Discrete Fourier Sequences
1. 离散傅里叶级数
可以计算周期序列的离散频率
设
x
(
n
)
x(n)
x(n)是长度为N的有限长序列,以N为周期对
x
(
n
)
x(n)
x(n)进行周期延拓得周期序列
x
~
(
n
)
\tilde{x}(n)
x~(n) ,
x
~
(
n
)
\tilde{x}(n)
x~(n)的离散傅里叶级数为
X
~
(
k
)
\tilde{X}(k)
X~(k)
DFS 变换对
D
F
S
:
X
~
(
k
)
=
D
F
S
[
x
~
(
n
)
]
=
∑
n
=
0
N
−
1
x
~
(
n
)
e
−
j
2
π
N
k
n
DFS: \ \tilde{X}(k)=DFS[\tilde{x}(n)] = \sum_{n=0}^{N-1}\tilde{x}(n)e^{-j\frac{2 \pi}{N}kn}
DFS: X~(k)=DFS[x~(n)]=n=0∑N−1x~(n)e−jN2πkn
I
D
F
S
:
x
~
(
n
)
=
I
D
F
S
[
X
~
(
k
)
]
=
1
N
∑
N
−
1
X
~
(
k
)
e
j
2
π
N
k
n
IDFS: \ \tilde{x}(n)=IDFS[\tilde{X}(k)] =\frac{1}{N} \sum_{}^{N-1}\tilde{X}(k)e^{j\frac{2 \pi}{N}kn}
IDFS: x~(n)=IDFS[X~(k)]=N1∑N−1X~(k)ejN2πkn
W
N
=
e
−
j
2
π
N
W_N = e^{-j\frac{2\pi}{N}}
WN=e−jN2π
性质 :共轭对称性:
W
N
n
=
(
W
N
−
n
)
∗
W_N^n=(W_N^{-n})^{*}
WNn=(WN−n)∗
表明将周期序列分解成N次谐波
- k k k 次谐波为 e j 2 π N k n e^{j\frac{2\pi}{N}kn} ejN2πkn 第 k k k 个谐波频率为 w k = 2 π N k , k = 0 , 1 , 2... N − 1 w_k = \frac{2\pi}{N}k, k=0,1,2...N-1 wk=N2πk,k=0,1,2...N−1, 幅度为 1 N X ~ ( n ) \frac{1}{N}\tilde{X}(n) N1X~(n) k = = 1 k==1 k==1时候是基波分量
主值序列 x ( n ) = x ~ ( k ) R N ( N ) x(n) = \tilde{x}(k)R_N(N) x(n)=x~(k)RN(N)
x ~ ( n ) = x ( ( n ) ) N \tilde{x}(n) = x((n))_N x~(n)=x((n))N 表示将 x ( n ) x(n) x(n) 以N为周期进行周期延拓
2. 与z变换的关系
z变换 X ( z ) = ∑ n = 0 N − 1 x ( n ) z − n X(z)=\sum_{n={0}}^{N-1}x(n)z^{-n} X(z)=n=0∑N−1x(n)z−n
DFS变换
X
~
(
k
)
=
∑
n
=
0
N
−
1
x
~
(
n
)
e
−
j
2
π
N
n
k
=
∑
n
=
0
N
−
1
x
(
n
)
[
e
j
2
π
N
k
]
−
n
\tilde{X}(k)=\sum_{n=0}^{N-1}\tilde{x}(n)e^{-j\frac{2\pi}{N}nk} = \sum_{n=0}^{N-1}x(n)[e^{j\frac{2\pi}{N}k}]^{-n}
X~(k)=n=0∑N−1x~(n)e−jN2πnk=n=0∑N−1x(n)[ejN2πk]−n
结论
X
~
(
k
)
=
X
(
z
)
∣
z
=
e
j
2
π
N
k
\tilde{X}(k) = X(z)|_{z=e^{j\frac{2\pi}{N}k}}
X~(k)=X(z)∣z=ejN2πk
周期序列
X
~
(
k
)
\tilde{X}(k)
X~(k)可以看成是 的一个周期
x
~
(
n
)
\tilde{x}(n)
x~(n)作 z变换,然后将
z
z
z变换在
z
z
z平面的单位圆上作等间隔
2
π
N
\frac{2\pi}{N}
N2π抽样得到的结果
3.与DTFT的关系
DTFT
X
(
e
j
w
)
=
∑
n
=
0
N
−
1
x
(
n
)
e
j
w
n
X(e^{jw}) = \sum_{n=0}^{N-1}x(n)e^{jwn}
X(ejw)=n=0∑N−1x(n)ejwn
DFS
X
~
(
k
)
=
∑
n
=
0
N
−
1
x
~
(
n
)
e
−
j
2
π
N
n
k
=
∑
n
=
0
N
−
1
x
(
n
)
[
e
−
j
2
π
N
k
]
n
\tilde{X}(k)=\sum_{n=0}^{N-1}\tilde{x}(n)e^{-j\frac{2\pi}{N}nk}=\sum_{n=0}^{N-1}x(n)[e^{-j\frac{2\pi}{N}k}]^n
X~(k)=n=0∑N−1x~(n)e−jN2πnk=n=0∑N−1x(n)[e−jN2πk]n
结论
X
~
(
k
)
=
X
(
e
j
w
)
∣
w
=
2
π
N
k
\tilde{X}(k)=X(e^{jw})|_{w=\frac{2\pi}{N}k}
X~(k)=X(ejw)∣w=N2πk
**DFS等价于对DTFT变换的结果以
w
1
=
2
π
N
w_1=\frac{2\pi}{N}
w1=N2π为间隔进行采样的结果
其中
w
1
w_1
w1为频率的采样间隔,也被称为频率分辨率
4.离散傅里叶级数的性质
-
线性
- D F S [ a ⋅ x ~ 1 ( n ) + b ⋅ x ~ 2 ( n ) ] = a ⋅ X ~ 1 ( k ) + b ⋅ X ~ 2 ( k ) DFS[a \cdot \tilde{x}_1(n) + b \cdot \tilde{x}_2(n)] = a \cdot \tilde{X}_1(k) + b \cdot \tilde{X}_2(k) DFS[a⋅x~1(n)+b⋅x~2(n)]=a⋅X~1(k)+b⋅X~2(k)
-
序列的移位
- D F S [ x ~ ( n + m ) ] = W N − k m X ~ ( k ) , I D F S [ X ~ ( k + l ) ] = W N l n x ~ ( n ) DFS[\tilde{x}(n+m)]=W_N^{-km}\tilde{X}(k), \ \ \ IDFS[\tilde{X}(k+l)] = W_N^{ln}\tilde{x}(n) DFS[x~(n+m)]=WN−kmX~(k), IDFS[X~(k+l)]=WNlnx~(n)
-
调制特性
- D F S [ W N n l x ~ ( n ) ] = X ~ ( k + l ) DFS[W_N^{nl}\tilde{x}(n)] = \tilde{X}(k+l) DFS[WNnlx~(n)]=X~(k+l)
-
共轭(反)对称分量
- x ~ e ( n ) = 1 2 [ x ~ ( n ) + x ~ ∗ ( − n ) ] = 1 2 [ x ( ( n ) ) N + x ∗ ( ( N − n ) ) N ] \tilde{x}_e(n) = \frac{1}{2}[\tilde{x}(n)+\tilde{x}^*(-n)] = \frac{1}{2}[x((n))_N + x^*((N-n))_N] x~e(n)=21[x~(n)+x~∗(−n)]=21[x((n))N+x∗((N−n))N] 反: x ~ o ( n ) = 1 2 [ x ~ ( n ) + x ~ ∗ ( − n ) ] = 1 2 [ x ( ( n ) ) N − x ∗ ( ( N − n ) ) N ] 反:\tilde{x}_o(n) = \frac{1}{2}[\tilde{x}(n)+\tilde{x}^*(-n)] = \frac{1}{2}[x((n))_N - x^*((N-n))_N] 反:x~o(n)=21[x~(n)+x~∗(−n)]=21[x((n))N−x∗((N−n))N] x ~ e ( n ) = x ~ e ∗ ( − n ) , x ~ o ( n ) = − x ~ o ∗ ( − n ) , x ~ ( n ) = x ~ e ( n ) + x ~ o ( n ) \tilde{x}_e(n) = \tilde{x}_e^*(-n),\ \ \tilde{x}_o(n) = -\tilde{x}_o^*(-n),\ \ \tilde{x}(n)=\tilde{x}_e(n)+\tilde{x}_o(n) x~e(n)=x~e∗(−n), x~o(n)=−x~o∗(−n), x~(n)=x~e(n)+x~o(n)
-
周期卷积和
- 如果 Y ~ ( k ) = X 1 ~ ( k ) ⋅ X 2 ~ ( k ) \tilde{Y}(k) = \tilde{X_1}(k) \cdot \tilde{X_2}(k) Y~(k)=X1~(k)⋅X2~(k) 则有 y ~ ( n ) = I D F S [ Y ~ ( k ) ] = ∑ m = 0 N − 1 x ~ 1 ( m ) x ~ 2 ( n − m ) = ∑ m = 0 N − 1 x ~ 2 ( m ) x ~ 1 ( n − m ) \tilde{y}(n)=IDFS[\tilde{Y}(k)] = \sum_{m=0}^{N-1}\tilde{x}_1(m)\tilde{x}_2(n-m) =\sum_{m=0}^{N-1}\tilde{x}_2(m)\tilde{x}_1(n-m) y~(n)=IDFS[Y~(k)]=m=0∑N−1x~1(m)x~2(n−m)=m=0∑N−1x~2(m)x~1(n−m)
-