SAMPLING THEOREM

SAMPLING THEOREM: STATEMENT [1/3]

1. Statement of Sampling Theorem

Given: Continuous-time signal x(t).

2. Derivation of Sampling Theorem

Thats: Bandlimited to B Hertz.

3. Ideal Reconstruction from Samples

Means: Maximum frequency is B Hertz.

4. Determining Signal Bandwidths

Means: X() = F{x(t)}=0 for || 2B.

5. Finite Pulse Width Sampling

RADIAN
Means: Bandwidth=B Hertz=2B SECOND
.

6. Undersampling and Aliasing

SAMPLING THEOREM: STATEMENT [2/3]

SAMPLING THEOREM: STATEMENT [3/3]

Given: Knowledge of only the samples {x(nT )} of x(t).

Then: x(t) can be reconstructed from its samples {x(nT )}

SAMPLE
.
Means: x(t) sampled every T seconds, at rate S= T1 SECOND

If: Sampling rate S =

Signal x(t)

Pulse Train p(t)

=
t

Note: Co-discovered by Claude Shannon (UM Class of 1938)

SAMPLING THEOREM: EXAMPLE #1

SAMPLE
1
x(t)=cos(21000t) sampled at S=8000 SECOND
. T = S1 = 8000
sec.
1000

Set t=nT =n/8000: x(nT )=cos(2 8000 n)=cos( 4 n).

Can reconstruct x(t) from its samples {x(nT )}: (How? See below.)
n
... 0 1 2
3
4 ...
x(nT ) . . . 1.0 .71 0.0 .71 1.0 . . .

Note: Digital Signal Processing is possible because of this.

SAMPLING THEOREM: EXAMPLE #2

Do we need S > 2B or only S 2B?
SAMPLE
1
x(t)=sin(2500t) sampled at S=1000 SECOND
. T = S1 = 1000
sec.
500
Set t=nT =n/1000: x(nT )=sin(2 1000 n)=sin(n)=0!
Cant reconstruct x(t) from its samples {x(nT )}!
1

0.5

1
0

0.5
0.5

0
1
0

0.5
1
0

0.5

1.5

2.5
sec

3.5

4.5

5
3

x 10

> 2B=2(bandwidth).

Where: S > 2B Here 2B is the Nyquist sampling rate.

Sampled Signal x(t)p(t)

1 SAMPLE
T SECOND

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.01

SAMPLING THEOREM: PROOF [1/3]

P

DEF: p(t)=

n= (tnT )=train

SAMPLING THEOREM: PROOF [2/3]

Fourier series expansion: p(t)= pk ej2kt/T = T1

(series) of impulses.

DEF: x(t)p(t)=sampled signal=train of impulses weighted by {x(nT )}.

Fourier coefficient formula: pk = T1

Note: x(t)p(t)=x(t) (tnT )= x(nT )(tnT ).

Then: F{x(t)p(t)}= T1

So: Only need samples {x(nT )} to create x(t)p(t).

S= T1 F{x(t)p(t)}= T1

Signal x(t)

Pulse Train p(t)

Sampled Signal x(t)p(t)

=
t

F{x(t)}

ej2kt/T .

R T /2

j2kt/T
dt= T1 .
T /2 (t)e

F{x(t)ej2kt/T }= T1

X(2(f Tk ))= T1

X(2 Tk ).

X(2(f kS)).

F{x(t)p(t)}=Spectrum of Sampled Signal

f

BS

BS B

B SB

SAMPLING THEOREM: PROOF [3/3]

SAMPLING THEOREM: EXAMPLE

Can: Reconstruct x(t) from x(t)p(t) if (SB) > B S > 2B .

Given: Continuous-time x(t) is bandlimited to 4 kHz.

By: Low-pass filtering x(t)p(t). Cutoff frequency=B Hertz.

SAMPLE
Sample: 10 kHz=10000 SECOND
> 2(4 kHz).

Formula: x(t)= 2B
2B T

X
|

x(nT )(t
nT )}
{z

SAMPLED SIGNAL x(t)p(t)

sin(2Bt)
.
t
|
{z
}

F{x(t)p(t)}=Spectrum of sampled signal. Repeats in f !

Reconstruct: Low-pass filter with cutoff=4 kHz.

LPF h(t)

F{x(t)}

2B(tnT )
.
Interpolation Formula: x(t)= x(nT )(2BT ) sin2B(tnT
)
P

SAMPLING AND RECONSTRUCTION: SUMMARY

p(t)
p(t)

O
O
x(t) x(t)p(t)
{x(nT )}} x(t)p(t) LOWPASS
FILTER x(t)
|
{z
|

SAMPLING

{z

BS

RECONSTRUCTION

BS B

x(t)p(t)

B SB

f
S+B

F{x(t)p(t)}
f
S+B

F{x(t)p(t)}=Spectrum of Sampled Signal

f

14

4 6

14

f kHz

DETERMINING SIGNAL BANDWIDTH: EXAMPLES

1. x(t)= 3t+7
sin(25t): Rewrite: x(t)=3 sin(25t)+7 sin(25t)
.
t
t
2. x(t)=sin(23t) sin(22t): Rewrite: x(t)= 12 [cos(21t)cos(25t)].
1
[rect( 6 ) rect( 4 )].
3. x(t)= sin(23t)t2sin(22t) : Then X()= 2
Convolve length=6 with length=4length=10.
SAMPLE
For all 3: Bandlimited to = 10 or f =5 Hertz. Need 10 SECOND
.

SAMPLING WITH FINITE-WIDTH PULSES [1/2]

SAMPLING WITH FINITE-WIDTH PULSES [2/2]

Problem: The ideal impulse train p(t)= (tnT ) doesnt exist!

Fourier series expansion: pq (t)= pq,k ej2kt/T . Pulse width: .

Soln: Use pq (t)= q(tnT ) where q(t)=short pulse (does exist).

Fourier coefficient formula: pq,k = T1

Then: pq (t)= q(tnT )= (tnT ) q(t)=p(t) q(t):

p(t)
pq (t)
q(t)
*
=

Then: F{x(t)pq (t)}= pq,k F{x(t)ej2kt/T }= pq,k X(2 Tk ).

2T

2T t

2T

2T t

2T

Sample: x(t)=cos(2600t) with

Get: Samples

Set

x(nT )=cos(2 600n

1000 )=cos(1.2n)=cos(0.8n).

Since: cos(1.2n)=cos(1.2n2n)=cos(0.8n)=cos(0.8n).

F{x(t)p(t)}

PPP
P

BS

BS B

PPP
P

B SB

f
S+B

UNDERSAMPLING AND ALIASING [2/4]

Image spectrum impersonates actual spectrum! Overlap!
Low-pass filtering this400 Hz, not 600 Hz, sinusoid!

Aliasing: image spectrum masquerades as actual spectrum!

n
t=nT = 1000
:

F{x(t)}

MAX.
= S < 2B = 2 FREQ
=Nyquist frequency?

1
T = 1000
.

2T t

SAMPLE
EX: 600 Hz sinusoid sampled at 1000 SECOND
. 1000<2(600).

sin( k
T ).

S= T1 F{x(t)p(t)}= pq,k X(2(f Tk ))= pq,k X(2(f kS)).

UNDERSAMPLING AND ALIASING [1/4]

SAMPLE
RATE

1
j2kt/T
dt= k
/2 e

What if

R /2

0.5
0.4
0.3
0.2
0.1
0
1500

1000

500

500

1000

1500

UNDERSAMPLING AND ALIASING [3/4]

UNDERSAMPLING AND ALIASING [4/4]

SAMPLE
500-Hertz sinusoid sampled at 450 SECOND
. Samples: circles.

SAMPLE
x(t)=cos(2300t+1) sampled at 500 SECOND
. Reconstructed=?

Did samples come from 450 Hertz or 50 Hertz sinusoid?

n
x(nT )=cos(2 300
Sample: t= 500
500 n+1)=cos(1.2n+1).

But: cos(1.2n+1)=cos(1.2n2n+1)=cos(0.8n+1)=cos(0.8n1).

Reconstruct: n=500t cos(0.8(500t)1)= cos(2200t1)

0.5

300 Hertz aliased down to 200 Hertz. Phase also changed.

0.5

1
0

0.005

0.01

0.015

0.02