當我直接去看chirp的MATLAB幫助文件時，始終不得要領，查看了很多博文上的說法，也還是不懂，直到我去查看了維基百科，並總結了下面這篇博文後，反過來看chirp的MATLAB幫助文件，才覺得明朗了一些。

因此，推薦看看上篇這篇博文，先從基礎上了解下chirp訊號。

MATLAB 中稱 chirp 為 Swept-frequency cosine，也即掃頻餘弦波。

MATLAB 中給出了 chirp 如下的語法結構：

下面一一簡單介紹：

y = chirp(t,f0,t1,f1) 在陣列t中定義的時間例項中產生線性掃頻餘弦訊號的樣本，f0為時間為0時刻的瞬時頻率，f1為時刻t1時的瞬時頻率，二者單位都是赫茲。如果未指定，則對於logarithmic chirp，f0為10^-6;對於其他的chirp，f0都為0,t1為1，f1為100.

y = chirp(t,f0,t1,f1,'method') 指定可選擇的掃頻方法，掃頻方法介紹如下：

如果看了上篇博文，這點介紹應該是能看懂的（儘管有些差異，差異在於叫法以及少了下面的第二種情況），我就不翻譯了，翻譯也許就沒那麼精妙了。

總結起來，掃頻方法分為三種，分別為linear、quadratic、logarithmic。

y = chirp(t,f0,t1,f1,'method',phi) allows an initial phase phi to be specified in degrees. If unspecified, phi is 0. Default values are substituted for empty or omitted trailing input arguments.

這種形式指定了初始相位，如果未指定初始相位為0。

y = chirp(t,f0,t1,f1,'quadratic',phi,'shape') specifies the shape of the quadratic swept-frequency signal's spectrogram. shape is either concave or convex, which describes the shape of the parabola in the positive frequency axis. If shape is omitted, the default is convex for downsweep (f

y = chirp（t，f0，t1，f1，'quadratic'，phi，'shape'）指定二次掃頻訊號的頻譜圖的形狀。 形狀是凹形或凸形，它描述了正頻率軸上拋物線的形狀。 如果省略形狀，則預設為下掃頻（f0> f1）為凸，而上掃頻（f0 <f1）為凹。

下面舉一個例子：

Linear chirp

Generate a chirp with linear instantaneous frequency deviation. The chirp is sampled at 1 kHz for 2 seconds. The instantaneous frequency is 0 at t = 0 and crosses 250 Hz at t = 1 second.

產生具有線性瞬時頻率偏差的chirp。取樣率為1KHz，時間為2s，初始頻率為0，時間為1時刻的頻率為250Hz。

% Generate a chirp with linear instantaneous frequency deviation. 
% The chirp is sampled at 1 kHz for 2 seconds. 
% The instantaneous frequency is 0 at t = 0 and crosses 250 Hz at t = 1 second.

clc
clear
close all


t = 0:1/1e3:2;
y = chirp(t,0,1,250);
subplot(2,1,1)
plot(t,y)
subplot(2,1,2)
spectrogram(y,256,250,256,1e3,'yaxis')

Quadratic chirp

Generate a chirp with quadratic instantaneous frequency deviation. The chirp is sampled at 1 kHz for 2 seconds. The instantaneous frequency is 100 Hz at t = 0 and crosses 200 Hz at t = 1 second.

% Generate a chirp with quadratic instantaneous frequency deviation. 
% The chirp is sampled at 1 kHz for 2 seconds. 
% The instantaneous frequency is 100 Hz at t = 0 and crosses 200 Hz at t = 1 second.

clc
clear
close all


t = 0:1/1e3:2;
y = chirp(t,100,1,200,'quadratic ');
subplot(2,1,1)
plot(t,y)


% Compute and plot the spectrogram of the chirp.
% Specify 128 DFT points, a Hamming window of the same length, and 120 samples of overlap.

subplot(2,1,2)
spectrogram(y,256,250,256,1e3,'yaxis')

中間流了兩種情況，等我弄明白了再填。

那個spectrogram我會單獨寫一篇部落格，弄懂它。暫時還不是太明白原理。

Logarithmic Chirp

% Generate a logarithmic chirp sampled at 1 kHz for 10 seconds. 
% The instantaneous frequency is 10 Hz initially and 400 Hz at the end.



clc
clear
close all


t = 0:1/1e3:10;
fo = 10;
f1 = 400;
y = chirp(t,fo,10,f1,'logarithmic');
subplot(2,1,1)
plot(t,y)

% Compute and plot the spectrogram of the chirp.
% Specify 256 DFT points, a Hamming window of the same length, and 200 samples of overlap.

subplot(2,1,2)
spectrogram(y,256,200,256,1e3,'yaxis')