[radar] radar signal online sorting based on kernel clustering with matlab code

1 Content introduction

Radar signal sorting is a necessary function of modern radar reconnaissance equipment ‚ It uses the measured radar signal parameters ‚ Separate randomly overlapping radar signals into independent radar signal columns ‚ Lay a solid foundation for the final radar signal recognition . In Modern Warfare ‚ The signal environment of radar reconnaissance receiver is becoming more and more dense ‚ More and more complex . The traditional signal sorting is mainly based on the idea of Statistics ‚ Through the pulse repetition period of the pulse sequence （PRI） analysis ‚ basis PRI Modulation mode of ‚ Sort out the corresponding radar signals . But this method has a large amount of computation 、 Slow speed 、 The accuracy and signal parameter utilization are not high . This paper will use kernel method to improve fuzzy clustering ‚ Applied to radar signal sorting ‚ And the simulation test is carried out stay Before kernel fuzzy clustering ‚ First, a large number of radar sequence signals are diluted . The reconnaissance opportunity receives signals from radiation sources in all directions at the same time ‚ Because the space position of radar is not It will mutate ‚ So the direction of arrival of the signal （DOA） Parameters will not jump ‚DOA From the physical position, we can distinguish the radiation sources in different directions . actually ‚ The signal to be sorted is the signal mixed by multiple radiation sources from the same direction . therefore ‚ accurate DOA It is a powerful sorting parameter ‚ With accurate DOA As a dense signal flow, pre sorting is an important way to solve the reliable sorting of various special radar signals . There will be a lot of 、 Dense sequence signal data press DOA Diluted into several data sets ‚ In pulse width modulation （PDW） Medium repetition rate （RF）‚ Pulse width （PW） Relatively fixed ‚ Pulse amplitude （PA） Is an uncertain parameter ‚ It cannot be used as the basis for sorting and judgment .

​

2 Simulation code

<span style="color:#333333"><span style="background-color:rgba(0, 0, 0, 0.03)"><code><span style="color:#dd1144">%% use RBF nucleus ,K(X1,X2)=exp(-a*||X1-X2||)</span></code><code>%% Initialization data </code><code>clc</code><code>close all</code><code>​</code><code>a=<span style="color:#0e9ce5">0</span>.<span style="color:#0e9ce5">05</span>;% Kernel function parameters </code><code>e=<span style="color:#0e9ce5">0</span>.<span style="color:#0e9ce5">05</span>;% Look for the similarity measurement threshold of the superior class </code><code>d=<span style="color:#0e9ce5">5</span>;% Share data to determine the number of vectors of merged classes </code><code>s=<span style="color:#0e9ce5">0</span>.<span style="color:#0e9ce5">1</span>;% Initial step </code><code>T=<span style="color:#0e9ce5">100</span>;% Eliminate noise sampling interval </code><code>n=<span style="color:#0e9ce5">3</span>;% Eliminate noise threshold </code><code>a<span style="color:#0e9ce5">0</span>=<span style="color:#0e9ce5">1</span>;% Initial parameter of penalty item </code><code>num=<span style="color:#0e9ce5">1000</span>;%</code><code>q=<span style="color:#0e9ce5">0</span>.<span style="color:#0e9ce5">01</span>;% Variance parameter </code><code>​</code><code><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span>% Simulation data is not sorted <span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span>%</code><code>signal1=signal_create(num,q);</code><code>figure(<span style="color:#0e9ce5">1</span>);</code><code>title(<span style="color:#dd1144">' The data distribution '</span>);</code><code>plot(signal1(<span style="color:#dd1144">:</span>,<span style="color:#0e9ce5">1</span>),signal1(<span style="color:#dd1144">:</span>,<span style="color:#0e9ce5">2</span>),<span style="color:#dd1144">'.b'</span>);</code><code>hold on;</code><code>​</code><code><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span>% Create a new class <span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%</span></code><code>Cdata=cell(1,1000);% Store each type </code><code>Adata=cell(<span style="color:#0e9ce5">1</span>,<span style="color:#0e9ce5">1000</span>);% Store the weight of class vector </code><code>Pdata=zeros(<span style="color:#0e9ce5">1</span>,<span style="color:#0e9ce5">1000</span>);% Offset </code><code>N=zeros(<span style="color:#0e9ce5">1</span>,<span style="color:#0e9ce5">1000</span>);% The total number of various vectors </code><code>share_kind=zeros(<span style="color:#0e9ce5">1000</span>,<span style="color:#0e9ce5">2</span>);% Shared data </code><code>share_num=<span style="color:#0e9ce5">0</span>;</code><code>​</code><code>m=<span style="color:#0e9ce5">1</span>;% Number of classes </code><code>Cdata{<span style="color:#0e9ce5">1</span>,<span style="color:#0e9ce5">1</span>}(<span style="color:#0e9ce5">1</span>,<span style="color:#dd1144">:</span>)=signal1(<span style="color:#0e9ce5">1</span>,<span style="color:#dd1144">:</span>);% The first <span style="color:#0e9ce5">1</span> The first vector of the class </code><code>Adata{<span style="color:#0e9ce5">1</span>,<span style="color:#0e9ce5">1</span>}(<span style="color:#0e9ce5">1</span>,<span style="color:#0e9ce5">1</span>)=s;</code><code>Pdata(<span style="color:#0e9ce5">1</span>)=s;</code><code>N(<span style="color:#0e9ce5">1</span>)=<span style="color:#0e9ce5">1</span>;</code><code>num_data=<span style="color:#0e9ce5">1</span>;% The number of signal vectors has been determined </code><code>f=<span style="color:#0e9ce5">0</span>;</code><code>​</code><code><span style="color:#ca7d37">for</span> j=<span style="color:#0e9ce5">1</span><span style="color:#dd1144">:</span><span style="color:#0e9ce5">2999</span></code><code> num_data=num_data+<span style="color:#0e9ce5">1</span>;     </code><code><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%% Eliminate noise %</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span><span style="color:#dd1144">%%%</span></code><code>%     <span style="color:#ca7d37">if</span> j mod <span style="color:#0e9ce5">100</span>==<span style="color:#0e9ce5">0</span></code><code>%         </code><code>%     <span style="color:#ca7d37">end</span></code><code>,Pdata,m ] = delete_kind( Cdata,Adata,Pdata,N,m,m_out(i));</code><code>     <span style="color:#ca7d37">end</span></code><code><span style="color:#ca7d37">end</span></code><code>​</code><code><span style="color:#ca7d37">end</span></code><code>​</code><code>​</code><code>​</code><code>​</code><code>​</code><code>​</code></span></span>

3 Running results

4 reference

[1] Yu Xinxing , Wang Yong . Radar signal sorting method based on on-line kernel clustering [J]. Computer engineering , 2012, 38(3):270-272.

[2] Ksssssss , Luo Luqin , Zhao Guixi . Radar signal sorting algorithm based on kernel fuzzy clustering [J]. Ship electronic countermeasure , 2009, 32(4):4.

[3] Wang Yili , Yang Yuming . A radar signal sorting algorithm based on improved spectral clustering . 

About bloggers ： Good at intelligent optimization algorithms 、 Neural networks predict 、 signal processing 、 Cellular automata 、 The image processing 、 Path planning 、 UAV and other fields Matlab Simulation , relevant matlab Code problems can be exchanged by private letter .

Some theories cite network literature , If there is infringement, contact the blogger to delete .