我们可以将一个载波调制信号写成如下形式:
x
c
(
t
)
=
y
(
t
)
c
o
s
[
2
π
f
c
t
+
ϕ
(
t
)
]
x_c(t)=y(t)cos[2\pi f_ct+\phi(t)]
xc(t)=y(t)cos[2πfct+ϕ(t)]
通过欧拉公式:
e
i
x
=
c
o
s
x
+
i
s
i
n
x
e^{ix}=cosx+isinx
eix=cosx+isinx
c
o
s
x
=
1
2
[
e
i
x
+
e
−
i
x
]
cosx=\dfrac{1}{2}[e^{ix}+e^{-ix}]
cosx=21[eix+e−ix]
s
i
n
x
=
1
2
i
[
e
i
x
−
e
−
i
x
]
sinx=\dfrac{1}{2i}[e^{ix}-e^{-ix}]
sinx=2i1[eix−e−ix]
我们可以把载波调制信号写成:
x
c
(
t
)
=
1
2
[
y
(
t
)
e
i
[
2
π
f
c
t
+
ϕ
(
t
)
]
+
y
(
t
)
e
−
i
[
2
π
f
c
t
+
ϕ
(
t
)
]
]
x_c(t)=\dfrac{1}{2}[y(t)e^{i[2\pi f_ct+\phi(t)]}+y(t)e^{-i[2\pi f_ct+\phi(t)]}]
xc(t)=21[y(t)ei[2πfct+ϕ(t)]+y(t)e−i[2πfct+ϕ(t)]]
把载波调制信号写成复数形式:
x
c
(
t
)
˚
=
y
(
t
)
e
i
[
2
π
f
c
t
+
ϕ
(
t
)
]
\mathring{x_c(t)}=y(t)e^{i[2\pi f_ct+\phi(t)]}
xc(t)˚=y(t)ei[2πfct+ϕ(t)]
载波调制信号的复数共轭形式就可以写成:
x
c
(
t
)
˚
‾
=
y
(
t
)
e
−
i
[
2
π
f
c
t
+
ϕ
(
t
)
]
\overline{\mathring{x_c(t)}}=y(t)e^{-i[2\pi f_ct+\phi(t)]}
xc(t)˚=y(t)e−i[2πfct+ϕ(t)]
所以,可以把载波调制信号的实数形式写成:
x
c
(
t
)
=
1
2
(
x
c
(
t
)
˚
+
x
c
(
t
)
˚
‾
)
x_c(t)=\dfrac{1}{2}(\mathring{x_c(t)}+\overline{\mathring{x_c(t)}})
xc(t)=21(xc(t)˚+xc(t)˚)
傅立叶变换
F
{
x
c
(
t
)
˚
}
=
X
c
(
w
)
˚
F\{\mathring{x_c(t)}\}=\mathring{X_c(w)}
F{xc(t)˚}=Xc(w)˚
F
{
x
c
(
t
)
˚
‾
}
=
X
c
(
−
w
)
˚
‾
F\{\overline{\mathring{x_c(t)}}\}=\overline{\mathring{X_c(-w)}}
F{xc(t)˚}=Xc(−w)˚
这个也比较好理解,比如:
对于10hz的复数信号:
e
i
(
2
π
×
10
t
+
ϕ
)
e^{i(2\pi \times10t+\phi)}
ei(2π×10t+ϕ)
当它乘以
e
−
i
(
2
π
×
10
t
)
e^{-i(2\pi \times10t)}
e−i(2π×10t)时,乘积为1,一个周期的积分结果为T,所以它的傅立叶变换就是10hz上的冲击函数,相位为
e
i
ϕ
e^{i\phi}
eiϕ
对于10hz的复数共轭信号:
e
−
i
(
2
π
×
10
t
+
ϕ
)
e^{-i(2\pi \times10t+\phi)}
e−i(2π×10t+ϕ)
当它乘以
e
i
(
2
π
×
10
t
)
e^{i(2\pi \times10t)}
ei(2π×10t)时,乘积为1,一个周期的积分结果为T,所以它的傅立叶变换就是-10hz上的冲击函数,相位为
e
−
i
ϕ
e^{-i\phi}
e−iϕ
所以
X
c
(
w
)
=
F
{
x
c
(
t
)
}
=
1
2
[
X
c
(
w
)
˚
+
X
c
(
w
)
˚
‾
]
X_c(w)=F\{x_c(t)\}=\dfrac{1}{2}[\mathring{X_c(w)}+\overline{\mathring{X_c(w)}}]
Xc(w)=F{xc(t)}=21[Xc(w)˚+Xc(w)˚]
上式说明,调制信号的傅立叶变换是两个部分的求和,一个部分是正频率
X
c
(
w
)
˚
\mathring{X_c(w)}
Xc(w)˚,一个部分是负频率
X
c
(
w
)
˚
‾
\overline{\mathring{X_c(w)}}
Xc(w)˚。我们因此可以发现,
X
c
(
w
)
˚
\mathring{X_c(w)}
Xc(w)˚可以通过
X
c
(
w
)
X_c(w)
Xc(w)获得。我们只需要提取
X
c
(
w
)
X_c(w)
Xc(w)的正频率部分,即。
X
c
(
w
)
˚
=
2
u
(
w
)
X
c
(
w
)
\mathring{X_c(w)}=2u(w)X_c(w)
Xc(w)˚=2u(w)Xc(w)
那么,如果我们获得了调制信号的傅立叶变换之后,如何将其解调呢?
这里引入包络的概念:
x
c
(
t
)
~
=
y
(
t
)
e
i
ϕ
(
t
)
=
y
(
t
)
(
c
o
s
(
ϕ
(
t
)
)
+
i
s
i
n
(
ϕ
(
t
)
)
)
\widetilde{x_c(t)}=y(t)e^{i\phi(t)}=y(t)(cos(\phi(t))+isin(\phi(t)))
xc(t)
=y(t)eiϕ(t)=y(t)(cos(ϕ(t))+isin(ϕ(t)))
x
c
(
t
)
~
=
x
c
(
t
)
˚
e
−
i
w
c
t
\widetilde{x_c(t)}=\mathring{x_c(t)}e^{-iw_ct}
xc(t)
=xc(t)˚e−iwct
F
{
x
c
(
t
)
~
}
=
F
{
x
c
(
t
)
˚
e
−
i
w
c
t
}
=
∫
−
∞
∞
x
c
(
t
)
˚
e
−
i
w
c
t
e
−
i
w
t
d
t
=
∫
−
∞
∞
x
c
(
t
)
˚
e
−
i
(
w
c
+
w
)
t
d
t
=
X
˚
(
w
+
w
c
)
F\{\widetilde{x_c(t)}\}=F\{\mathring{x_c(t)}e^{-iw_ct}\}=\int_{-\infty}^{\infty}\mathring{x_c(t)}e^{-iw_ct}e^{-iwt}dt=\int_{-\infty}^{\infty}\mathring{x_c(t)}e^{-i(w_c+w)t}dt=\mathring{X}(w+w_c)
F{xc(t)
}=F{xc(t)˚e−iwct}=∫−∞∞xc(t)˚e−iwcte−iwtdt=∫−∞∞xc(t)˚e−i(wc+w)tdt=X˚(w+wc)
所以
F
{
x
c
(
t
)
~
}
F\{\widetilde{x_c(t)}\}
F{xc(t)
}相当于将
X
˚
(
w
)
\mathring{X}(w)
X˚(w)向左平移
w
c
w_c
wc。