Signals & Systems
7 SAMPLING
7.1. Introduction
Instantaneous sampling in time domain:
ms (t ) = m(t )c(t )
ms (t ) = m(t ) (t − nTs )
1
TS =
n =− fs
Ts : sampling interval
f s : Sampling frequency
Instantaneous sampling in frequency domain
m(t ) ⎯→ M ( )
m(t ) (t − nTS ) ⎯
→ fs M ( − n ) s
n=− n=−
m(t ) ⎯→ M ( f )
m(t ) (t − nTS ) ⎯
→ fS M ( f − nf S )
n=− n=−
Spectral analysis of Instantaneous Frequency
2
(t − nT ) ⎯→ T ( − n )
S S
S
(t − nTS ) ⎯→ fS
n =− n=−
( f − nf S )
If f S 2 f m : − oversampling
Tx : No aliasing PBG = TS
Rx: practical LPF , Ideal LPF with f m fc f s − f m
Recovery - ( f S 2 f m ) ( f m fC f S − f m )
GATE WALLAH ELECTRONICS & COMMUNICATION HANDBOOK 3.65
Signals & Systems
If f S = 2 f m : critical sampling
Tx : Aliasing on verge (No aliasing)
Rx : Ideal LPF with ( f s = f m ) & PBG = TS
Recovery - ( fS = 2 fm ) ( fC = f m )
Case 3: f S 2 f m under sampling
Tx : Aliasing
Rx : Recovery not possible.
Low Pass Sampling Theorem–
A lowpass signal bandlimited to f m Hz can be sampled and reconstructed form its samples if and only
If f S 2 f m f m fc ( f S − f m )
Sampling rate. f s 2 f m
Nyquist rate = minimum sampling rate
( f S )min = 2 f m
1 1
Nyquist interval Ts = =
( f s )min 2 fm
m(t ) f NY
sin c (t ) 1Hz
sin c(at ) a Hz
sin c K (at ) Ka Hz
sin c(at ) + sin c(bt ) Max (aHz, bHz )
sin c(at ) sin c(bt ) (a + b) Hz
sin c(at ) *sin c(bt ) min(aHz, bHz)
d 1Hz
sin c(t )
dt
t 1Hz
sin c( )d
−
Sampling using general carrier pulse train–
m(t ) ⎯→ M ( f )
c(t ) ⎯
→C( f ) = C ( f − nf )
n=−
n s
GATE WALLAH ELECTRONICS & COMMUNICATION HANDBOOK 3.66
Signals & Systems
M s ( f ) = Cn M ( f − nf s )
n−
If ( f s 2 f m ) ( f m f c f s − f m )
L.P.F(P.B.G) y(t)
1 c0 m(t )
1 / Co m(t )
L L C0 m(t )
When c(t) is rectangular pulse train –
2A 2
Cn = sin c n
a a
2A 2n
MS ( f ) =
n =−
a
sin c ( f − nf s )
a
Sampling of Sinusoidal Signal:
Note: f s 2 f m Recovery is possible through BPF
f s 2 f m Recovery not possible through BPF
Calculation of Frequency:
(i) m(t ) = Am cos 2 f mt
C (t ) : Impulse train with period TS → 0, f s ,2 f s ,3 f s , − − −
0 + fm
ms (t ) = m(t )c(t ) ⎯⎯
→ 0 fm same
0 − fm
fs + fm
fs fm
fs − fm
2 fs + fm
2 fs fm
2 fs − fm
(ii) m(t ) = A1 cos 2 f1t + A2 cos 2 f 2t ⎯⎯
→ f1 , f 2
C(t ) = Impulse train , 0, f s 2 f s ,3 f s
0 f1 0 f2
f s f1 f s f 2
2 fs f1 2 fs f 2
GATE WALLAH ELECTRONICS & COMMUNICATION HANDBOOK 3.67
Signals & Systems
(iii) m(t ) = Am cos 2 f mt ⎯⎯
→ fm
mS (t ) = 0 f m , f s f m ,2 f s f m ,3 f s f m
(iv) m(t ) = Am cos 2 f mt
fs fm
2 fs fm
3 fs fm
Band pass sampling
fH
→ GIF
2 fH
fS K =
K fH − fL
Nyauist rate = 2 f H
GATE WALLAH ELECTRONICS & COMMUNICATION HANDBOOK 3.68