CN105891797A - Small target detection method based on sea clutter nonlinear characteristics - Google Patents

Abstract

The invention provides a small target detection method based on sea clutter nonlinear characteristics. The method comprises the following steps of: a, receiving radar sea clutter data after radar irradiates a detected sea area; b, dividing, according to data points in the radar sea clutter data, the radar sea clutter data into data segments including a set number of data points; c, for each data segment, generating corresponding substitute data by using an IAAFT method; and d, analyzing the nonlinear characteristic of the substitute data of each data segment and determining whether a target exists in the radar sea clutter data. The small target detection method has beneficial effects of no demands for prior information of the sea clutter and extensive applicability.

Description

A kind of small target detecting method based on sea clutter nonlinear characteristic

Technical field

The invention belongs to radar data processing technology field, more particularly to one based on the non-linear spy of sea clutter The small target detecting method of property.

Background technology

Chaos is the origin of non-linear dynamic model, and the nonlinear characteristic therefore detecting sea clutter is mixed to judging Ignorant feature, and utilize nonlinear model carry out sea clutter modeling there is important application.Therefore sea is being judged Clutter has chaotic characteristic, first should check whether data have nonlinear characteristic.

The chaos time sequence arrived for actual observation, basic problem is to judge that Time-series System is linear Or the most nonlinear, this is the problem of non-linear test of this sequence.Replacement based on assumed statistical inspection The principle of data method, in recent years, the time series at non-linear nature is tested and is widely used in.Sinology Person is also with its underwater noise and economic data, traffic flow data etc..1996, Shi Leibai proposed amplitude The Fourier transformation (IAAFT) that coupling iteration amplitude adjusts is that a kind of stable alternate data generates method. This method can well match with original Fourier spectrum data and probability density distribution, so by extensively It is applied to nonlinear data test generally.

Research finds, the non-linear nature of sea clutter is to have stable time and space.The more important thing is, The non-linearity confrontation Small object of sea clutter has the strongest sensitivity, though weak signal target energy appreciable impact non-thread Property characteristic.By analyzing containing target data and the non-linear nature of sea clutter data, it is proposed that a kind of new The alternative method of target detection data.By the test of radar data under various sea conditions, the method is to weak The detection performance of target is good.

Therefore, it is necessary to provide a kind of small target detecting method based on sea clutter nonlinear characteristic.

Summary of the invention

It is an object of the invention to provide a kind of small target detecting method based on sea clutter nonlinear characteristic.

Technical scheme is as follows: a kind of small target detecting method bag based on sea clutter nonlinear characteristic Include following steps:

A, reception radar are to the radar sea clutter data after the irradiation of detection marine site；

B, according to the data point in described radar sea clutter data, described radar sea clutter data are divided into and include Set multiple data segments of the data point of quantity；

C, the alternate data corresponding to the employing IAAFT method generation of each described data segment；

D, analyze the nonlinear characteristic of the alternate data of each described data segment, and judge described radar sea clutter Whether data exist target.

In the small target detecting method based on sea clutter nonlinear characteristic that the embodiment of the present invention provides, in step In rapid c, the process utilizing IAAFT method to generate corresponding alternate data generation alternate data includes as follows Step:

C1, remember that original observation sea clutter is { X (n) }, calculate its amplitude rank { (K) } respectively, and Fu In the amplitude spectrum of vertical leaf transformation discrete square:

${\left|S_k\right|}^2={\left|\frac{1}{\sqrt{N}}{\mathrm{\Σ}}_{n=0}^{N-1}{s}_n{e}^{i2\mathrm{\π}kn/N}\right|}^2;$

C2, random scrambling former sea clutter sequence { X (n) }, obtain new sea clutter random sequence

C3, calculatingFourier dissipate fromMakePlural number is equal toThus obtain equation:

${\stackrel{\‾}{S_n}}^{\left(0\right)}=\frac{1}{\sqrt{N}}{\mathrm{\Σ}}_{n=0}^{N-1}{e}^{{{\mathrm{i\ψ}}_k}^{\left(0\right)}}\left|S_k\right|e^{i2\mathrm{\π}kn/N};$

C4, allow sequenceA new sequence is regained according to { c (k) }Wherein, And { x (n) } has an identical probability density distribution, and withPower spectral density distribution closer to | S_k|²；

C5, repeat step c3 and c4, give the sequence { x (n) } limited, thus the corresponding sequence producedAnd by described sequenceIt is denoted asAlternate data.

In the small target detecting method based on sea clutter nonlinear characteristic that the embodiment of the present invention provides, described Alternate data eliminates the non-linear nature of initial data, but maintains the linear character of initial data.

In the small target detecting method based on sea clutter nonlinear characteristic that the embodiment of the present invention provides, described Radar sea clutter data have Third order statistic t^C3(τ):

t^C3(τ)=mean (x_nx_n-τx_n-2τ)。

In the small target detecting method based on sea clutter nonlinear characteristic that the embodiment of the present invention provides, in step In rapid d, use non-linear to the alternate data of described data segment of Gaussian Distribution Parameters method or sort method method Characteristic carries out statistical analysis.

The beneficial effects of the present invention is: described small target detecting method based on sea clutter nonlinear characteristic is adopted It is replaced with data method the nonlinear characteristic of sea clutter is analyzed, and utilizes test statistics method to carry out The detection of Small object in described detection marine site.Therefore, described Small object based on sea clutter nonlinear characteristic inspection Survey method need not the prior information to sea clutter, has the widely suitability.

Accompanying drawing explanation

Fig. 1 is the stream of the small target detecting method based on sea clutter nonlinear characteristic that the embodiment of the present invention provides Journey block diagram.

Detailed description of the invention

In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing and reality Execute example, the present invention is further elaborated.Only should be appreciated that specific embodiment described herein Only in order to explain the present invention, it is not intended to limit the present invention.

The description of specific distinct unless the context otherwise, the element in the present invention and assembly, quantity both can be single Presented in individual, it is also possible to presented in multiple, this is not defined by the present invention.In the present invention Although step arranged with label, but be not used to limit the precedence of step, unless specifically Based on understanding that the order of step or the execution of certain step need other steps, otherwise step is the most secondary Sequence is adjustable in.It is appreciated that term "and/or" used herein relates to and contains to be associated One or more of any and all possible combination in Listed Items.

Refer to Fig. 1, be the small target deteection side based on sea clutter nonlinear characteristic of embodiment of the present invention offer The integrated schematic diagram of method.The small target detecting method based on sea clutter nonlinear characteristic 100 that the present invention provides Be applicable to the nonlinear characteristic by analyzing the radar sea clutter data radar return to the Small object on sea Detect.Wherein, showing according to measured data, the sea clutter of radar has nonlinear character, and Stable and the outside weather conditions change of this character has the strongest sensitivity.

The small target detecting method based on sea clutter nonlinear characteristic 100 provided in the present invention uses replacement number Non-linear nature according to method test sea clutter.The basic thought of described surrogate data technique is: first specify One linear stochastic process is null hypothesis, and assumes that use generating algorithm generates the alternate data of a group based on this, Then calculate initial data and the statistic of test of proxy data, finally use a statistical test, according to former Difference between time series and the alternate data come determines to accept or refusal null hypothesis.

And, described surrogate data technique includes four aspects: null hypothesis, alternate data generating algorithm, inspection Test statistic and Statistical Identifying Method.

Wherein, in described null hypothesis, described surrogate data technique often has three null hypothesiss:

Null hypothesis 1: observation data are randomly generated by variable, and this is independent.

Null hypothesis 2: observation data are to be produced by linear stochastic process.

Null hypothesis 3: the stochastic process after static non linear conversion is linear.

In three null hypothesiss, because the data in the form of sea clutter are complicated, intuitively, it can not The hypothesis 1 that can use and hypothesis 2.Additionally, null hypothesis 3 contained front 2 it is assumed that the most only make By null hypothesis 3, it is assumed that the static non linear after the Gaussian noise of the linear correlation that sea clutter amplitude data produces becomes Change.Static non linear conversion is the static non linear conversion with nonlinear characteristic, and its static state refers to observation Observed result during result is only dependent upon the value of dynamic process, and unrelated with former value and former time.

Such as: represent the output time series of system with x (t), observing function g (n) is:

G (n)=g (x (t))

Actual radar data is nonlinear.Because the data in discrete sampling are nonlinear functions, and By the linear process of nonlinear observer, he will have nonlinear characteristic.

Will be as described alternate data generating algorithm, described statistic of test and described Statistical Identifying Method Next the description to described small target detecting method 100 based on sea clutter nonlinear characteristic is introduced.

Refer to Fig. 1, be the small target deteection side based on sea clutter nonlinear characteristic of embodiment of the present invention offer The FB(flow block) of method.In the present embodiment, described small target detecting method based on sea clutter nonlinear characteristic 100 comprise the steps:

Step S1, reception radar are to the radar sea clutter data after the irradiation of detection marine site.

Specifically, utilize radar that described detection marine site is irradiated, receive and store described radar return, From the little mesh determining surface, described detection marine site by analyzing the radar sea clutter data described radar return Mark information.

Step S2, according to the data point in described radar sea clutter data, described radar sea clutter data are divided Become the multiple data segments including setting the data point of quantity.

Specifically, each described radar sea clutter data includes multiple data point, uses and sets rule by institute Stating multiple data point to carry out splitting in multiple data segments, each described data segment comprises the data point setting quantity. Such as, the radar data of each radar length (each range gate) is 130000 data points, according to described Described data are divided into the data segment of 260 sections, each described data segment by the chaos time sequence of radar data There are 1000 data points, between two adjacent described data segments, repeat 500 data points.

Step S3, the alternate data corresponding to the employing IAAFT method generation of each described data segment.

Specifically, the process using IAAFT method to generate corresponding alternate data described data segment includes Following steps:

(1) remember that original observation sea clutter is { X (n) }, calculate its amplitude rank { (K) } respectively, and In the amplitude spectrum of Fourier transform discrete square:

${\left|S_k\right|}^2={\left|\frac{1}{\sqrt{N}}{\mathrm{\Σ}}_{n=0}^{N-1}{s}_n{e}^{i2\mathrm{\π}kn/N}\right|}^2;$

(2) random scrambling former sea clutter sequence { X (n) }, obtains new sea clutter random sequence

(3) calculateFourier dissipate fromMakePlural number is equal toThus obtain equation:

${\stackrel{\‾}{S_n}}^{\left(0\right)}=\frac{1}{\sqrt{N}}{\mathrm{\Σ}}_{n=0}^{N-1}{e}^{{{\mathrm{i\ψ}}_k}^{\left(0\right)}}\left|S_k\right|e^{i2\mathrm{\π}kn/N};$

(4) sequence is allowedA new sequence is regained according to { c (k) }Wherein,And { x (n) } has an identical probability density distribution, and withPower spectral density distribution closer to | S_k|²；

(5) repeat step 3 and step 4, give the sequence { x (n) } limited, thus the corresponding sequence producedAnd by described sequenceIt is denoted asAlternate data.

It should be noted that the alternate data obtained in described step S3, eliminate the non-thread of initial data Property character, but maintain the linear character of initial data.

Step S4, analyze the nonlinear characteristic of the alternate data of each described data segment, and judge described radar Whether sea clutter data exist target.

Specifically, in described step S4, use statistical methods for experiment to the alternate data of described data segment Nonlinear characteristic carries out statistical analysis, such as: use Gaussian Distribution Parameters method or sort method method to described The nonlinear characteristic of the alternate data of data segment carries out statistical analysis.

And, for radar sea clutter, described radar sea clutter data have Third order statistic t^C3(τ):

t^CB(τ)=mean (x_nx_n-τx_n-2τ)。

Based on described Third order statistic t^C3(τ), use statistical methods for experiment to the alternate data of described data segment Nonlinear characteristic carries out statistical analysis.

For example, it is assumed that the normal distribution of statistic of test, initial data and alternate data collection use Gauss distribution Parametric method is tested.But it is true that the statistics of alternate data might not meet normal distribution, so This time will there is deviation with parametric test.

Such as, the sort method (or grade) using Shi Leibai to propose, then carry out test statistics.The method First 1/1a-1 (one-sided test) or the alternate data of 2/a-1 group (detection of two-tailed test dual edge) are produced, Then initial data and alternate data and the test statistics of sequence are calculated.If the original series of statistic of test Being minimum or maximum, we refuse null hypothesis.

And, by described statistical methods for experiment, the non-linear nature of each section of described data segment is added up Analyze, and filter out the data segment with non-linear nature.The non-linear nature that can destroy due to Small object Sea clutter, and the nonlinear characteristic of sea clutter has the strongest sensitivity to target, therefore, it can lead to The non-linear nature crossing described data segment judges whether whether there is Small object in described detection marine site.

Compared to prior art, the small target detecting method based on sea clutter nonlinear characteristic that the present invention provides 100 use surrogate data technique to be analyzed the nonlinear characteristic of sea clutter, and utilize test statistics method Carry out the detection of Small object in described detection marine site.Therefore, described little mesh based on sea clutter nonlinear characteristic Mark detection method 100 need not the prior information to sea clutter, has the widely suitability.

It is obvious to a person skilled in the art that the invention is not restricted to the details of above-mentioned one exemplary embodiment, And without departing from the spirit or essential characteristics of the present invention, it is possible to realize in other specific forms The present invention.Therefore, no matter from the point of view of which point, embodiment all should be regarded as exemplary, and right and wrong Restrictive, the scope of the present invention is limited by claims rather than described above, it is intended that will fall All changes in the implication of equivalency and scope of claim are included in the present invention.Should will not weigh Any reference during profit requires is considered as limiting involved claim.

Moreover, it will be appreciated that although this specification is been described by according to embodiment, but the most each enforcement Mode only comprises an independent technical scheme, and this narrating mode of description is only for clarity sake, Those skilled in the art should be using description as an entirety, and the technical scheme in each embodiment can also be through Appropriately combined, form other embodiments that it will be appreciated by those skilled in the art that.

Claims (5)

1. a small target detecting method based on sea clutter nonlinear characteristic, it is characterised in that: include as follows Step:

A, reception radar are to the radar sea clutter data after the irradiation of detection marine site；

B, according to the data point in described radar sea clutter data, described radar sea clutter data are divided into and include Set multiple data segments of the data point of quantity；

C, the alternate data corresponding to the employing IAAFT method generation of each described data segment；

D, analyze the nonlinear characteristic of the alternate data of each described data segment, and judge described radar sea clutter Whether data exist target.

Small target detecting method based on sea clutter nonlinear characteristic the most according to claim 1, it is special Levy and be: in step c, utilize IAAFT method to generate corresponding alternate data and produce alternate data Process comprises the steps:

C1, remember that original observation sea clutter is { X (n) }, calculate its amplitude rank { (K) } respectively, and Fu In the amplitude spectrum of vertical leaf transformation discrete square:

${\left|S_k\right|}^2={\left|\frac{1}{\sqrt{N}}{\mathrm{\Σ}}_{n=0}^{n=1}{s}_n{e}^{i2\mathrm{\πkn}/N}\right|}^2;$

C2, random scrambling former sea clutter sequence { X (n) }, obtain new sea clutter random sequence

C3, calculatingFourier dissipate fromMakePlural number is equal toThus obtain equation:

${\stackrel{\‾}{S_n}}^{\left(0\right)}=\frac{1}{\sqrt{N}}{\mathrm{\Σ}}_{n=0}^{N-1}{e}^{{{\mathrm{i\ψ}}_k}^{\left(0\right)}}\left|S_k\right|e^{i2\mathrm{\π}kn/N};$

C4, allow sequenceA new sequence is regained according to { c (k) }Wherein, And { x (n) } has an identical probability density distribution, and withPower spectral density distribution closer to | S_k|²；

C5, repeat step c3 and c4, give the sequence { x (n) } limited, thus the corresponding sequence producedAnd by described sequenceIt is denoted asAlternate data.

Small target detecting method based on sea clutter nonlinear characteristic the most according to claim 2, it is special Levy and be: described alternate data eliminates the non-linear nature of initial data, but maintains initial data Linear character.

Small target detecting method based on sea clutter nonlinear characteristic the most according to claim 2, it is special Levy and be: described radar sea clutter data have Third order statistic t^C3(τ):

t^C3(τ)=mean (x_nx_n-τx_n-2τ)。

Small target detecting method based on sea clutter nonlinear characteristic the most according to claim 4, it is special Levy and be: in step d, use Gaussian Distribution Parameters method or the sort method method replacement to described data segment The nonlinear characteristic of data carries out statistical analysis.