CN105467374B - Object detection method based on prewhitening ratio cell average detector under sea clutter background - Google Patents
Object detection method based on prewhitening ratio cell average detector under sea clutter background Download PDFInfo
- Publication number
- CN105467374B CN105467374B CN201610012175.3A CN201610012175A CN105467374B CN 105467374 B CN105467374 B CN 105467374B CN 201610012175 A CN201610012175 A CN 201610012175A CN 105467374 B CN105467374 B CN 105467374B
- Authority
- CN
- China
- Prior art keywords
- detection
- observation vector
- target
- unit
- reference unit
- 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
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
- G01S7/414—Discriminating targets with respect to background clutter
Abstract
The invention discloses the object detection method based on prewhitening ratio cell average detector under a kind of sea clutter background.Its technical scheme is：Build the observation vector z of the detection unit and observation vector z of reference unit in radar return datak(k=1,2 ..., P), and maximum likelihood estimate is utilized, obtain the estimate of the normalization sample covariance matrix of this P+1 observation vectorUtilizeTo z and zkAlbefaction is carried out, obtains the observation vector of detection unit after albefactionWith the observation vector of reference unitCalculateMean power e andMean power ek；Utilize ekSeek the mean power of P reference unitUsing e andCalculate detection statistic ξ；Judge that target whether there is using detection statistic ξ and the detection threshold T set.The present invention improves the detection performance of target, available for the target detection under sea clutter background.
Description
Technical field
The invention belongs to Radar Targets'Detection technical field, and in particular to prewhitening ratio is based under a kind of sea clutter background
The object detection method of cell average detector, available for the target detection under sea clutter background.
Background technology
In the case that high-resolution radar is operated in small grazing angle, sea clutter shows stronger non-Gaussian feature, can
To simulate sea clutter using complex Gaussian model.According to sea clutter complex Gaussian model, sea clutter is described as speckle component
With the product of texture component.Speckle component meets zero-mean complex Gaussian process, related to the Bragg scatterings become soon；Texture component
Related to the texture information that radar illumination sea becomes slowly for non-negative random process, coherence time is about 100ms.Early in previous generation
Record the forties, CFAR has been just the important research content in radar signal processing field.At present, existing phase in the world
To a whole set of CFAR etection theories and method of maturation." Hu Wenlin airborne radar constant false alarm rates detection algorithm studies [D] to document
Xi'an：Xian Electronics Science and Technology University, 2007. " the middle ratio cell average detectors introduced, is averaged using the sampled value of reference unit
Method estimate the power level of background clutter.The use of this method has two important hypotheses：Detection unit clutter
Statistical property it is identical with the statistical property of reference unit；It is separate between detection unit and reference unit.But in reality
In the environment of border, because the echo-signal that radar receives has stronger correlation, second hypotheses can not meet, cause
The performance degradation of the equal value detection method of ratio.
The content of the invention
It is an object of the invention to the deficiency for above-mentioned prior art, proposes to be based on prewhitening under a kind of sea clutter background
The object detection method of ratio cell average detector, to improve the detection performance under sea clutter background.
To realize above-mentioned technical purpose, technical scheme includes as follows：
(1) number of echoes formed by surface scattering is received using radar transmitter transmitting pulse signal, radar receiver
According to the observation vector z of the detection unit and observation vector z of reference unit in structure echo datakFor：
K=1,2 ..., P
Wherein, H0Represent to only have clutter in the absence of target it is assumed that H1It is indicating clutter and target be present it is assumed that c is represented
The sea clutter vector of detection unit, ckThe sea clutter vector of reference unit is represented, s represents target echo signal, and P is represented with reference to single
First number；
(2) utilize maximum Likelihood, obtain the observation vector z and reference unit of detection unit in echo data
Observation vector zkNormalization sample covariance matrix estimate
(3) estimate for normalizing sample covariance matrix is utilizedRespectively to the sight of detection unit in echo data
Direction finding amount z and reference unit observation vector zkAlbefaction is carried out, obtains the observation vector of the detection unit after albefactionAfter albefaction
Reference unit observation vector
(4) assume to have clutter and target be present, calculate the observation vector of detection unit after albefactionMean power e and white
The observation vector of reference unit after changeMean power ek, according to mean power e and ek, calculate test statistics ξ：
Wherein,Represent the mean power of the observation vector of P reference unit；
(5) false-alarm probability P is setfa, calculate corresponding detection threshold T；
(6) by detection statistic ξ compared with detection threshold T, judge that target whether there is in detection unit：If ξ
>=T, illustrate that detection unit has target, if ξ ＜ T, illustrate no target.
The present invention has advantages below compared with prior art：
1) because the present invention is directed to the correlation of sea clutter, albefaction is carried out to sea clutter using whitening approach before detection,
Improve the detection performance of subsequent detectors.
2), can be adaptive because the present invention is using clutter data real-time update sea clutter normalization sample covariance matrix
With detection environment in noise performance match, more preferable detection performance can be obtained.
Brief description of the drawings
Fig. 1 is the implementation process figure of the present invention；
Fig. 2 is the target detection probability of the emulation data obtained with of the invention and existing ratio cell average detector in different letters
The miscellaneous comparison diagram than under.
Embodiment
The invention will be further described below in conjunction with the accompanying drawings：
Reference picture 1, step is as follows for of the invention realizing：
Step 1, the observation vector z of the detection unit and observation vector z of reference unit in echo data is builtk。
Radar transmitter launches pulse signal, and radar receiver receives the echo data formed by surface scattering, this time
Wave number is according to being to include the three-dimensional data that distance is tieed up, the peacekeeping pulse of ripple position is tieed up, and each apart from peacekeeping ripple position, dimension forms one and differentiates list
Member；
A resolution cell is arbitrarily chosen in echo data as detection unit, and P are chosen around detection unit
Resolution cell is as reference unit；
Build the observation vector z of the detection unit and observation vector z of reference unit in echo datak:
When detection unit only has clutter, and target is not present, z=c, zk=ck,
When detection unit has clutter and target be present, z=s+c, zk=ck, i.e.,：
K=1,2 ..., P<1>
Wherein, H0Represent to only have clutter in the absence of target it is assumed that H1It is indicating clutter and target be present it is assumed that c is represented
The sea clutter vector of detection unit, ckThe sea clutter vector of reference unit is represented, s represents target echo signal, and P is represented with reference to single
First number.
Step 2, using maximum Likelihood, the observation vector z of detection unit in echo data is obtained and with reference to single
The observation vector z of memberkNormalization sample covariance matrix estimate
(2.1) assume to have clutter and target be present, by the observation vector z of detection unit in echo data and reference unit
Observation vector zkIt is normalized respectively, obtains the observation vector of normalized detection unitWith the ginseng of power normalization
Examine the observation vector of unit
K=1,2 ..., P<2>
Wherein, P ＞ 2N, N are accumulation umber of pulse；
(2.2) according to above-mentionedCalculate the covariance matrix M of P power normalization reference unit：
K=1,2 ..., P<3>
Wherein, ()HTo take conjugate transposition, M dimension is N × N；
(2.3) according to above-mentioned covariance matrix M, calculateWithJoint probability density function
<4>
(2.4) by joint probability density functionNatural logrithm is taken, obtains log-likelihood function
<5>
(2.5) by log-likelihood functionTo M derivations, and it is zero to make derivative, obtains estimating for sample covariance matrix
Evaluation
<6>
(2.6) by (2.1)WithSubstitute into the estimate of above-mentioned sample covariance matrixObtain normalizing sample
The estimate of this covariance matrix
<7>
Step 3, the observation vector of the detection unit after albefaction is calculatedWith the observation vector of the reference unit after albefaction
Due to having phase in echo data between detection unit and reference unit and between reference unit and reference unit
Closing property to detection unit and reference unit, it is necessary to carry out albefaction before detection using whitening approach, i.e., using normalizing sample association
The estimate of variance matrixSight to reference unit in the observation vector z and echo data of detection unit in echo data
Direction finding amount zkAlbefaction is carried out respectively；
Albefaction is carried out to the observation vector z of detection unit, obtains the observation vector after albefaction
<8>
To the observation vector z of reference unitkAlbefaction is carried out, obtains the observation vector after albefaction
<9>
Step 4, test statistics ξ is calculated.
(4.1) assume to have clutter and target be present, calculate the observation vector of the detection unit after albefactionMean power e：
<10>
(4.2) assume to have clutter and target be present, calculate the observation vector of the reference unit after albefactionMean power
ek：
<11>
(4.3) mean power of the observation vector of P reference unit is calculated
<12>
(4.4) formula is utilized<10>With<12>Calculate test statistics ξ：
<13>
Step 5, false-alarm probability P is setfa, calculate corresponding detection threshold T.
Detection threshold T is calculated by Monte Carlo experiment, and its step is as follows：
(5.1) natural number for being more than 1 for making C be setting, the detection statistic ξ of the 1st target is calculated1To the C target
Detection statistic ξC；
(5.2) C obtained detection statistic is arranged in descending order, the C target detection statistic after descending arrangement
In, take [[CPfa]] individual element value is as detection threshold T, [[CPfa]] represent to be no more than real number CPfaMaximum integer.
Step 6, by detection statistic ξ compared with detection threshold T, judge that target whether there is in detection unit.Such as
Fruit ξ >=T, then judge that detection unit has target, if ξ ＜ T, judge that detection unit does not have target.
The effect of the present invention is described further with reference to emulation experiment.
1. simulation parameter
The emulation data used in experiment is the three-dimensional datas including distance dimension, ripple position peacekeeping pulse dimension, each distance dimension
Tieed up with ripple position and form a resolution cell, wherein apart from dimension 1600, ripple position dimension 600, pulse dimension 5；Signal to noise ratio SCR=[-
10dB, 30dB], false-alarm probability Pfa=10-4。
2. emulation experiment content
The inventive method is respectively adopted in emulation experiment, existing ratio cell average detector is detected to emulation data, led to
The Detection results of two kinds of detection methods of detection probability com-parison and analysis are crossed, detection probability shows that more greatly detector detection performance is better.
Emulation experiment step：
First, the transformation range for setting signal to noise ratio is SCR=[- 10dB, 30dB]；
Then, 10000 different resolution cells are randomly selected in data are emulated, and are added on each resolution cell
The echo signal of corresponding signal to noise ratio；
Finally, under the conditions of different signals to noise ratio, using of the invention and existing ratio cell average detector to above-mentioned 10000 points
Distinguish that unit carries out target detection, the detection probability for obtaining of the invention and existing ratio cell average detector is bent with the change of signal to noise ratio
Line, as a result as shown in Fig. 2 transverse axis represents signal to noise ratio in Fig. 2, the longitudinal axis represents detection probability.
Figure it is seen that the target based on prewhitening ratio cell average detector under sea clutter background proposed by the present invention
The detection performance of detection method is substantially better than the detection performance of existing ratio cell average detector.
Claims (2)
1. the object detection method based on prewhitening ratio cell average detector under sea clutter background, it is characterised in that including：
(1) echo data formed by surface scattering, structure are received using radar transmitter transmitting pulse signal, radar receiver
Build the observation vector z of detection unit in echo data and the observation vector z of reference unitkFor：
Wherein, H0Represent to only have clutter in the absence of target it is assumed that H1It is indicating clutter and target be present it is assumed that c represents detection
The sea clutter vector of unit, ckThe sea clutter vector of reference unit is represented, s represents target echo signal, and P represents number of reference
Mesh；
(2) maximum Likelihood is utilized, obtains observation vector z and the observation of reference unit of detection unit in echo data
Vectorial zkNormalization sample covariance matrix estimate
(2.1) assume there is clutter and target be present, by the observation vector z of detection unit in echo data and the observation of reference unit
Vectorial zkPower normalization processing is carried out respectively, obtains the observation vector of the detection unit of power normalizationAnd power normalization
Reference unit observation vector
Wherein, P ＞ 2N, N are accumulation umber of pulse；
(2.2) according to above-mentionedCalculate the covariance matrix M of P power normalization reference unit：
Wherein, ()HTo take conjugate transposition, M dimension is N × N；
(2.3) according to above-mentioned covariance matrix M, calculateWithJoint probability density function
(2.4) by joint probability density functionNatural logrithm is taken, obtains log-likelihood function
(2.5) by log-likelihood functionTo M derivations, and it is zero to make derivative, obtains the estimate of sample covariance matrix
(2.6) by (2.1)WithSubstitute into the estimate of above-mentioned sample covariance matrixObtain normalizing sample association
The estimate of variance matrix
(3) estimate for normalizing sample covariance matrix is utilizedRespectively the observation to detection unit in echo data to
Measure the observation vector z of z and reference unitkAlbefaction is carried out, obtains the observation vector of the detection unit after albefactionWith the ginseng after albefaction
Examine the observation vector of unit
(4) assume to have clutter and target be present, calculate the observation vector of detection unit after albefactionMean power e and albefaction after
The observation vector of reference unitMean power ek, according to the two mean powers e and ek, calculate detection statistic ξ：
Wherein,The mean power of the observation vector of P reference unit is represented,N
To accumulate umber of pulse；
(5) false-alarm probability P is setfa, calculate corresponding detection threshold T；
(6) by detection statistic ξ compared with detection threshold T, judge that target whether there is in detection unit：If ξ >=T,
Then judge that detection unit has target, if ξ ＜ T, judge that detection unit does not have target.
2. the object detection method based on prewhitening ratio cell average detector under sea clutter background as claimed in claim 1, its
It is characterised by, false-alarm probability P is set in the step 5fa, corresponding detection threshold T is calculated, is carried out as follows：
(5.1) C is set as the natural number more than 1, calculates the detection statistic ξ of the 1st target1To the detection system of the C target
Measure ξC；
(5.2) C obtained detection statistic is arranged in descending order, in the C target detection statistic after descending arrangement, taken
[CPfa] individual element value is as detection threshold T, [CPfa] represent to be no more than real number CPfaMaximum integer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610012175.3A CN105467374B (en) | 2016-01-08 | 2016-01-08 | Object detection method based on prewhitening ratio cell average detector under sea clutter background |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610012175.3A CN105467374B (en) | 2016-01-08 | 2016-01-08 | Object detection method based on prewhitening ratio cell average detector under sea clutter background |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105467374A CN105467374A (en) | 2016-04-06 |
| CN105467374B true CN105467374B (en) | 2018-03-20 |
Family
ID=55605278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610012175.3A Active CN105467374B (en) | 2016-01-08 | 2016-01-08 | Object detection method based on prewhitening ratio cell average detector under sea clutter background |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105467374B (en) |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6150474B2 (en) * | 2012-08-09 | 2017-06-21 | 三菱電機株式会社 | Signal processing apparatus and signal processing method in wind profiler |
| CN104614714B (en) * | 2015-03-05 | 2017-05-10 | 北京航空航天大学 | Double calibration treatment method based on minimum weighted mean square error |
| CN104730507B (en) * | 2015-04-01 | 2017-06-16 | 苏州闻捷传感技术有限公司 | Vehicle-mounted roadblock alarm method based on the imaging of premodulated AIC distance by radars |
| CN104793200B (en) * | 2015-04-09 | 2017-05-10 | 电子科技大学 | Dynamic planning track-before-detect method based on iterative processing |
-
2016
- 2016-01-08 CN CN201610012175.3A patent/CN105467374B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105467374A (en) | 2016-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kim et al. | Human activity classification based on micro-Doppler signatures using a support vector machine | |
| CN104569948B (en) | Sub-band adaptive GLRT LTD detection methods under sea clutter background | |
| Hu et al. | Detection of low observable targets within sea clutter by structure function based multifractal analysis | |
| Luo et al. | The fractal properties of sea clutter and their applications in maritime target detection | |
| US9057783B2 (en) | Change detection method and system for use in detecting moving targets behind walls, barriers or otherwise visually obscured | |
| US9188666B2 (en) | System and method for distribution free target detection in a dependent non-Gaussian background | |
| CN105425223B (en) | The detection method of sparse extended distance radar target under broad sense Pareto clutter | |
| CN103558597B (en) | Based on weak target detection method in the sea clutter of spectrum kurtosis | |
| CN106569193B (en) | The small targets detection in sea clutter method filtered based on anterior-posterior to income reference particle | |
| CN111580091B (en) | Sea surface weak target detection method based on AR spectrum singular intensity function | |
| CN109212503B (en) | Offshore target detection method and device based on AR spectrum fractal correlation characteristics | |
| Park et al. | Simulation and analysis of polarimetric radar signatures of human gaits | |
| CN106772302A (en) | A kind of knowledge assistance STAP detection methods under complex Gaussian background | |
| Cotter et al. | Classification of broadband target spectra in the mesopelagic using physics-informed machine learning | |
| Rutten et al. | Efficient particle-based track-before-detect in Rayleigh noise | |
| Hu et al. | A new way to model nonstationary sea clutter | |
| CN112147584A (en) | MIMO radar extended target detection method based on non-uniform clutter | |
| Hu et al. | Statistic characteristic analysis of forward scattering surface clutter in bistatic radar | |
| CN105548986B (en) | Object detection method based on prewhitening median ratio detector under sea clutter background | |
| Salmasi et al. | Design and analysis of fractal detector for high resolution radars | |
| CN106019250A (en) | Repeating false target discriminating method based on angular glint | |
| CN105467374B (en) | Object detection method based on prewhitening ratio cell average detector under sea clutter background | |
| CN104215960A (en) | Target tracking method based on improved particle filters | |
| Tian et al. | Multiresolution jamming recognition with few-shot learning | |
| Oyedokun | Sea clutter simulation |
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 |