傅立叶级数与变换

参考：DR_CAN

相同领域的文章：

• 动态系统的建模与分析
• 自动控制原理
• Advanced控制理论
• 傅里叶级数与变换
• 工程数学

1.三角函数的正交性

三角函数系是一个集合：
{ 1 , sin ⁡ x , cos ⁡ x , sin ⁡ 2 x , cos ⁡ 2 x , ⋯   , sin ⁡ n x , cos ⁡ n x , ⋯   } \left \{ 1,\sin x, \cos x, \sin 2x,\cos 2x,\cdots,\sin nx, \cos nx,\cdots \right \} {1,sinx,cosx,sin2x,cos2x,⋯,sinnx,cosnx,⋯}

三角函数系更一般的形式：
{ sin ⁡ 0 x , cos ⁡ 0 x , sin ⁡ x , cos ⁡ x , sin ⁡ 2 x , cos ⁡ 2 x , ⋯   , sin ⁡ n x , cos ⁡ n x , ⋯   } \left \{ \sin 0x, \cos 0x,\sin x, \cos x, \sin 2x,\cos 2x,\cdots,\sin nx, \cos nx,\cdots \right \} {sin0x,cos0x,sinx,cosx,sin2x,cos2x,⋯,sinnx,cosnx,⋯}

集合中任意两个三角函数乘积的积分为0，即正交：
$${\int }_{-\pi }^{\pi }\sin nx\cos mxdx=0$$
$${\int }_{-\pi }^{\pi }\cos nx\cos mxdx=\{\begin{array}{lc}0& m\ne n\\ \pi & m=n\ne 0\\ 2\pi & m=n=0\end{array}$$
$${\int }_{-\pi }^{\pi }\sin nx\sin mxdx=\{\begin{array}{lc}0& m\ne n\\ \pi & m=n\ne 0\end{array}$$

---

证明过程：
$\vec{a}\cdot \vec{b}=(2,1)\cdot (-1,2)=0$

$$\begin{gathered} \vec{a}=(a_1,a_2,\cdots ,a_n) \\ \vec{b}=(b_1,b_2,\cdots ,b_n) \\ \vec{a}\cdot \vec{b}=a_1{b}_1+a_2{b}_2+\cdots +a_n{b}_n={\sum }_{i=1}^n{a}_i{b}_i \end{gathered}$$
离散为求和，连续为积分

写成积分的形式，向量变为函数
$$\begin{gathered} a=f(x),b=g(x) \\ a\cdot b={\int }_{x_0}^{x_1}f(x)g(x)dx=0 \end{gathered}$$

例证：
$${\int }_{-\pi }^{\pi }\cos 0x\sin xdx={\int }_{-\pi }^{\pi }\sin xdx=0$$
$\sin x是奇函数，因此对称区间面积抵消$

$$\begin{array}{rl}& {\int }_{-\pi }^{\pi }\cos nx\cos mx\\ & =\frac{1}{2}\left[{\int }_{-\pi }^{\pi }\cos (n-m)x+{\int }_{-\pi }^{\pi }\cos (n+m)x\right]\\ & =\frac{1}{2}\left[{\frac{1}{n-m}\sin (n-m)x|}_{-\pi }^{\pi }+{\frac{1}{n+m}\sin (n+m)x|}_{-\pi }^{\pi }\right]\\ & =0\end{array}$$

$$\begin{gathered} {\int }_{-\pi }^{\pi }\cos nx\sin mxdx=0,n\ne m \\ {\int }_{-\pi }^{\pi }\sin nx\sin mxdx=0,n\ne m \end{gathered}$$

如果n=m
$$\begin{array}{rl}& {\int }_{-\pi }^{\pi }\cos nx\cos mxdx\\ & =\frac{1}{2}{\int }_{-\pi }^{\pi }1dx+{\int }_{-\pi }^{\pi }\cos 2mxdx\\ & =\frac{1}{2}{\int }_{-\pi }^{\pi }1dx=\pi \end{array}$$

2.周期为“$2\pi$”的函数展开为傅立叶级数

$T=2\pi ,f(x)=f(x+2\pi )$

傅立叶级数展开式
$$\begin{array}{rl}f(x)& =\sum _{n=0}^{\infty }{a}_n\cos nx+\sum _{n=0}^{\infty }{b}_n\sin nx\\ & =a_0\cos 0x+\sum _{n=1}^{\infty }{a}_n\cos nx+b_0\sin 0x+\sum _{n=1}^{\infty }{b}_n\sin nx\\ & =\frac{a_0}{2}+\sum _{n=1}^{\infty }{a}_n\cos nx+\sum _{n=1}^{\infty }{b}_n\sin nx\end{array}$$

确定$a_0$
$$\begin{gathered} {\int }_{-\pi }^{\pi }f(x)dx={\int }_{-\pi }^{\pi }{a}_{0}dx=2\pi a_0 \\ {a}_0=\frac{1}{\pi }{\int }_{-\pi }^{\pi }f(x)dx \end{gathered}$$

确定$a_n$
$$\begin{gathered} {\int }_{-\pi }^{\pi }f(x)\cos mxdx={\int }_{-\pi }^{\pi }\sum _{\infty }^{n=1}{a}_n\cos nx\cos mxdx=\pi a_n \\ {a}_n=\frac{1}{\pi }{\int }_{-\pi }^{\pi }f(x)\cos nxdx \end{gathered}$$

确定b_n
同理可得
$$b_n=\frac{1}{\pi }{\int }_{-\pi }^{\pi }f(x)\sin nxdx$$

$$\begin{array}{rl}f(x)& =f(x+2\pi )T=2\pi \\ f(x)& =\frac{a_0}{2}+\sum _{n=1}^{\infty }{a}_n\cos nx+\sum _{n=1}^{\infty }{b}_n\sin nx\\ a_0& =\frac{1}{\pi }{\int }_{-\pi }^{\pi }f(x)dx\\ a_n& =\frac{1}{\pi }{\int }_{-\pi }^{\pi }f(x)\cos nxdx\\ b_n& =\frac{1}{\pi }{\int }_{-\pi }^{\pi }f(x)\sin nxdx\end{array}$$

3.周期为“2L”的函数展开为傅立叶级数

$f(t)=f(t+2L)$
换元
$$\begin{gathered} x=\frac{\pi }{L}t \\ t=\frac{L}{\pi }x \end{gathered}$$

$t$	$x$
0	0
2L	2$\pi$
4L	$4\pi$
$⋮$	$⋮$

用新的函数表示原函数
$f(t)=f(\frac{L}{\pi }x)=g(x)$

$$\begin{array}{rl}g(x)& =\frac{a_0}{2}+\sum _{n=1}^{\infty }{a}_n\cos nx+\sum _{n=1}^{\infty }{b}_n\sin nx\\ a_0& =\frac{1}{\pi }{\int }_{-\pi }^{\pi }g(x)dx\\ a_n& =\frac{1}{\pi }{\int }_{-\pi }^{\pi }g(x)\cos nxdx\\ b_n& =\frac{1}{\pi }{\int }_{-\pi }^{\pi }g(x)\sin nxdx\end{array}$$

$$\begin{gathered} x=\frac{\pi }{L}t \\ \cos nx=\cos \frac{n\pi }{L}t \\ \sin nx=\sin \frac{n\pi }{L}t \\ g(x)=f(t) \\ \frac{1}{\pi }{\int }_{-\pi }^{\pi }dx=\frac{1}{\pi }{\int }_{-L}^{L}d\frac{\pi }{L}t=\frac{1}{L}{\int }_{-L}^{L}dt \end{gathered}$$

$$\begin{array}{rl}f(t)& =\frac{a_0}{2}+\sum _{n=1}^{\infty }{a}_n\cos \frac{n\pi }{L}t+\sum _{n=1}^{\infty }{b}_n\sin \frac{n\pi }{L}t\\ a_0& =\frac{1}{L}{\int }_{-L}^{L}f(t)dt\\ a_n& =\frac{1}{L}{\int }_{-L}^{L}f(t)\cos \frac{n\pi }{L}tdt\\ b_n& =\frac{1}{L}{\int }_{-L}^{L}f(t)\sin \frac{n\pi }{L}tdt\end{array}$$

工程中t从0开始，周期为T=2L，$\omega =\frac{\pi }{L}=\frac{2\pi }{T}$
${\int }_{-L}^{L}dt\to {\int }_0^{2L}dt\to {\int }_0^{T}dt$

$$\begin{array}{rl}f(t)& =\frac{a_0}{2}+\sum _{n=1}^{\infty }{a}_n\cos nwt+\sum _{n=1}^{\infty }{b}_n\sin nwt\\ a_0& =\frac{2}{T}{\int }_0^{T}f(t)dt\\ a_n& =\frac{2}{T}{\int }_0^{T}f(t)\cos nwtdt\\ b_n& =\frac{2}{T}{\int }_0^{T}f(t)\sin nwtdt\end{array}$$

4.傅立叶级数的复数形式

欧拉公式
$$\begin{gathered} e^{i\theta }=\cos \theta +i\sin \theta \\ \cos \theta =\frac{1}{2}(e^{i\theta }+e^{-i\theta }) \\ \sin \theta =-\frac{1}{2}i(e^{i\theta }-e^{-i\theta }) \end{gathered}$$

$$\begin{array}{rl}f(t)& =\frac{a_0}{2}+\sum _{n=1}^{\infty }\left[a_n\frac{1}{2}(e^{inwt}+e^{-inwt})-b_n\frac{1}{2}(e^{inwt}+e^{-inwt})\right]\\ & =\frac{a_0}{2}+\sum _{n=1}^{\infty }\left[\frac{a_n-i{b}_n}{2}{e}^{inwt}+\frac{a_n+i{b}_n}{2}{e}^{-inwt}\right]\\ & =\frac{a_0}{2}+\sum _{n=1}^{\infty }\frac{a_n-i{b}_n}{2}{e}^{inwt}+\sum _{n=1}^{\infty }\frac{a_n+i{b}_n}{2}{e}^{-inwt}\\ & =\sum _{n=0}^0\frac{a_0}{2}{e}^{inwt}+\sum _{n=1}^{\infty }\frac{a_n-i{b}_n}{2}{e}^{inwt}+\sum _{n=-\infty }^{-1}\frac{a_n+i{b}_n}{2}{e}^{-inwt}\\ & =\sum _{-\infty }^{\infty }{C}_n{e}^{inwt}\end{array}$$

$$C_n=\{\begin{array}{lc}\frac{a_0}{2}& ,n=0\\ \frac{a_n-i{b}_n}{2}& ,n=1,2,3,\cdots \\ \frac{a_n+i{b}_n}{2}& ,n=-1,-2,-3,\cdots \end{array}$$

n=0
$$C_n=\frac{a_0}{2}=\frac{1}{2}\cdot \frac{2}{T}{\int }_0^{T}f(t)dt=\frac{1}{T}{\int }_0^{T}f(t)e^{0}dt$$

n=1,2,3,$\cdots$
$$\begin{array}{rl}{C}_n& =\frac{1}{2}\left(\frac{2}{T}{\int }_0^{T}f(t)\cos nwtdt-i\frac{2}{T}{\int }_0^{T}f(t)\sin nwtdt\right)\\ & =\frac{1}{T}{\int }_0^{T}f(t)(\cos nwt-i\sin nwt)dt\\ & =\frac{1}{T}{\int }_0^{T}f(t)(\cos nwt+i\sin (-nwt))dt\\ & =\frac{1}{T}{\int }_0^{T}f(t)e^{-inwt}dt\end{array}$$

n=-1,-2,-3,$\cdots$
$$\begin{array}{rl}{C}_n& =\frac{1}{2}\left(\frac{2}{T}{\int }_0^{T}f(t)\cos (-n)wtdt+i\frac{1}{T}{\int }_0^{T}f(t)\sin (-n)wtdt\right)\\ & =\frac{1}{T}{\int }_0^{T}f(t)e^{-inwt}dt\end{array}$$

$$\begin{array}{rl}& f(t)=f(t+T)\\ & f(t)=\sum _{-\infty }^{\infty }{C}_n{e}^{inwt}\\ & C_n=\frac{1}{T}{\int }_0^{T}f(t)e^{-inwt}dt\end{array}$$

5.傅立叶变换

$$\begin{array}{rl}& f(t)=f(t+T)\\ & f_T(t)=\sum _{-\infty }^{\infty }{C}_n{e}^{in{w}_{0}t}\\ & C_n=\frac{1}{T}{\int }_{T/2}^{T/2}{f}_T(t)e^{-in{w}_{0}t}dt\\ & w_0=\frac{2\pi }{T}为基频率\end{array}$$

在工程中，信号一般都是非周期信号，即$T\to \infty$
$$\begin{gathered} \underset{T\to \infty }{\lim }{f}_T(t)=f(t) \\ \Delta w=(n+1)w_0+n{w}_0=w_0=\frac{2\pi }{T} \\ \frac{1}{T}=\frac{\Delta }{2\pi } \end{gathered}$$
当$T\to \infty$时 ，$\Delta w\to 0$，离散$\to$ 连续

则有$$\begin{gathered} {\int }_{T/2}^{T/2}dt\to {\int }_{-\infty }^{\infty }dt \\ n{w}_0\to w \\ \sum _{n=-\infty }^{infty}\Delta w\to {\int }_{-\infty }^{\infty } \end{gathered}$$

$$f(t)=\frac{1}{2\pi }{\int }_{-\infty }^{\infty }{\int }_{-\infty }^{\infty }f(t)e^{-iwt}dt{e}^{iwt}dw$$
其中$$F(w)={\int }_{-\infty }^{\infty }f(t)e^{-iwt}dt$$为傅立叶变换，
$$f(t)=\frac{1}{2\pi }{\int }_{-\infty }^{\infty }F(w)e^{iwt}dw$$为傅里叶逆变换。