CN102176000A - Sea clutter suppression method for marine radar - Google Patents

Abstract

The invention discloses a method for suppressing sea clutter of marine radar. Most of the existing marine radars do not deal with the nearby sea clutter. In the existing sea clutter suppression methods, the sea clutter suppression is generally performed according to a fixed template, which will greatly attenuate the signal amplitude of the nearby target, and even cause leakage. Check happens. The present invention proposes in view of above-mentioned shortcoming. The method of the invention includes the following steps: initializing parameters and data, estimating parameters, suppressing sea clutter and updating estimated samples. Users can choose different sea clutter distributions to estimate sea clutter according to different sea conditions, and judge whether the sample data needs to be updated for re-estimation according to the suppression effect. While suppressing sea clutter as much as possible, the target The influence is reduced to the minimum, and the problem of greatly attenuating the signal amplitude of the nearby target is overcome.

Description

A method of suppressing sea clutter in marine radar

technical field

The invention belongs to the technical field of marine radar, in particular to the suppression of sea clutter therein.

Background technique

Marine radar is one of the indispensable navigation equipment for ships, but the detection and tracking performance of marine radar is often affected by the scattered electromagnetic waves on the surface of the waves (that is, sea clutter), especially when detecting small targets flying at low altitudes skimming the sea. The clutter problem is more serious. It is well known that clutter increases the false alarm probability or correspondingly reduces the detection probability for a constant false alarm rate. This situation may be even more serious for sea clutter, because compared with ground clutter, sea clutter is moving and undulating, so it is difficult to eliminate. Most of the existing marine radars do not deal with the sea clutter near the radar, which leads to the so-called "big pie" phenomenon in the center. In the existing sea clutter suppression methods, the sea clutter suppression is generally performed according to a fixed template, which will greatly attenuate the signal amplitude of nearby targets, and even lead to missed detection.

Contents of the invention

The purpose of the present invention is to solve the problem that in the existing marine radar sea clutter suppression method, while suppressing the sea clutter near the radar, the signal amplitude of nearby targets may be greatly attenuated. Suppression method of radar sea clutter.

In order to achieve the above object, the technical solution of the present invention is: a method for suppressing marine radar sea clutter, comprising the steps of:

S1. Initialization, complete the initialization setting of parameters and data initialization, specifically including the following sub-steps:

S11. Select the distribution of sea clutter according to the actual sea situation and the performance of the radar;

S12. The setting of suppression method parameters, N: the number of scan lines processed in the sample update period, M: the number of scan lines required for sample estimation, α: threshold for re-estimation, β: threshold for binarization;

计算一条扫描线上海杂波的数据点数，其中n是一条扫描线上海杂波的数据点数，D为雷达附近的海杂波距离雷达的最远距离，Range为雷达的探测距离，Num为一条扫描线上数据点的个数，得到M条扫描线的前n个扫描线数据点{T_i(k)，i＝1，2...M，k＝1，...，n}，取其平均值，记为

并保存为样本数据；S13. Read M scanning line data, denoted as {T _i (k), i=1, 2...M, k=1,..., Num}, by the formula

Calculate the number of data points of sea clutter in a scan line, where n is the number of data points of sea clutter in a scan line, D is the farthest distance between the sea clutter near the radar and the radar, Range is the detection distance of the radar, and Num is a scan The number of data points on the line, to obtain the first n scan line data points {T _i (k), i=1, 2...M, k=1,..., n} of the M scan lines, take its average value, denoted as

and saved as sample data;

进行数据映射，得到映射后的海杂波数据D(k)；S2. parameter estimation and data mapping, select the distribution of sea clutter by sample data C(k) and step S11, estimate the parameters of the distribution, and obtain sea clutter data {B(k), k=1,...,n }, using the formula

Carry out data mapping to obtain the mapped sea clutter data D(k);

并保存；S3. Read N scan line data and perform sea clutter suppression processing, that is, subtract the N scan line data from the sea clutter data D(k) estimated in step S2, and then binarize the result according to the threshold β Processing, when the suppression is carried out to NM pieces, while performing sea clutter suppression, read the first n point data of N-M+1 to N pieces of scanning data, take the average value, and record it as

and save;

S4. Judging whether the radar detection distance has changed, if changed, execute step S13, otherwise execute step S5;

则不需要重新估计海杂波数据，执行步骤S3；如果不满足，用步骤S3中保存的M条前n个的扫描线数据点的平均值C′(k)更新样本数据C(k)，执行步骤S2，重新估计参数。S5. Update of sample data, if satisfied

Then there is no need to re-estimate the sea clutter data, execute step S3; if not satisfied, update the sample data C(k) with the average value C'(k) of the M first n scan line data points saved in step S3, Execute step S2 to re-estimate the parameters.

Wherein, the distribution of sea clutter in step S11 is Rayleigh distribution, lognormal distribution, Weibull distribution and K distribution.

It is assumed here that the sample data sequence used to estimate the sea clutter distribution on one scanning line of the radar is Z={z _i , i=1, Ln}, and n is the number of data points. The specific process of parameter estimation described in step S2 is as follows:

(1) Rayleigh distribution: The probability density function of Rayleigh distribution is as follows

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; )</span>& {\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>\mathrm{<span\; class="notranslate">\; 0</span>}\\ \text{<span class="notranslate"> 0</span>}& {\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right.$

Among them, ^σ2 is the average power, then the parameter σ of the Rayleigh distribution can be estimated according to the sample data;

${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>$

(2) Lognormal distribution: The probability density function of the lognormal distribution is as follows

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}\mathrm{<span\; class="notranslate">\; \&sigma;</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>\frac{{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; ln</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&sigma;</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>$

Among them, σ is the shape parameter, which is the standard deviation of the lognormal distribution ln z _i . As σ increases, the tail of the probability density distribution becomes longer; μ is the scale parameter, which is the lognormal distribution ln z the mean of _i . According to the different sea conditions, the parameters σ and μ of the lognormal distribution can be estimated according to the sample data by taking between 0.5 and 1.2;

$\left\{\begin{array}{c}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; ln</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; ln</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\end{array}\right.$

(3) Weibull distribution: The probability density function of Weibull distribution is as follows,

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; q</span>}}{<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; q</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ]</span>{<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>$

Among them, q is a scale parameter, p is a shape parameter, and the value range is 0<p≤2. The distribution parameters (p, q) are estimated as follows:

method one:

The Weibull distribution of sea clutter has empirical values, here it is set as (p ₀ , q ₀ ), and the iteration interval of p, q is designed (p ₀ -δ _p , p ₀ +δ _p ), (q ₀ -δ p _q , q ₀ +δ _q ), where δ _p , δ _q are pre-set values according to the actual situation, which determine the size of the iteration interval of p and q. Here, q is the outer loop and p is the inner loop to perform curve fitting. Combine, record the difference between the fitted curve data and the sample data each time, when the difference is less than a given threshold, the p, q values of this iteration are used as parameters.

Method Two:

<math><mfenced open='{' close=''><mtable><mtr><mtd><mover><mi>p</mi><mo>^</mo></mover><mo> =</mo><msup><mrow><mo>{</mo><mfrac><mn>6</mn><msup><mi>&amp;pi;</mi><mn>2</ mn></msup></mfrac><mfrac><mi>n</mi><mrow><mi>n</mi><mo>-</mo><mn>1</mn></ mrow></mfrac><mtext>[&lt;</mtext><msup><mrow><mo>(</mo><mi>ln</mi><mrow><mo>(</mo>< mi>Z</mi><mo>)</mo></mrow><mo>)</mo></mrow><mn>2</mn></msup><mo>></mo ><mo>-</mo><msup><mrow><mo>&lt;</mo><mi>ln</mi><mrow><mo>(</mo><mi>Z</mi ><mo>)</mo></mrow><mo>></mo></mrow><mn>2</mn></msup><mo>]</mo><mo>}< /mo></mrow><mrow><mo>-</mo><mfrac><mn>1</mn><mn>2</mn></mfrac></mrow></msup>< /mtd></mtr><mtr><mtd><mrow><mover><mi>q</mi><mo>^</mo></mover><mo>=</mo><mi> exp</mi><mo>[</mo><mo>&lt;</mo><mi>ln</mi><mrow><mo>(</mo><mi>Z</mi>< mo>)</mo></mrow><mo>></mo><mo>+</mo><mn>0.5772</mn><msup><mover><mi>p</mi>< mo>^</mo></mover><mrow><mo>-</mo><mn>1</mn></mrow></msup><mo>]</mo></mrow> </mtd></mtr></mtable></mfenced></math>

Where <.> represents moment estimation, which is common knowledge in this field and will not be described in detail here.

(4) K distribution: The probability density function of K distribution is as follows

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; c</span>}}{\mathrm{<span\; class="notranslate">\; \&Gamma;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; c</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; v</span>}}{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; c</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}$

Among them, K _v (.) is the second type of modified Bessel function, c is the scale parameter, which affects the average power of clutter, the smaller c is, the smaller the clutter intensity is, and the shape parameter v reflects the degree of bias of the K distribution , the smaller the v, the more obvious the asymmetry of the distribution, and the greater the deviation from the Rayleigh distribution. v generally varies between 0.1 and 10, when v=∞ is a Rayleigh distribution, where Γ(v) refers to the Γ function.

The parameters of the K distribution adopt the method of moment estimation. For known samples, the origin moment of the sample is used to estimate the corresponding origin moments of each order of the population:

表示z_i的K次方， ${\hat{m}}_K=\frac{1}{\mathrm{no}}\underset{i=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{z}_i^K,K\&Greater\; Equal;0,$ in

Indicates the Kth power of z _i ,

Then the empirical estimation formula of moment estimation of v and c is as follows:

$\left\{\begin{array}{c}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>\frac{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\\ \stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}\sqrt{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; /</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}}\end{array}\right.$

分别为样本数据的二阶，四阶原点矩。in,

are the second-order and fourth-order origin moments of the sample data, respectively.

Beneficial effects of the present invention: With the sea clutter suppression method proposed by the present invention, the user can select different sea clutter distributions for sea clutter estimation according to different sea conditions, and judge whether the sample data needs to be updated for re-estimation according to the suppression effect. While suppressing the sea clutter as much as possible, the impact on the target is minimized, and the problem of greatly attenuating the signal amplitude of the nearby target is overcome.

Description of drawings

Fig. 1 is a schematic flow chart of the method for suppressing marine radar sea clutter of the present invention.

FIG. 2 is a schematic diagram of marine radar scan data of the present invention.

Fig. 3 is a schematic diagram of a result of sea clutter suppression using Weibull distribution on a piece of scanning data according to an embodiment of the present invention.

Fig. 4 is a schematic diagram of measured data of a marine radar according to an embodiment of the present invention.

FIG. 5 is a schematic diagram of sea clutter suppression results using Weibull distribution on measured data according to an embodiment of the present invention.

Detailed ways

Below in conjunction with accompanying drawing, provide the specific embodiment of the present invention. It should be noted that the parameters in the examples do not affect the generality of the present invention.

The characteristics of sea clutter include two aspects: one is to determine the amplitude distribution and power spectrum type of sea clutter; the other is to determine the parameters of the amplitude distribution and power spectrum model according to the specific radar system and radar working environment. The sea clutter amplitude distribution function is very important for designing signal and information processing algorithms such as target detection, estimation, tracking and recognition. The schematic flow chart of the embodiment of the present invention is shown in Figure 1, specifically as follows:

S1. Initialization, complete the initialization of parameters and data initialization, as follows:

S11. According to the actual sea situation and the performance of the radar, the Weibull distribution is selected here,

S12. Parameter setting: the number of scan lines processed in each sample update period N=100, the number of scan lines required for each sample estimation M=15, the threshold of re-estimating parameters α=0.2, and the threshold of binarization β=60,

计算一条扫描线上海杂波的数据点数，其中n是一条扫描线上海杂波的数据点数，如图2所示，D为雷达附近的海杂波距离雷达的最远距离，这里D＝3海里，雷达的探测距离Range＝60海里，一条扫描线上数据点的个数Num＝1000，即可得到15条扫描线的前50个点扫描线数据{T_i(k)，i＝1，2...15，k＝1，...，50}，取其平均值，记为

并保存为样本数据。S13. Initialization of sample data: read 15 scan lines, recorded as {T _i (k), i=1, 2...15, k=1,..., Num}, by the formula

Calculate the number of data points of sea clutter on a scan line, where n is the number of data points of sea clutter on a scan line, as shown in Figure 2, D is the farthest distance from the sea clutter near the radar to the radar, where D=3 nautical miles , the radar detection range Range=60 nautical miles, the number of data points Num=1000 on one scan line, the first 50 point scan line data {T _i (k), i=1, 2 of 15 scan lines can be obtained ...15, k=1, ..., 50}, take the average value, and write it as

and save as sample data.

S2. parameter estimation, select the distribution of sea clutter by sample data C(k) and step S1, estimate the parameter of distribution, obtain sea clutter data;

From the sample data C(k) and the Weibull distribution obeyed by the sea clutter, the required parameters p, q are estimated, and the specific process is as follows:

The Weibull distribution probability density function is as follows,

${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; q</span>}}{<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; q</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ]</span>{<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>$

In the formula, q is a scale parameter, which is generally smaller than the peak position of the curve distribution; p is a shape parameter, and the value range is 0<p≤2. When the shape parameter p=1, the Weibull distribution degenerates into an exponential distribution; when the shape parameter When p=2, Weibull distribution degenerates into Rayleigh distribution. As the shape parameter p decreases, the tail of the probability density distribution becomes longer; here it is assumed that the data sequence used to estimate the sea clutter distribution on a radar scan line is Z={z _i , i=1, L n}, n is the number of data points.

时，就以此次迭代的p，q值作为估计的参数，即得到海杂波数据{B(k)，k＝1，...，50}。Generally, the Weibull distribution of sea clutter has empirical values, here it is set to (1, 15), here it is taken as the iterative interval (0, 2), (0, 20) of the design p, q, and the iterative operation is carried out. Curve fitting, record the difference between each fitted curve and the actual radar data, when the difference is less than a given threshold

When , the p and q values of this iteration are used as the estimated parameters to obtain the sea clutter data {B(k), k=1,...,50}.

Because the estimated sea clutter data is a sampling of the probability distribution, and the data interval is (0, 1), it is necessary to map the data, that is, to map the obtained sea clutter data to a range equivalent to the scan line data, that is, according to formula Perform data mapping to obtain the mapped sea clutter data {D(k), k=1,...,50};

The results of sea clutter suppression for a piece of radar scan data are shown in Figure 3. It can be seen that the sea clutter near the radar is well estimated, so that the target submerged by the sea clutter can be effectively detected.

S3. Read 100 scan line data and perform sea clutter suppression processing, that is, the sea clutter data {T _i (k), i=1, 2...100, k=1, ..., 50}, subtracted from the sea clutter data {D(k), k=1, ..., 50} estimated in step S2, and then binarized the result according to the threshold 60. When 85 lines are suppressed, while performing clutter suppression, save the average value of the last 15 scanning line data of the first 50 points, which is recorded as

S4. Judging whether the radar detection distance has changed, if changed, execute step S13, otherwise execute step S5;

则不需要重新估计海杂波数据，执行步骤S3；如果不满足，用步骤S3中保存的15条前50个点的扫描线数据的平均值C′(k)更新样本数据C(k)，执行步骤S2，重新估计参数。S5. Update of sample data, if satisfied

Then there is no need to re-estimate the sea clutter data, execute step S3; if not satisfied, update the sample data C(k) with the average value C'(k) of the scan line data of the 15 first 50 points saved in step S3, Execute step S2 to re-estimate the parameters.

The scanning data of one frame of radar is shown in Figure 4, and the suppression result is shown in Figure 5. It can be seen that the sea clutter near the radar has been effectively suppressed, and the target submerged by the sea clutter has been effectively detected.

What selected in this embodiment is Weibull distribution, but the present invention is not limited to Weibull distribution, in actual processing, also can select other three distributions-Rayleigh distribution, logarithmic normal according to different sea situation distribution, K distribution.

The Rayleigh distribution is a classic function to describe the amplitude distribution of sea clutter, which is suitable for describing the sea clutter of low-resolution radar. When there are a large number of independent scattering sources in a sea clutter unit, there is no contribution When there are obvious individuals, the envelope amplitude of radar sea clutter obeys the Rayleigh distribution; the lognormal distribution is suitable for high-resolution sea clutter data in some flat areas; in the case of high-resolution radar and low incident angle, The sea clutter of general sea conditions can be accurately described by the Weibull distribution; the K distribution is a good fit for the sea clutter envelope data obtained when the radar works at a low grazing angle, and it can not only simulate the sea clutter amplitude distribution The "long tail" characteristic can also correctly simulate its time correlation. This characteristic is very important for accurately predicting the detection performance after echo pulse accumulation. At the same time, the K distribution also has a wide range of adaptation and is suitable for different types of radar sea clutter.

It can be seen from this embodiment that with the sea clutter suppression method proposed by the present invention, the user can select different sea clutter distributions for sea clutter estimation according to different sea conditions, and judge whether the sample data needs to be updated for re-according to the suppression effect. It is estimated that while the sea clutter is suppressed as much as possible, the impact on the target is minimized, and the problem of greatly attenuating the signal amplitude of the nearby target is overcome.

Those skilled in the art will appreciate that the embodiments described herein are to help readers understand the principles of the present invention, and it should be understood that the protection scope of the invention is not limited to such specific statements and embodiments. All possible equivalent replacements or changes made according to the above description are considered to belong to the protection scope of the claims of the present invention.

Claims (7)

1. A method for suppressing marine radar sea clutter, comprising the steps of: S1. Initialization, complete the initialization setting of parameters and data initialization, specifically including the following sub-steps: S11. Select the distribution of sea clutter according to the actual sea situation and the performance of the radar; S12. The setting of suppression method parameters, N: the number of scan lines processed in the sample update period, M: the number of scan lines required for sample estimation, α: threshold for re-estimation, β: threshold for binarization; S13. Read M scanning line data, denoted as {T _i (k), i=1, 2...M, k=1,..., Num}, by the formula

Calculate the number of data points of sea clutter in a scan line, where n is the number of data points of sea clutter in a scan line, D is the farthest distance between the sea clutter near the radar and the radar, Range is the detection distance of the radar, and Num is a scan The number of data points on the line, to obtain the first n scan line data points {T _i (k), i=1, 2...M, k=1,..., n} of the M scan lines, take its average value, denoted as

and saved as sample data; S2. parameter estimation and data mapping, select the distribution of sea clutter by sample data C(k) and step S11, estimate the parameters of the distribution, and obtain sea clutter data {B(k), k=1,...,n }, using the formula Carry out data mapping to obtain the mapped sea clutter data D(k); S3. Read N scan line data and perform sea clutter suppression processing, that is, subtract the N scan line data from the sea clutter data D(k) estimated in step S2, and then binarize the result according to the threshold β Processing, when the suppression is carried out to NM pieces, while performing sea clutter suppression, read the first n point data of N-M+1 to N pieces of scanning data, take the average value, and record it as

and save; S4. Judging whether the radar detection distance has changed, if changed, execute step S13, otherwise execute step S5; S5. Update of sample data, if satisfied

Then there is no need to re-estimate the sea clutter data, execute step S3; if not satisfied, update the sample data C(k) with the average value C'(k) of the M first n scan line data points saved in step S3, Execute step S2 to re-estimate the parameters. 2. The method for suppressing marine radar sea clutter according to claim 1, wherein the distribution of said sea clutter is Rayleigh distribution, lognormal distribution, Weibull distribution and K distribution. 3. the suppressing method of marine radar sea clutter according to claim 2, is characterized in that, the concrete process of the parameter estimation of described Rayleigh distribution is: Rayleigh distribution probability density function is as follows: ${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; )</span>& {\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>\mathrm{<span\; class="notranslate">\; 0</span>}\\ \text{<span class="notranslate"> 0</span>}& {\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; <</span>\mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right.$ Among them, σ ² is the average power, {z _i , i=1, L n} is the sample data sequence used to estimate the sea clutter distribution on one scanning line of the radar, and n is the number of data points, then according to the sample The data estimate the parameter σ of the Rayleigh distribution; ${\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>$ 4. the suppression method of marine radar sea clutter according to claim 2 is characterized in that, the concrete process of the parameter estimation of described logarithmic normal distribution is: the probability density function of logarithmic normal distribution is as follows ${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\sqrt{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;</span>}}\mathrm{<span\; class="notranslate">\; \&sigma;</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>\frac{{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; ln</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; ]</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&sigma;</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>$ Among them, σ is the shape parameter, which is the standard deviation of the lognormal distribution ln z _i . As σ increases, the tail of the probability density distribution becomes longer; μ is the scale parameter, which is the lognormal distribution ln z The mean value of _i , {z _i , i=1, L n}, is the sample data sequence used to estimate the sea clutter distribution on a radar scan line, and n is the number of data points, which is taken as 0.5 according to the different sea conditions Between 1.2 and 1.2, the parameters σ and μ of the lognormal distribution can be estimated according to the sample data; $\left\{\begin{array}{c}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; ln</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>\\ {\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&sigma;</span>}}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; no</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; ln</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&mu;</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}\end{array}\right.<span\; class="notranslate">\; .</span>$ 5. the suppression method of marine radar sea clutter according to claim 2 is characterized in that, the concrete process of the parameter estimation of described Weibull distribution is: Weibull distribution probability density function is as follows, ${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; q</span>}}{<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; q</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ]</span>{<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>$ Among them, q is the scale parameter, p is the shape parameter, the value range is 0<p≤2, {z _i , i=1, L n} is the sample data used to estimate the sea clutter distribution on one scanning line of the radar sequence, n is the number of data points, and the distribution parameters (p, q) are estimated as follows: the Weibull distribution of sea clutter has empirical values, here it is set to (p ₀ , q ₀ ), and the design of p, q Iteration interval (p ₀ -δ _p , p ₀ +δ _p ), (q ₀ -δ _q , q ₀ +δ _q ), where δ _p and δ _q are pre-set values according to the actual situation and determine p, The size of the q iteration interval. Here, q is used as the outer loop and p is the inner loop. Curve fitting is performed, and the difference between the fitted curve data and the sample data is recorded each time. When the difference is less than a given threshold, it is calculated as The p and q values of this iteration are parameters. 6. the suppression method of marine radar sea clutter according to claim 2 is characterized in that, the concrete process of the parameter estimation of described Weibull distribution is: Weibull distribution probability density function is as follows, ${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; q</span>}}{<span\; class="notranslate">\; (</span>\frac{{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; q</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; [</span><span\; class="notranslate">\; -</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; ]</span>{<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; \&Greater\; Equal;</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; q</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>$ Among them, q is a scale parameter, p is a shape parameter, and the value range is 0<p≤2, Z={z _i , i=1, L n}, which is used to estimate the sea clutter distribution on one scanning line of the radar The sample data sequence, n is the number of data points, and the distribution parameters (p, q) are estimated as follows:

where <.> represents moment estimation. 7. the suppression method of marine radar sea clutter according to claim 2 is characterized in that, the concrete process of the parameter estimation of described K distribution is: K distribution probability density function is as follows, ${\mathrm{<span\; class="notranslate">\; f</span>}}_{\mathrm{<span\; class="notranslate">\; Z</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; c</span>}}{\mathrm{<span\; class="notranslate">\; \&Gamma;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; )</span>}{<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; c</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; v</span>}}{\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; c</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; z</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; c</span>}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0</span>}$ Among them, K _v (.) is the second kind of modified Bessel function, c is the scale parameter, which affects the average power of clutter, and the shape parameter v reflects the bias degree of K distribution, {z _i , i=1, L n}, is the sample data sequence used to estimate the sea clutter distribution on one scanning line of the radar, n is the number of data points, The parameters of the K distribution adopt the method of moment estimation. For known samples, the origin moment of the sample is used to estimate the corresponding origin moments of each order of the population: ${\hat{m}}_K=\frac{1}{\mathrm{no}}\underset{i=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{z}_i^K,K\&Greater\; Equal;0,$ in

Indicates the Kth power of z _i , Then the empirical estimation formula of moment estimation of v and c is as follows: $\left\{\begin{array}{c}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>\frac{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}}{\mathrm{<span\; class="notranslate">\; 2</span>}{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}\\ \stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; c</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}\sqrt{{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; m</span>}}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; /</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; v</span>}}}\end{array}\right.$ in, are the second-order and fourth-order origin moments of the sample data, respectively.

Images