0% found this document useful (0 votes)
61 views

AM Notes

- SSB modulation transmits only one sideband (USB or LSB) to reduce bandwidth usage, since the sidebands contain redundant information. - SSB signals can be represented in the time domain using the baseband signal m(t) and its Hilbert transform mh(t). A general SSB signal is sSSB(t) = m(t)cos(ωct) ∓ mh(t)sin(ωct). - SSB signals can be generated using selective filtering of a DSB-SC signal or using a phase-shift method. Demodulation involves multiplying with a carrier and lowpass filtering to recover m(t).

Uploaded by

Adarsh Damodaran
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
61 views

AM Notes

- SSB modulation transmits only one sideband (USB or LSB) to reduce bandwidth usage, since the sidebands contain redundant information. - SSB signals can be represented in the time domain using the baseband signal m(t) and its Hilbert transform mh(t). A general SSB signal is sSSB(t) = m(t)cos(ωct) ∓ mh(t)sin(ωct). - SSB signals can be generated using selective filtering of a DSB-SC signal or using a phase-shift method. Demodulation involves multiplying with a carrier and lowpass filtering to recover m(t).

Uploaded by

Adarsh Damodaran
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

• s(t) > Vc :

– diode conducts (forward biased)


– C charges quickly to the max value of s(t)

• s(t) < Vc :
– diode does not conduct (reverse biased)
– C discharges over RL until s(t) > Vc

• The output of the ED is then lowpass filtered to eliminate the ripple, followed by
blocking out the DC component.

29

Single Sideband Modulation (SSB)

Standard AM and DSB-SC techniques are wasteful of bandwidth because they both require
transmission bandwidth of 2B Hz, where B is the bandwidth of the baseband modulating
signal m(t).
In both cases the transmission bandwidth (BT ) is occupied by the upper sideband (USB) and
lower sideband (LSB).
Observations

• USB and LSB are uniquely related to each other, as they are symmetric wrt fc .
Therefore, to transmit information contained within m(t) we used to transmit only one
side band.

• As far as demodulation is concerned, we can coherently demodulate SSB (as we did the
DSB-SC signal) by multiplying SSB with cos(ωc t) followed by LPF.

Frequency domain representation of SSB signals


Given the baseband signal m(t) with spectrum M (ω), the spectrum of DSB-SC and SSB are
shown below (textbook, p174):

30
In either of the USB/LSB cases we can demodulate and extract m(t) from SSB+ or SSB− ,
by regular coherent demodulation, shown as below.

M(omega)

DSB−SC

SSB+ (USB)

SSB− (LSB)

Figure 6: SSB - frequency domain.

31

Demodulator
USB signal
C k m(t)
LPF

carrier

spectrum @ input

fc −2 fc

spectrum of carrier

fc 2 fc

spectrum @ C

−2 fc 2 fc

Figure 7: SSB demodulation, frequency domain.

32
Time domain representation of SSB signals – Hilbert Transform

First, define some notations:


• M (f ), m(t) baseband modulating signal (real)
• M+ (f ), m+ (t) upper sideband (USB) signal (cannot be real)
• M− (f ), m− (t) lower sideband (LSB) signal (cannot be real)

33

Time domain representation of SSB signals - Hilbert Transform

From the spectrum relationship, we have


1 1
M+ (f ) = M (f )u(f ) = M (f ) [1 + sgn(f )] = [M (f ) + jMh (f )] (11)
2 2
1 1
M− (f ) = M (f )u(−f ) = M (f ) [1 − sgn(f )] = [M (f ) − jMh (f )] (12)
2 2
where
1 1
jMh (f ) = M (f )sgn(f ) (13)
2 2
which implies
Mh (f ) = M (f ) · [−jsgn(f )] (14)
Considering the Fourier transform pair 1/(πt) → −jsgn(f ), taking inverse Fourier
transform then we have
-
1 1 ∞ m(α)
mh (t) = m(t) ∗ = dα (15)
πt π −∞ t − α

34
On the other hand, the transfer function can be written as

 −j f ≥ 0
H(f ) = −jsgn(f ) =
 j f <0

H(f ): wideband phase shifter (Hilber Transform).

Thus, if we delay the phase of every component of m(t) by π/2 (without changing its
amplitude), the resulting signal is mh (t), the Hilbert transform of m(t). Therefore, a Hilbert
transformer is an ideal phase shifter that shifts the phase of every spectral component by
−π/2.

35

Based on the Hilbert transform, we have


1
m+ (t) = [m(t) + jmh (t)] (16)
2
1
m− (t) = [m(t) − jmh (t)] (17)
2
where mh (t) is called Hilbert Transform of m(t).

36
Time domain representation of SSB signals using Hilbert Transform

The USB spectrum is

ΦU SB (f ) = M+ (f − fc ) + M− (f + fc ) (18)
1 1
= [M (f − fc ) + M (f + fc )] − [Mh (f − fc ) − Mh (f + fc )]
2 2j
The inverse Fourier transform is then

sU SB (t) = m(t) cos(ωc t) − mh (t) sin(ωc t) (19)

Similarly, we can show that

sLSB (t) = m(t) cos(ωc t) + mh (t) sin(ωc t) (20)

Hence, a general SSB signal can be expressed as

sSSB (t) = m(t) cos(ωc t) ∓ mh (t) sin(ωc t) (USB and LSB) (21)

37

Generation of SSB Signals

A. Selective Filtering Method


It is the most common method of generation SSB. The basic idea is the following
• Using m(t) to generate DSB-SC (m(t) cos ωc t)
• DSB-SC goes through a BPF

m(t) SSB
BPF

carrier

For successful implementation of this method, we must have


• B << fc
• m(t) must have little or no low-frequency content, i.e., M (ω) has a “hole” at
zero-frequency. For example, voice grade speech signal [0.3 3.4] khz.
• Why do we need “frequency hole”? If not, low frequency component cannot be kept.

38
Ideal filter (not realizable).

B. Phase-shift Method

φSSB (t) = m(t) cos(ωc t) ∓ mh (t) sin(ωc t)

39

Demodulation of SSB Signals

A. Coherent demodulation
Observe that

sSSB (t) cos ωc t = m(t) cos2 (ωc t) ∓ mh (t) sin(ωc t) cos(ωc t)


1 1 1
= m(t) + m(t) cos 2ωc t ± mh (t) sin 2ωc t
2 2 2
If we filter φSSB cos ωc t with a LPF, we can eliminate the components centered at 2fc and
the filter output will be ∼ m(t)
Hence, any of the coherent demodulation techniques applicable for DSB-SC signals can be
used.
B. Envelope Detection with a Carrier (SSB+C)
As a variation to the basic SSB case, we can add a carrier to the SSB signal and attempt to

40
use envelope detector

sSSB+C (t) = A cos(ωc t) + m(t) cos(ωc t) ∓ mh (t) sin(ωc t)


= [A + m(t)] cos(ωc t) ∓ mh (t) sin(ωc t)
= E(t) cos(ωc t + Θ(t))

where E(t) is envelope, given as


.
E(t) = [A + m(t)]2 + [mh ]2

and the phase is given as


" #
−1 mh
Θ(t) = tan
A + m(t)

At the receiver, a properly designed envelope detector will extract E(t) from sSSB+C .
Observe that

E(t) = [A2 + 2m(t)A + m2 (t) + m2h (t)]1/2 (22)


+ ,1/2
2m(t) m2 (t) m2h (t)
= A 1+ + +
A A2 A2

41

If A >> |m(t)| or |mh (t)|, E(t) can be approximated as


+ ,1/2
2m(t)
E(t) ≈ A 1 +
A
Using a series expansion and discarding higher order terms due to m/A << 1, we have
+ ,
m(t)
E(t) ≈ A 1 + + · · · = A + m(t)
A
It is evident that for a large carrier, the SSB+C can be demodulated by an envelope detector.
In AM, we need A > mp = |m(t)|, while SSB+C, we need A >> |m(t)|. Therefore,
SSB+C is very inefficient.

42
Vestigial Sideband Modulation (VSB)

Some observations

• SSB modulation is well suited for transmission of voice signals (or for all signals which
exhibit a lower component at f ≈ 0).

• DSB generation is much simpler, but requires twice the signal bandwidth.

• VSB modulation represents a compromise between SSB and DSB modulation systems.

• Simply stated VSB: one sideband is passed almost completely whereas just a trace (or
vestige) of the other sideband is retained.

43

m(t) VSB
M(f) HPF

carrier VSB generation

DSB

fc

SSB (USB)

fc

Hi(f): HPF

fc
VSB

fc fc+B
fc−f1

Figure 8: Illustration – VSB.

44
where Hi (f ) is vestigial shaping filter that produces VSB from DSB. It allows the
transmission of one sideband, but suppresses the other sideband, not completely, but
gradually. How do we design Hi (f ) to generate VSB signal?

A(w) O(w)

ΦV SB (f ) = ΦDSB−SC · Hi (f ) = [M (f + fc ) + M (f − fc )] · Hi (f ) (23)
It can be demodulated by multiplying the carrier:

A(f ) = [ΦV SB (f + fc ) + ΦV SB (f − fc )]
≈ [M (f − 2fc ) + M (f )]Hi (f − fc ) + [M (f + 2fc ) + M (f )] · Hi (f + fc )
= [Hi (f − fc ) + Hi (f + fc )]M (f ) + other terms
O(f ) ≈ H0 (f )[Hi (f − fc ) + Hi (f + fc )]M (f ) = M (f )

45

Thus, we require
1
H0 (f ) = |f | ≤ B (24)
Hi (f + fc ) + Hi (f − fc )
Furthermore, if we choose

Hi (f − fc ) + Hi (f + fc ) = 1 |f | ≤ B

the output filter is just a simple low-pass filter.

Further observations

• limfv →0 [V SB] = SSB, limfv →B [V SB] = DSB

• VSB is demodulated using coherent detector.

• As an alternative, we can use VSB+C (envelope detector)

• BT (VSB)≈ 1.25BT (SSB)

• 1/3 ≥ η(AM ) > η(V SB + C) > η(SSB + C)

46
Superheterodyne Receiver

• Consider the FDM signal, to receive this signal and to tune in to a particular “channel”,
we may require a receiver with the following structure

fc1 fc2 fc3

Tunable Audio
Demodulator
RF stage amplifier

• The above system will function as required. However, the design and implementation of
a tunable front-end, the RF stage, with sharp cut-off frequencies and high gain over a
wide-range of frequencies, is a difficult task.

• Consider the following scenarios

47

Signal mixed
fc1 fc2 fc3

Tunable RF, move over the entire spectrum


[540 1600]kHz

Signal move over the spectrum by mixing


fc1 fc2 fc3

RF−stage FIXED!

• We know how to “mix” the input signal, i.e., how to move the spectrum up and down
(multiplying the input signal with the output of a local oscillator).

• We can also design a fixed frequency RF stage which has all the desired filtering and
amplication properties.

• Heterodyning: translating or shifting in frequency. The concept which we described in


very general terms above is called heterodyning. This technique consists of either
down-converting or up-converting the input signal to some convenient frequency.

• We use a fixed Intermediate Frequency (IF) band. IF is fixed and is independent of the
fc (the carrier frequency) of the signal we receive.

48
• Commercial AM broadcast fIF = 455 kHz. Carrier frequency assignment
fc ∈ [540, 1600] kHz.

• Let us consider an AM radio station broadcasting at the carrier frequency of fc = 1000


kHz, then
fLO = fc + fIF = 1000 + 455 = 1455 kHz

• Image station: 2fIF above fc ,

fimage = fc + 2fIF = 1000 + 2 · 455 = 1910 kHz

would also appear simultaneously at the IF output if it were not filtered out by the RF
filter.

• The RF filter is hard to provide selectivity against adjacent stations separated by 10 kHz,
but it can provide reasonable selectivity against a station separated by 910 kHz.

49

[Ac + m(t)] cos ωct [Ac + m(t)] cos ωIF t

RF amplifier
with bandpass Frequency IF Audio
filters tunable converter amplifier Detector amplifier
to desired ωc (mixter)
Km(t)

ωc + ωIF
Local oscillator

Figure 9: FM stereo transmitter.

50

You might also like