(Amplitude modulation with pulse-train carrier)
(Amplitude modulation with pulse-train carrier)
Line 1: Line 1:
=== Amplitude modulation with pulse-train carrier ===
+
 
 +
== === Amplitude modulation with pulse-train carrier ===
  
 
y(t)=x(t)c(t) with c(t) be a pulse train
 
y(t)=x(t)c(t) with c(t) be a pulse train
Line 9: Line 10:
 
where <math>C(W)= \sum^{\infty}_{k = -\infty} a_k 2\pi\delta(\omega-\frac{2\pi}{T})</math>
 
where <math>C(W)= \sum^{\infty}_{k = -\infty} a_k 2\pi\delta(\omega-\frac{2\pi}{T})</math>
  
For k=0,<math>a_0 = \frac{\delta}{T} where \delta is the average of signal over 1 period.</math>
+
For k=0,<math>a_0 = \frac{\delta}{T} </math> where <math>\delta</math> is the average of signal over 1 period.
  
 
Else <math> a_k = \frac{1}{T} \int^{\frac{T}{2}}_{-\frac{T}{2}} c(t) e^{-jk\frac{T}{2}t} dt </math>
 
Else <math> a_k = \frac{1}{T} \int^{\frac{T}{2}}_{-\frac{T}{2}} c(t) e^{-jk\frac{T}{2}t} dt </math>
  
<math> a_k = \frac{1}{T} \int^{\frac{\delta}{2}}_{-\frac{\delta}{2}} e^{-jk\frac{T}{2}t} dt </math>
+
    <math> = \frac{1}{T} \int^{\frac{\delta}{2}}_{-\frac{\delta}{2}} e^{-jk\frac{T}{2}t} dt </math>
  
<math> a_k = \frac{2sin(k\frac{2\pi}{T}\frac{\delta}{2})}{TK\frac{T}{2}}</math>
+
    <math> = \frac{2sin(k\frac{2\pi}{T}\frac{\delta}{2})}{TK\frac{T}{2}}</math>
  
<math> a_k = \frac{sin(k\omega_0 \frac{\delta}{2})}{k\pi}</math>
+
    <math> = \frac{sin(k\omega_0 \frac{\delta}{2})}{k\pi}</math>
  
 
To recover x(t), use a low pass filter with gain <math> \frac {1}{a_0}=\frac {T}{\delta} </math>, and cut-off frequency <math> \omega_m < \omega_c < \omega_0 - \omega_m </math>
 
To recover x(t), use a low pass filter with gain <math> \frac {1}{a_0}=\frac {T}{\delta} </math>, and cut-off frequency <math> \omega_m < \omega_c < \omega_0 - \omega_m </math>

Revision as of 16:10, 16 November 2008

=== Amplitude modulation with pulse-train carrier =

y(t)=x(t)c(t) with c(t) be a pulse train

$ C(t)= \sum^{\infty}_{k = -\infty} a_k e^{jk\frac{2\pi}{T}t} $

Thus Y(W)= $ \frac{1}{2\pi}X(W)*C(W) $

where $ C(W)= \sum^{\infty}_{k = -\infty} a_k 2\pi\delta(\omega-\frac{2\pi}{T}) $

For k=0,$ a_0 = \frac{\delta}{T} $ where $ \delta $ is the average of signal over 1 period.

Else $ a_k = \frac{1}{T} \int^{\frac{T}{2}}_{-\frac{T}{2}} c(t) e^{-jk\frac{T}{2}t} dt $

   $  = \frac{1}{T} \int^{\frac{\delta}{2}}_{-\frac{\delta}{2}} e^{-jk\frac{T}{2}t} dt  $
   $  = \frac{2sin(k\frac{2\pi}{T}\frac{\delta}{2})}{TK\frac{T}{2}} $
   $  = \frac{sin(k\omega_0 \frac{\delta}{2})}{k\pi} $

To recover x(t), use a low pass filter with gain $ \frac {1}{a_0}=\frac {T}{\delta} $, and cut-off frequency $ \omega_m < \omega_c < \omega_0 - \omega_m $

Alumni Liaison

EISL lab graduate

Mu Qiao