微信公众号：EW Frontier

关注可了解更多的雷达、通信、人工智能相关代码。问题或建议，请公众号留言;

如果你觉得EW Frontier对你有帮助，欢迎加入我的知识星球或面包多，更多代码等你来学

知识星球：https://wx.zsxq.com/dweb2/index/group/15552518881412

面包多：https://mbd.pub/o/author-a2mYl2tsbA==/work

QQ交流群：949444104

常见的通信干扰信号[12][13]主要分为两类：欺骗式干扰和压制式干扰，压制式干扰依据干扰带宽的选择又可分为瞄准式干扰和阻塞式干扰，像单音、多音和窄带噪声干扰属于瞄准式干扰，而宽带噪声干扰、扫频式干扰和梳状谱干扰则属于阻塞式干扰。本文主要针对这6类干扰信号进行识别研究。

单音干扰

单音干扰是针对目标信号单一频点进行干扰，干扰信号能量集中于固定频率，其表达式如下：

多音干扰

多音干扰是针对多个目标频率进行干扰，干扰信号能量相对于单音干扰分散，表现形式为多个单音干扰叠加，其表达式如下：

宽带噪声干扰

宽带噪声干扰是将高斯白噪声通过一个宽带滤波器，在每一个宽带范围内产生噪声。

窄带噪声干扰

窄带噪声干扰是干扰信号能量主要集中在一段频带，表现形式为高斯白噪声，其表达式如下：

梳状谱干扰

梳状谱干扰信号频谱与多音干扰相似，但是其谱峰为等间隔排列，且每个谱峰存在一定带宽，其表达式如下

扫频干扰

扫频干扰是扫频干扰中常见形式，干扰信号频率在一定带宽内逐渐改变，其表达式如下

MATLAB代码

% produce a fragment of single sinusoid jam
clear;close all;
Pl = 1;
Fs = 1000;                    % Sampling frequency
T = 1/Fs;                     % Sample time
L = 1000;                     % Length of signal
t = (0:L-1)*T;                % Time vector

% produce a fragment of multiple sinusoid jam
clear;close all;jnr = 200; % jam-noise ratio
Q = 10; % number of multiple sinusoid
Pl = 1-0.1*rand(1,Q);                       % Jam power
​
Fs = 2000;                    % Sampling frequency
T = 1/Fs;                     % Sample time
L = 2*Fs;                     % Length of signal
t = (0:L-1)*T;                % Time vector
% Sum of a 50 Hz sinusoid
fs0 = 100;deltafs = 10;
fs = fs0:deltafs:fs0+(Q-1)*deltafs; % produce same-distance fs
theta = 2*pi*rand(1,Q);% random phase

% produce a fragment of narrow band noise 
clear;close all;jnr = 20; % jam-noise ratio
Fs = 2000;                    % Sampling frequency
T = 1/Fs;                     % Sample time
L = 2*Fs;                     % Length of signal
t = (0:L-1)*T;                % Time vector
P=1;% power of noise
y = wgn(L,1,P); % white noise
% 

​
​

% produce a fragment of broad band noise 
clear;close all;
Fs = 2000;                    % Sampling frequency
T = 1/Fs;                     % Sample time
L = 2*Fs;                     % Length of signal
t = (0:L-1)*T;                % Time vector
P=1;% power of noise
y = wgn(L,1,P/600); % white noise
% 
% y = awgn(x,jnr,'measured');    % Sinusoids plus noise

% produce a fragment of comb-like noise 
clear;close all;
Fs = 2000;                    % Sampling frequency
T = 1/Fs;                     % Sample time
L = 2*Fs;                     % Length of signal
t = (0:L-1)*T;                % Time vector
P=1;% power of noise
y = wgn(L,1,P); % white noise
% 

% produce a fragment of scanning frequency jam
clear;close all;jnr = 1000; % jam-noise ratio
Fs = 2000;                    % Sampling frequency
T = 1/Fs;                     % Sample time
L = 2*Fs;                     % Length of signal 2Second
t = (0:L-1)*T;                % Time vector
om = 2*pi*10;
be = 2*pi*50;

完

相关学习资料见面包多链接https://mbd.pub/o/author-a2mYl2tsbA==/work。

欢迎加入我的知识星球：https://wx.zsxq.com/dweb2/index/group/15552518881412，永久获取更多相关资料、代码。