Band Pass Systems, Phasors and Complex Representation of Systems

KEY LEARNING OBJECTIVES

I. Phasors (complex envelope) representation for

sinusoidal signal
narrow band signal

II. Complex Representation of Linear Modulated Signals & Bandpass

System

Phasors and Complex Representation are useful for analyzing

baseband component of a signal
eliminates high frequency carrier components

1
I. Phasors for monochromatic & narrow band signals

x(t) is a narrowband signal (aka bandpass signal) if

X(f) 0 in some small neighborhood of f0 , a high frequency

X(f) 0 for | f f0 | W where W < f0

f0 is usually referred to as center frequency, but need not be

center frequency or in signal bandwidth at all
2W X(f)

-f0 -W -f0 - f0 +W f0 -W f0 f0 +W

h(t) is a Bandpass System,, that passes signals with frequency components in

the neighborhood of some frequency, f0
X(f) H(f) X(f)
H(f) = 1 for | f f0 | W otherwise H(f) 0

bandpass system h(t) passes a bandpass signal x(t)

2
Consider LTI system driven by input x(t)
output determined by multiplying X & frequency response of
system computed at input frequency, f0
input & output frequencies are same output phasor gives output
signal

X(f) H(f) Y(f)

determine the phasor for sinusoida1 signal and narrowband signal

capture phase and magnitude of base band signal
ignore effects of the carrier

3
1. determination of phasor, X for sinusoidal input signal x(t)
x(t) = Acos(2f0 t + )
xq(t) = Asin(2f0 t + )
quadrature component shifted 90o from x(t)
(i) define a signal z(t) as a vector rotating with angular frequency 2f0
z(t) = Aexp(j(2f0t + )) I
xq(t) Aexp(j)
= Acos(2f0t + ) + jAsin(2f0t + )
2f0
= x(t) + jxq(t)
R
(ii) obtain phasor X from z(t) by eliminating 2f0 rotation x(t)

- rotate z(t) at an angular frequency = 2f0 in opposite direction

- equivalent to multiplying z(t) by exp(2f0t)
X = z(t) exp(-j2f0t ) = Aexp(j(2f0t + ))exp(-j2f0t )
= Aexp(j)
4
1a. determine Frequency Domain equivalent of z(t) and X
(i) obtain Z(f), using either or two methods

(1) determine X(f) = F[x(t)], delete negative frequencies & multiply by 2

x(t) = Acos(2f0t + ) = Acos()cos(2f0t) + Asin()sin(2f0t)
A A
X(f) = cos()[(ff0 ) + (f+f0)] - j sin()[(f+f0) - (f-f0)]
2 2
2A
Z(f) = [cos()(ff0 ) + jsin()(ff0 )]
2

(2) determine Z(f) = F[z(t)]

z(t) = Aexp(j(2f0t + )) = Aexp(j)exp(j2f0t )

since F[exp(j2t)] = {(f-)} Z(f) = Aexp(j)(f f0 )

(ii) then shift Z(f) by f0 X = Aexp(j)

5
2. determine phasor for a narrowband signal, x(t)
based on definition of z(t) in sinusoid case: z(t) = x(t) + jxq(t)
find Z(f) by deleting negative frequencies of X(f) & multiply result by 2

Z(f) = 2u-1(f)X(f)
z(t) is known as the analytic signal or pre-envelope of x(t)

find z(t) using IFT find signal whose Fourier transform = u-1(f)
1 1
we know that F[u-1(t)] = ( f )
2 j 2f
1 j
by duality F (t ) = u-1(f)
2 2t
j
by convolution z(t) = ( t ) x (t )
t
1
let x ( t ) x(t ) then z(t) = x ( t ) jx ( t )
t
6
pre-envelope for two types of signals

(i) sinusoid case z(t) = x(t) + jxq(t)

x(t) = Acos(2f0 t+)
xq(t)= Asin(2f0 t+)

(ii) narrowband case

z(t) = x(t) + j x (t )

1
Hilbert Transform of x(t) is given by
x ( t ) x(t )
t

phase shift x(t) by for positive frequencies
2

phase shift x(t) by for negative frequencies
2

7
determine phasor, xl(t) of bandpass signal x(t)
xl(t) = low pass representation of x(t)
determined by shifting spectrum of z(t) left by f0

Xl(f) = Z(f + f0) = 2u-1(f + f0)X(f + f0)

X(f)
A
xl(t) = z(t)exp(-j2f0t)
f0 f0 f
Z(f)
xl(t) is a low pass signal 2A

Xl(f) 0 for all | f | W f

f0
phasor for band pass signal Xl(f)
2A

8
Generally xl(t) is complex signal with real (in phase) & imaginary
(quadrature) components
xl(t) = xc(t) + jxs(t)

z(t) = x ( t ) jx ( t ) rewrite in terms of quadrature & in-phase components

z(t) = xl(t)exp(j2f0t)

= [xc(t) + jxs(t)]exp(j2f0t)

= xc(t)cos(2f0t) - xs(t)sin(2f0t) + j[xc(t)sin(2f0t)+xs(t)cos(2f0t)]

equate real & imaginary parts of z(t) and xl(t)
x(t) = Re{z(t)} = xc(t)cos(2f0t) - xs(t)sin(2f0t)

x ( t ) = Im{z(t)} = xc(t)sin(2f0t)+xs(t)cos(2f0t)

bandpass to lowpass transform describes relationship of x(t) & x ( t )in terms

of xc(t) & xs(t)
9
Define xl(t) in terms of phase & envelope

define envelope of xl(t) as V(t) = xc2 (t ) xs2 (t )

xs (t ) I
define phase of xl(t) as (t) = tan 1
xc ( t )
xl(t)
then xl(t) = V(t)exp( j(t) ) V(t)
xs (t ) (t)
= x (t ) x (t )
2
tan
2
1
c s
xc ( t ) R

V(t) & (t) are slowly time varying

monochromatic phasor has constant amplitude & phase
bandpass signals phase & envelope vary slowly with time vector
representation moves on a curve in the complex plane

10
II. Complex Representation of Linear Modulated Signals & Bandpass
System

canonical representation of any bandpass signal, s(t) has 2 components

s(t) = sI(t)cos(2fct) - sQ(t)sin(2fct)

sI(t) = in-phase component of s(t)

sQ(t) = quadrature component of s(t)

properties of sI(t) & sQ(t)

are real valued functions
are orthogonal to each other
are uniquely defined in terms of the baseband signal m(t)
two components can be used to synthesize modulated signal s(t)

11
circuit used to synthesize s(t) from sI(t) & sQ(t)

sI(t)
cos(2fct)
oscillator
s(t)
90o
sin(2fct)
sQ(t)

circuits used to analyze sI(t) & sQ(t) based on s(t),

LPF sI(t)
2cos(2fct)
oscillator
s(t)
90o
-2sin(2fct)
LPF sQ(t)

12
1. Complex Envelope of a Band-Pass Signal s(t) is given as

ss (t) = sI(t) + jsQ(t)

ss (t) preserves information content of s(t), except for fc(t)

sss (t)e(2fct) = [sI(t) + jsQ(t)] [cos(2fct) + jsin(2fct)]

= sI(t)cos(2fct) - sQ(t)sin(2fct) + j[sI(t)sin(2fct)+sQ(t)cos(2fct)]

real imag

then, s(t) = Re{ss (t)e(2fct)}

= sI(t)cos(2fct) - sQ(t)sin(2fct)

13
2. Consider a narrowband linear band-pass system
system is narrowband if bandwidth W << fc , the systems center
frequency
x(t) h(t) y(t)
input x(t) is modulated by carrier, fc
output = y(t)
canonical representation of systems impulse response given by:

h(t) = hI(t)cos(2fct) - hQ(t)sin(2fct)

use equivalent complex baseband model to simplify analysis

impulse response given by
hs (t) = hI(t) + jhQ(t) xs (t) hss (t) 2(t)

14
2.1 Passband Analysis of LTI System

y(t) = x( )h(t )d

y(t) = [xI()cos(2fc )-xQ(t)sin(2fc )]

[hI(t-)cos(2fct-)-hQ(t-)sin(2fct-)]d

= xI(t) hI(t-) cos(2fct)cos(2fct-)

d

+ xQ(t) hQ(t-) sin(2fct)sin(2fct-) d

- xI(t)hQ(t-)cos(2fct)sin(2fct-) d

-

xQ(t)hI(t-)cos(2fct-)sin(2fct) d

15
Passband Analysis of LTI System (continued)

y(t) = xI(t) hI(t-) [ cos() + cos(4fc t-) ] d

+ xQ(t) hQ(t-) [ cos() - cos(4fc t-) ] d

- xI(t)hQ(t-)[ sin(4fc t) + sin() ]

d

- xQ(t)hI(t-)[ sin(4fc t) - sin() ]

d

16
2.2 Equivalent Complex Baseband Model
complex input & output are complex envelopes of bandpass systems
input & output
xs (t) = xI(t) + jxQ(t) is the complex envelope of x(t)
(t) = yI(t) + jyQ(t) is the complex envelope of y(t)

complex envelopes are related by complex convolution

1 ~ ~ 1 ~ ~
(t) = x ( )h (t )d = x (t )h ( )d
2 2

= [xI(t) + jxQ(t)] [hI(t-) + jhQ(t-)]d

= hI(t-)xI(t) - hQ(t-)xQ(t) + j[xQ(t)hI(t-) + hQ(t-)xI(t)]d

17
Equivalent Notation for complex baseband model ( = convolution)
(t) = (xss (t) hs (t)) = (hs (t) xss (t))
factor added to maintain equivalence between real & complex models
fc is omitted from complex baseband model simplifies analysis
without loss of information

Impulse response of band-pass system given by

h(t) = Re{hs (t)exp(2fct)}
= Re{ (hI(t) + jhQ(t)) (cos(2fct) + jsin(2fct) ) }
= hI(t)cos(2fct) - hQ(t) sin(2fct)

Passband signals are readily determined from (t) and xss (t)
x(t) = Re{xss (t)exp(2fct)}

y(t) = Re{(t)exp(2fct)}

18
Appendix: More on Complex Envelope - viewed as an extension of phasor
for a real harmonic signal x(t)
x(t) = x cos(2f0t + x) tR
assume x 0 and phase is 0 x < 2, then:
(i) exp( j(2f0t+x )) = cos(2f0t +x) + jsin(2f0t +x)

(ii) x(t) = Re[x ( cos(2f0t +x) + jsin(2f0t +x) )] tR

= Re [x exp(j(2f0t + x))] tR
= Re [x exp(jx) exp(j2f0t )] tR

phasor representing phase & magnitude of x(t) = complex envelope:

x exp(jx) = x cos(x) + jx sin(x)
x = magnitude
x = argument (phase of x(t))
19
derive complex envelope for any real continuous signal, x(t)
assume x(t) = Re [xe(t) exp(j2f0t )] tR

where xe(t)= x exp(jx),

i. Take Fourier Transform of x(t)

X(f) = F[x cos(2f0t+ x)] = 1 x exp(jx)(f-f0) + 1x exp(-jx)(f+f0)
2 2

ii. suppress negative frequencies & multiply by 2

xs p(f) = x exp(jx)(f-f0) fR

iii. shift left by f0 to obtain frequency signal

xss e(f) = x exp(jx)(f0) fR

iv. take Inverse Fourier Transform

F-1[xs e(f) ] = xe(t) = x exp(jx)
20
e.g. Pure Harmonic signal given by x(t) = cos(2f1t + x) tR
where x 0
0 x < 2

i. FT yields X(f) = exp(jx)(f-f1) + exp(-jx)(f+f1)

ii. xs p(f) = exp(j)(f-f1)

iii. xss e(f) = exp(j)(f-f1+f0)

iv xe(t) = exp(j)exp(2j(f1-f0))t tR

if f1 = f0 complex envelope = phasor

if |f1-f0| << f0 xe varies slowly compared to exp(2jf0t)

21
If x(t) = real, continuous function, & F(x) has no delta function at f = 0

pre-envelope (aka analytical) of x is complex valued signal xp with

F[xss p] = xss p(f) = 2X(f)1(f) fR

complex-envelope of x with respect to frequency f0 is signal xe

xs e(f) = xs p(f+f0) = 2X(f+f0) 1(f+f0) fR

xe(t) = F-1[ xss e(f) ]

22
Complex Envelope for let x(t) = real, band-pass, band-limited signal
fc = center frequency & W = bandwidth
where W < fc, are positive real numbers (W << fc x(t) is narrowband)
X(f) = 0 for | f | < fc-W and | f | > fc+W fR

X(f) W W

0
-fc 0 fc
x p ( f )

xp = analytical
0 fc

x e ( f )
xe = complex envelope with respect to f0
contains only low frequencies
f0 R+ xe is not uniquely defined 0
23