语音信号预处理3——计算原始chirp信号与滤波后chirp信号的相关性

article/2025/10/4 0:16:40

写在前面：仅作为个人学习笔记，便于以后查阅，如对你有帮助，荣幸之至，如有错误，欢迎评论指正

编程软件：matlab R2018b
系统： win10

系列文章目录

chirp信号的生成与接收
对接收到的chirp信号进行滤波、分帧处理及端点检测
计算原始chirp信号与滤波后chirp信号的相关性
绘制接滤波后chirp信号的时域图和频域图
从1-4完整走一遍

文章目录

- - 系列文章目录
  - 相关性计算
  - 参考文献

相关性计算

本案例计算发射的信号与接收到的回声的相关性，以此判断回声信号的哪些部分包含了我们的发射信号，换言之，我们采集的回声信号里除了环境杂音，还有我们自己发出的声音信号。
相关性计算的具体代码如下corr_compute.m：

function corrSeq = corr_compute(data1, sampleSignal) %%%data initilization
sizeOfSample = length(sampleSignal); 
sizeOfData = length(data1); 
output = zeros(sizeOfData, 1); %%%Compute the correlation 
for i = 1 :1 : sizeOfData-sizeOfSampleoutput(i,1) = abs(dot(sampleSignal', data1(i:i+sizeOfSample-1, 1)));   
end%%%Return the final values
corrSeq = output; % figure
% t1 = 1:length( output );
% %t2 =  t1/samplingRate;
% plot(t1, output,'LineWidth',2,'Color','b');
% hold on;%%%Conduct the envelop detection 
window_size_for_envelope  = 30;  %%%The length of each frame%%%Count the number of windows we need 
counter=0;
for i=1:(window_size_for_envelope - 1) : (length(corrSeq) - window_size_for_envelope)counter=counter+1;
end
envelope_m=zeros(counter,1);%%%COmpute the values for each frame
counter=1;
for i=1:( window_size_for_envelope - 1 ) : (length(corrSeq) - window_size_for_envelope)current_window = corrSeq(i:i+window_size_for_envelope-1,:);min_value=min(current_window);max_value=max(current_window);envelope_m(counter,1) = max_value; %%upper part% envelope_m(counter,1) = min_value; %%lower partcounter=counter+1;
end%%%Conduct the interpolation 
x=1:( window_size_for_envelope - 1 ):(length(corrSeq) - window_size_for_envelope);  %%%The position of the sampled points
y=envelope_m;  %%%The value of sampled points
x_i=1:(length(corrSeq) - window_size_for_envelope); %%%The vector size after interpolationy_i = interp1(x,y,x_i,'spline');%%%Return the results
corrSeq = y_i;