This preview shows page 1. Sign up to view the full content.
Unformatted text preview: soid is c haracterized by its amplitude a nd angle (which includes its frequency and phase). In amplitude modulated signals, t he i nformation content of the
baseband (message) signal m (t) a ppears in t he a mplitude variations of the carrier.
In a ngle m odulation discussed in this section, t he i nformation content of m (t)
is c arried by t he angle of t he carrier. Angle modulation also goes by t he n ame
e xponential m odulation.
T he generalized angle modulated (or exponentially modulated) carrier can be
described as 'PEM(t) = A cos [wet o 200 3200 :r ( Hz) Fig. 4.39 Voice Spectrum. + k,p(t)] (4.80) where k is a n a rbitrary c onstant a nd ,p(t), which is a measure of m (t), is o btained
by a n invertible linear operation on m (t). I n other words, ,p(t) is t he o utput of
some linear system with a suitable transfer function H (s) w hen t he i nput is m (t),
as depicted in Fig. 4.40.t I f h (t) is t he u nit impulse response of this system: t hat
is, if h (t) { = } H (s), t hen Generation o f SSB Signals
T wo m ethods a re commonly used t o g enerate SSB signals. T he first method,
t he s electivefiltering m ethod uses s harp cutoff filters t o e liminate t he undesired
sideband, a nd t he second method uses phaseshifting networks 4 t o achieve t he same
goaJ.t We s hall consider here only t he first method.
.
T he selectivefiltering Method is t he m ost commonly used m ethod of generatmg
SSB signals. In this method, a DSBSC signal is passed through a s harp cutoff filter
to eliminate t he undesired sideband.
To o btain t he USB, t he filter should pass all components above W e u nattenuated
a nd c ompletely suppress all components below We' Such a n o peration requires an
ideal filter which is unrealizable. I t can, however, be realized closely if there is
some s epa;ation between t he p assband a nd t he s topband. Fortunately, t he voice
signal provides this condition, because its spectrum shows little power cont.ent a t
t he origin (Fig. 4.39). Moreover, articulation tests show t hat for speech Signals,
frequency components below 300 Hz are n ot i mportant. In other words, we may
suppress all speech components below 300 Hz without affecting t he intelligibility
tYet another method, known as Weaver's method, is also used to generate SSB signals. ~L._ _______) r~·~(t)~L._ __A_n_g_le_ _~t~_~E"(t)_ __
 I H ( S ~.
M odulator"
F ig. 4 .40 Generation of angle modulated signal. ,p(t) = loo m (o:)h(t  0:) do: (4.81) By selecting suitable h (t), we c an obtain a variety o f subclasses of angle modulation. For instance, if we select h (t) = u (t), t he r esulting form is t he wellknown
f requency m odulation ( FM). I n contrast, use of h (t) = o(t) leads to p hase
m odulation ( PM). These are b ut two of t he infinite possibilities. Although in
:j:Similarly, suppression of speechsignal components above 3500 Hz causes no appreciable change
in intelligibility.
tBecause H (s) is required to be invertible, we can obtain met) by...
View
Full
Document
This note was uploaded on 04/14/2013 for the course ENG 350 taught by Professor Bayliss during the Spring '13 term at Northwestern.
 Spring '13
 Bayliss
 Signal Processing, The Land

Click to edit the document details