:
The first step, establishing has M target in the radar target storehouse, utilize actual measurement or emulation to obtain a plurality of azimuthal echo datas of each target, sets up the radar target database, and the echo data in the storehouse is all as training sample;
Second step utilizes actual measurement or emulation to obtain the arbitrarily individual azimuthal echo data of each target in the first step radar target storehouse, as test sample book;
The 3rd step, utilize the training sample in multi-faceted E pulse training algorithm and the first step, be a plurality of E pulses of each target formation in the radar target storehouse;
In the 4th step, utilize the test sample book in the second step to test the recognition effect of multi-faceted E pulse work.
2. the radar target identification method based on multi-faceted E pulse work according to claim 1, it is characterized in that: the echo data in the first step is based on the echo data that the short-pulse radar system gets access to, and the rule that a plurality of position angles angle is chosen is to choosing an angle every 5 ~ 10 degree in the azimuth range of target.
3. the radar target identification method based on multi-faceted E pulse work according to claim 1, it is characterized in that: azimuthal angular range of supposing a plurality of echo datas of each target in the first step is φ
i MinTo φ
i Max, angle intervals is △ φ
i, the 3rd step concrete steps of multi-faceted E pulse training algorithm are as follows:
The first step, initialization i=1, j=1, wherein i represents target designation, j represents E pulse numbering, makes n
i=0, n
iRepresent the E pulse number that i target constructed, for each target in the 2nd to M the target, in its azimuthal angle variation range, evenly choose 5 ~ 8 angles, from the radar target database, obtain the echo of corresponding angle, recycling E pulse work is according to E pulse of echo structure of each angle, thus a plurality of interim E pulses that have been the 2nd to M each target formation in the target;
Second step, the whole echo angular regions (φ of i target of mark
i Min, φ
i Max) be inactive area, show that the current E of not having pulse is effective to this area echo;
In the 3rd step, the center angle of the invalid angular regions of i target is designated as φ
j, utilize the echo of this angle by E pulse of E pulse work structure, and be designated as E
i j
The 4th step, calculate the convolution of the E pulse of certain echo of inactive area of i target and all targets, if the E pulse that E pulse of convolution results minimum is i target, show that then i target is correctly validated, the angle that this echo is corresponding is put into the effective coverage, shows that to have the E pulse of i target effective to the echo of this angle; When all echoes in the inactive area of i target being executed this operation in step, storage E
i j, n
i=n
i+ 1;
In the 5th step, if all angles of i target all have been put into the effective coverage, then record n
iBe the E pulse sum of i target, forwarded for the 6th step to; Otherwise j=j+1 forwarded for the 3rd step to;
The 6th step, i=i+1; If i＜M then removes all interim E pulses of i target, and makes j=1, forward second step to; If i〉M, then finish to carry out.
4. the radar target identification method based on multi-faceted E pulse work according to claim 1, it is characterized in that: make test sample book in the second step and a plurality of E pulse convolution of each target, target corresponding to the minimum value in all convolution results is recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210574293.5A CN103076600B (en) | 2012-12-26 | 2012-12-26 | Radar target recognition method based on multi-azimuth pulse-E technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210574293.5A CN103076600B (en) | 2012-12-26 | 2012-12-26 | Radar target recognition method based on multi-azimuth pulse-E technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103076600A true CN103076600A (en) | 2013-05-01 |
| CN103076600B CN103076600B (en) | 2014-12-17 |
Family
ID=48153187
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210574293.5A Active CN103076600B (en) | 2012-12-26 | 2012-12-26 | Radar target recognition method based on multi-azimuth pulse-E technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103076600B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103235296A (en) * | 2013-05-05 | 2013-08-07 | 西安电子科技大学 | Power spectral feature correction-based radar target identification method under noise background |
| CN104280722A (en) * | 2014-10-17 | 2015-01-14 | 南京理工大学 | Radar target pole extraction method based on sparse representation theory |
| CN106443629A (en) * | 2016-09-30 | 2017-02-22 | 西安空间无线电技术研究所 | Radar object identification method base on Fourier primary function |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101561865A (en) * | 2009-05-22 | 2009-10-21 | 西安电子科技大学 | Synthetic aperture radar image target identification method based on multi-parameter spectrum feature |
-
2012
- 2012-12-26 CN CN201210574293.5A patent/CN103076600B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101561865A (en) * | 2009-05-22 | 2009-10-21 | 西安电子科技大学 | Synthetic aperture radar image target identification method based on multi-parameter spectrum feature |
Non-Patent Citations (4)
| Title |
|---|
| HUANHUAN ZHANG ET AL.: "Radar Target Recognition Based on Multi-Directional E-Pulse Technique", 《IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION》 * |
| PONNIAH ILAVARASAN ET AL.: "Performance of an automated radar target discrimination scheme using E pulses and S pulses", 《IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION》 * |
| 焦琳琳: "超宽带雷达目标的极点特征提取与E脉冲方法研究", 《中国优秀博硕士学位论文全文数据库 (硕士)信息科技辑》 * |
| 王晓丹 等: "雷达目标识别技术综述", 《现代雷达》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103235296A (en) * | 2013-05-05 | 2013-08-07 | 西安电子科技大学 | Power spectral feature correction-based radar target identification method under noise background |
| CN104280722A (en) * | 2014-10-17 | 2015-01-14 | 南京理工大学 | Radar target pole extraction method based on sparse representation theory |
| CN106443629A (en) * | 2016-09-30 | 2017-02-22 | 西安空间无线电技术研究所 | Radar object identification method base on Fourier primary function |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103076600B (en) | 2014-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104267379B (en) | A kind of active radar and passive radar based on Waveform Design works in coordination with anti-interference method | |
| CN104076355B (en) | Tracking before Dim targets detection in strong clutter environment based on dynamic programming | |
| CN103217679B (en) | Full-waveform laser radar echo data gaussian decomposition method based on genetic algorithm | |
| CN103869311B (en) | Real beam scanning radar super-resolution imaging method | |
| CN103744076B (en) | MIMO radar moving target detection method based on non-convex optimization | |
| CN103176168B (en) | A kind of airborne non-working side battle array radar short range clutter cancellation method | |
| CN107015205B (en) | False target elimination method for distributed MIMO radar detection | |
| CN103268496B (en) | SAR image target recognition method | |
| CN104714225B (en) | Dynamic programming tracking-before-detection method based on generalized likelihood ratios | |
| CN101776748B (en) | Method and device for extracting pole characteristic of complicated target | |
| CN102914769A (en) | Joint fractal-based method for detecting small target under sea clutter background | |
| CN109597046A (en) | Metre wave radar DOA estimation method based on one-dimensional convolutional neural networks | |
| CN104280725A (en) | Full-waveform laser radar data waveform decomposing method | |
| Lv et al. | Deep neural network-based interrupted sampling deceptive jamming countermeasure method | |
| CN106772299B (en) | One kind is based on apart from matched PD radar weak target Dynamic Programming detection method | |
| CN103076600B (en) | Radar target recognition method based on multi-azimuth pulse-E technology | |
| CN106226753A (en) | Wave arrival direction estimating method based on least variance method spectral function second dervative | |
| CN106371083A (en) | Radar monitoring point selection method and device, and echo signal processing method and device | |
| CN104881561A (en) | Hough transform-based track-before-detect method of multidimensional parameters | |
| CN106019257A (en) | Interpolation method based on high-frequency ground-wave radar ocean current observation result space-time characteristics | |
| CN106324576B (en) | Radar target adaptive matched filter method based on autoregression model | |
| CN104181518B (en) | A kind of method utilizing ripple to reach angular difference Sexual behavior mode training sample | |
| CN108845318A (en) | Spaceborne high score wide cut imaging method based on Relax algorithm | |
| CN108169744A (en) | A kind of ground wave radar and satellite ocean dynamic inversion use processing method | |
| CN103631990A (en) | Simulated scene model establishment method and system for SAR irradiation region |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |