CN105866748B - A kind of long CFAR detection method of fixation window based on detection priori - Google Patents
A kind of long CFAR detection method of fixation window based on detection priori Download PDFInfo
- Publication number
- CN105866748B CN105866748B CN201610166232.3A CN201610166232A CN105866748B CN 105866748 B CN105866748 B CN 105866748B CN 201610166232 A CN201610166232 A CN 201610166232A CN 105866748 B CN105866748 B CN 105866748B
- Authority
- CN
- China
- Prior art keywords
- detection
- sliding window
- cfar
- target
- testing result
- 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.)
- Expired - Fee Related
Links
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
- G01S7/28—Details of pulse systems
- G01S7/285—Receivers
- G01S7/292—Extracting wanted echo-signals
- G01S7/2923—Extracting wanted echo-signals based on data belonging to a number of consecutive radar periods
- G01S7/2927—Extracting wanted echo-signals based on data belonging to a number of consecutive radar periods by deriving and controlling a threshold value
-
- 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
- G01S7/35—Details of non-pulse systems
- G01S7/352—Receivers
- G01S7/354—Extracting wanted echo-signals
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
A kind of fixation window long CFAR detection method based on detection priori of the disclosure of the invention, belongs to signal processing technology, and in particular to CFAR detection method.By the mean power for obtaining multiple aimless detection units, the titleization factor is multiplied by with the mean power obtain target detection threshold value again, the power of next detection unit is compared with the threshold value, judge that the detection unit whether there is target, if there is no targets, then the power of the detection unit is brought into the mean power for calculating threshold value, is cycled successively.The invention, which has, calculates simple, the high effect of accuracy.
Description
Technical field
The invention belongs to signal processing technologies, and in particular to CFAR detection method.
Background technology
Detection is a basic operation of signal processing.Constant false alarm algorithm is a kind of particularly important side in detection technique
Method in recent decades by in-depth study, is widely used in the fields such as radar, communication.
So-called constant false alarm (false alarm rate, CFAR), that is, the false-alarm kept constant in detection process are general
Rate, from the influence of background clutter.In actual detections of radar, in order to obtain stable detection performance and constant false-alarm probability,
Background clutter power level is usually estimated in obtained echo data from measuring in real time, so as to be adaptively adjusted detection threshold.
The key of CFAR processing is exactly that background clutter power is estimated.Corresponding processing method is divided into two major classes：One
Class is spatial domain CFAR, estimates clutter power using the reference unit spatially adjacent with unit to be detected；It is another kind of when being
Sequence CFAR by carrying out power estimation to same unit Multiple-Scan, has main steps that the iteration in time domain.Cell-average
(cell average, CA) CFAR, MAXIMUM SELECTION (greatest of, GO) CFAR, minimum selection (smallest of, SO)
CFAR and Ordered Statistic (ordered statistics, OS) CFAR is several spatial domain CFAR algorithms the most classical.They
Estimate to be to exclude the object element being detected during background power.
The content of the invention
The technical problem to be solved by the invention is to provide a kind of calculation amount is small, accuracy of detection is high based on detection priori
The long CFAR detection method of fixation window.
The present invention is to solve above-mentioned technical problem the technical scheme adopted is that a kind of fixation window based on detection priori is long
(former detection information-constant window, FDI-CW) CFAR detection method, this method bag
It includes：
Step 1：By the two paths of signals of in-phase component and quadrature component by matched filtering, acquired and worked as by square-law detector
The performance number of echo samples in preceding detection unit；
Step 2：Current power value is compared with detection threshold, performance number is not less than thresholding, is judged to detecting target；
Otherwise, it is determined that it is no target；Initial threshold acquires by the following method：Directly the R space cell of beginning is taken away as original reference
Unit acquires their average, is multiplied by after normalized factor T as detection threshold, the performance number of this R unit is directly stored in
With reference in sliding window, while 0 is all stored in testing result sliding window, the normalized factor T is determines according to actual conditions；
Step 3：If detecting target, by 0 deposit with reference to sliding window, testing result sliding window is stored in by 1；If mesh is not detected
Current power value deposit is then stored in testing result sliding window by mark with reference to sliding window by 0；It is described to refer to sliding window and testing result sliding window
Length is all mutually R, and data therein follow first in first out；
Step 4：It is sum to reference sliding window summation at this time, the summation of testing result sliding window is m, with sum divided by (R-m), is obtained
It is multiplied by normalized factor T to detection statistic Z, Z and obtains the thresholding of detection next time；
Step 5：The performance number of next detection unit is acquired according to step 1, then carries out step 2, at this time the door in step 2
Limit value acquires threshold value for step 4, then carries out following step successively；Final Xun Huan solves, and target is carried out to echo signal
Detection.
Further, normalized factor T is acquired using following two formula is combined in the step 2：
Pfa=(1+T/ (R-m))-(R-m)
P in formulafaFor false-alarm probability, PdFor detection probability, λ is signal to noise ratio.
The beneficial effects of the invention are as follows：
1st, under homogeneous background single goal environment, detection performance of the invention is better than GO-CFAR, SO-CFAR and OS-
CFAR, and it is suitable with CA-CFAR, but calculation amount of the present invention is small, and it is easy to operate.As shown in Figure 2, in homogeneous background single goal environment
Under, the detection probability curve of the present invention and CA-CFAR are completely superposed.And in such circumstances, performance of the invention is better than GO-
CFAR, SO-CFAR and OS-CFAR.
2nd, under homogeneous background target-rich environment, detection performance of the invention is better than CA-CFAR, GO-CFAR, SO-CFAR
With OS-CFAR algorithms.So-called multiple target refers to refer in sliding window in the presence of one or more jamming targets.In the present context, it is known that
The performance of SO-CFAR is better than GO-CFAR and CA-CFAR.From the figure 3, it may be seen that the detection performance of the present invention is better than SO-CFAR and OS-
CFAR.Therefore, under homogeneous background target-rich environment, detection performance of the invention is better than above-mentioned four kinds of algorithms.
Description of the drawings
Fig. 1 is a kind of long CFAR detection method schematic diagram of fixation window based on detection priori of the present invention；
Fig. 2 is the detection probability curve synoptic diagram of the present invention and CA-CFAR under homogeneous background single goal environment；Two in Fig. 2
Curve is completely superposed, and always, but calculation amount smaller of the present invention, speed is faster for testing result；
Fig. 3 is the detection performance comparison diagram of the present invention under homogeneous background target-rich environment；Wherein (a) is the present invention and SO-
CFAR is compared, and (b) is the present invention and OS-CFAR comparison diagrams.
Specific embodiment
Enter detector of the present invention comprising I/Q two ways of digital signals in the present invention.
When initial, I roads and Q roads signal acquire echo in current detection unit by matched filtering by square-law detector
The performance number of sampling.
Current power value is compared with detection threshold, performance number is not less than thresholding, is judged to detecting target, i.e. H1；
Otherwise, it is determined that it is no target, i.e. H0。
After judgement, if H1, by 0 deposit with reference to sliding window, the reference unit as next detection unit；If H0, ought
Preceding performance number deposit is with reference to sliding window.In this way, the target having been detected by is eliminated with reference to sliding window.Meanwhile by this testing result
Testing result sliding window is stored in, if H1, deposit 1；H0, deposit 0.
The detection threshold of this detection is obtained by following methods：
It is with reference to sliding window and testing result sliding window respectively, window length is all fixed as R there are two sliding window in the present invention.
It is respectively sum and m to summing and being worth with reference to sliding window and testing result sliding window before this judges to complete.Then (R-
M) non-targeted number in current reference sliding window is represented.With sum divided by (R-m), detection statistic Z is obtained, detection statistic is pair
The estimation of background clutter.Z is multiplied by the thresholding that normalized factor T can be obtained by detection next time.
Concrete structure of the present invention and workflow are as shown in Figure 1：
In Fig. 1, D represents the performance number of current detection unit, Xi(i=1 ..., R) represents reference unit, collectively forms ginseng
Examine sliding window.H (i) represents XiTesting result during as detection unit collectively constitutes testing result sliding window.M is represented with reference to sliding window
Middle target number, Z are detection statistic, and T is the normalized factor, and S represents detection threshold.
For detector of the present invention, false-alarm probability PfaWith detection probability PdRespectively：
Pfa=(1+T/ (R-m))-(R-m)
In formula, λ represents signal to noise ratio.
1st, under homogeneous background single goal environment, detection performance of the invention is better than GO-CFAR, SO-CFAR and OS-
CFAR, and it is suitable with CA-CFAR.As shown in Figure 2, under homogeneous background single goal environment, the detection of the present invention and CA-CFAR are general
Rate curve is completely superposed.And in such circumstances, it is known that the performance of CA-CFAR is better than GO-CFAR, SO-CFAR and OS-CFAR.
Therefore, under homogeneous background single goal environment, detection performance of the invention is also superior to above-mentioned three kinds of algorithms.
2nd, under homogeneous background target-rich environment, detection performance of the invention is better than CA-CFAR, GO-CFAR, SO-CFAR
With OS-CFAR algorithms.So-called multiple target refers to refer in sliding window in the presence of one or more jamming targets.In the present context, it is known that
The performance of SO-CFAR is better than GO-CFAR and CA-CFAR.From the figure 3, it may be seen that the detection performance of the present invention is better than SO-CFAR and OS-
CFAR.Therefore, under homogeneous background target-rich environment, detection performance of the invention is better than above-mentioned four kinds of algorithms.
Claims (2)
1. a kind of long CFAR detection method of fixation window based on detection priori, this method include：
Step 1：By the two paths of signals of in-phase component and quadrature component by matched filtering, current inspection is acquired by square-law detector
Survey the performance number of echo samples in unit；
Step 2：Current power value is compared with detection threshold, performance number is not less than thresholding, is judged to detecting target；It is no
Then, it is determined as no target；Initial threshold acquires by the following method：Directly the R space cell of beginning is taken away as original reference list
Member acquires their average, is multiplied by after normalized factor T as detection threshold, the performance number of this R unit is directly stored in ginseng
It examines in sliding window, while 0 is all stored in testing result sliding window, the normalized factor T is determines according to actual conditions；
Step 3：If detecting target, by 0 deposit with reference to sliding window, testing result sliding window is stored in by 1；If target is not detected,
Current power value deposit is then stored in testing result sliding window with reference to sliding window by 0；It is described to refer to sliding window and testing result sliding window length
It is all mutually R, and data therein follow first in first out；
Step 4：It is sum to reference sliding window summation at this time, the summation of testing result sliding window is m, with sum divided by (R-m), is examined
It surveys statistic Z, Z and is multiplied by the thresholding that normalized factor T obtains next time and detects；
Step 5：The performance number of next detection unit is acquired according to step 1, then carries out step 2, at this time the threshold value in step 2
Threshold value is acquired for step 4, then carries out following step successively；Final Xun Huan solves, and target detection is carried out to echo signal.
A kind of 2. long CFAR detection method of fixation window based on detection priori as described in claim 1, it is characterised in that institute
Normalized factor T in step 2 is stated to acquire using following two formula is combined：
Pfa=(1+T/ (R-m))-(R-m)
P in formulafaFor false-alarm probability, PdFor detection probability, λ is signal to noise ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610166232.3A CN105866748B (en) | 2016-03-22 | 2016-03-22 | A kind of long CFAR detection method of fixation window based on detection priori |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610166232.3A CN105866748B (en) | 2016-03-22 | 2016-03-22 | A kind of long CFAR detection method of fixation window based on detection priori |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105866748A CN105866748A (en) | 2016-08-17 |
| CN105866748B true CN105866748B (en) | 2018-05-18 |
Family
ID=56625493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610166232.3A Expired - Fee Related CN105866748B (en) | 2016-03-22 | 2016-03-22 | A kind of long CFAR detection method of fixation window based on detection priori |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105866748B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021007704A1 (en) * | 2019-07-12 | 2021-01-21 | Huawei Technologies Co., Ltd. | Method and apparatus for object detection system |
| CN110954879B (en) * | 2019-12-02 | 2022-09-20 | 北京无线电测量研究所 | Digital detection method and system for moving threshold |
| CN111856406B (en) * | 2020-06-02 | 2023-07-07 | 珠海微度芯创科技有限责任公司 | Target detection method and device based on FMCW radar echo |
| CN115032606B (en) * | 2022-08-11 | 2022-11-04 | 湖北工业大学 | Constant false alarm detector based on local minimum selected unit average |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3968490A (en) * | 1974-05-03 | 1976-07-06 | General Electric Company | Radar MTI processor with CFAR |
| GB2299910B (en) * | 1995-04-11 | 1999-01-06 | Commw Of Australia | Improved target detection |
| CN102012503A (en) * | 2010-07-29 | 2011-04-13 | 中国人民解放军海军航空工程学院 | Big-choosing and small-choosing nonparametric constant false alarm rate detector based on rank and local estimation |
| CN102353939A (en) * | 2011-06-28 | 2012-02-15 | 清华大学 | Improved constant false alarm method |
| CN104035090A (en) * | 2014-05-16 | 2014-09-10 | 奇瑞汽车股份有限公司 | Radar target detection-based free slide window type constant false alarm rate (CFAR) control method |
| CN105116380A (en) * | 2015-08-13 | 2015-12-02 | 电子科技大学 | Calculation method of sort type constant false alarm threshold |
-
2016
- 2016-03-22 CN CN201610166232.3A patent/CN105866748B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3968490A (en) * | 1974-05-03 | 1976-07-06 | General Electric Company | Radar MTI processor with CFAR |
| GB2299910B (en) * | 1995-04-11 | 1999-01-06 | Commw Of Australia | Improved target detection |
| CN102012503A (en) * | 2010-07-29 | 2011-04-13 | 中国人民解放军海军航空工程学院 | Big-choosing and small-choosing nonparametric constant false alarm rate detector based on rank and local estimation |
| CN102353939A (en) * | 2011-06-28 | 2012-02-15 | 清华大学 | Improved constant false alarm method |
| CN104035090A (en) * | 2014-05-16 | 2014-09-10 | 奇瑞汽车股份有限公司 | Radar target detection-based free slide window type constant false alarm rate (CFAR) control method |
| CN105116380A (en) * | 2015-08-13 | 2015-12-02 | 电子科技大学 | Calculation method of sort type constant false alarm threshold |
Non-Patent Citations (3)
| Title |
|---|
| CFAR Detection Method in Multi-target Environments for Foreign Object Debris Surveillance Radar;Jing Wu et.al;《THE PROCEEDINGS OF THE SECOND INTERNATIONAL CONFERENCE ON COMMUNICATIONS,SIGNAL PROCESSING,AND SYSTEMS》;20141230;第533-540页 * |
| CFAR方法在机场跑道FOD检测中的性能分析;李华琼等;《无线电工程》;20150914;第45卷(第9期);第53-57页 * |
| 非高斯背景下跑道异物监测雷达的杂波图CFAR技术;吴静等;《现代雷达》;20140130;第36卷(第1期);第35-42页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105866748A (en) | 2016-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104569948B (en) | Sub-band adaptive GLRT LTD detection methods under sea clutter background | |
| CN108802722B (en) | It is a kind of based on tracking before the Faint target detection virtually composed | |
| Davey et al. | Using phase to improve track-before-detect | |
| CN105866748B (en) | A kind of long CFAR detection method of fixation window based on detection priori | |
| CN103076602B (en) | For the radar self-adaption constant false alarm rate fusion detection method of multiple goal background | |
| KR101109150B1 (en) | Os-cfar detection method which stimulated computational complexity | |
| CN107843876A (en) | A kind of method for separating and equipment of radar pulse repetition | |
| CN111965632B (en) | Radar target detection method based on Riemann manifold dimensionality reduction | |
| CN103995259B (en) | Radar target adaptive-filtering fusion detection method under intensive interference environment | |
| CN105425223B (en) | The detection method of sparse extended distance radar target under broad sense Pareto clutter | |
| CN107024682A (en) | A kind of object detection method based on adaptive rejecting algorithm | |
| CN106872958A (en) | Radar target self-adapting detecting method based on linear fusion | |
| CN108872961B (en) | Radar weak target detection method based on low threshold | |
| CN106093903B (en) | Multiple target CFAR detection method based on unilateral detection unit cumulative mean | |
| CN107271973A (en) | CFAR detection method based on degree of skewness and average ratio under Weibull clutter environment | |
| CN109521412A (en) | Radar network composite airspace object detection method based on local statistic fusion | |
| CN106526545B (en) | A kind of detection method of robust CFAR detector | |
| CN106772357B (en) | AI-PHD filter multi-object tracking method under signal-to-noise ratio unknown condition | |
| CN110940977B (en) | Constant false alarm detector adaptive to sea condition change and constant false alarm detection method | |
| Vu et al. | Stability in SAR change detection results using bivariate Rayleigh distribution for statistical hypothesis test | |
| CN115032606B (en) | Constant false alarm detector based on local minimum selected unit average | |
| CN106019250A (en) | Repeating false target discriminating method based on angular glint | |
| CN105738882B (en) | To the Whitened degree evaluation method of actual measurement clutter covariance matrix estimation performance | |
| RU2646857C1 (en) | Method of moving target selection | |
| Kabakchiev et al. | Cell averaging constant false alarm rate detector with Hough transform in randomly arriving impulse interference |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180518 Termination date: 20210322 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |