Mo deling of QPSK transmitter and receiver
using MATLAB
y
Chang-Gene Woo
, Sung-Wook Jung
, and Hwan-Mok Jung
Abstract
|This report describes fundamentals of QPSK
and its implementation. And functional model of the block
has developed using MATLAB.
Keywords
| Digital Communications, QAM, QPSK, Mod-
eling.
I. Introduction
T
HE increasing demand for digital transmission chan-
nels has led to the investigation of spectrally ecient
modulation techniques which is to maximize bandwidth ef-
ciency and thus help ameliorate the spectral congestion
problem.
Quadrature phase shift keying(QPSK) mo dulation is a
well-known technique for achieving bandwidth reduction.
And this rep ort fo cused on designing, and mo deling of a
QPSK transmitter and receiver. Design specications are
given as follows:
Symbol rate: 5Msps, Given channel bandwidth: 7MHz
Pulse shaping lter: root raised cosine lter
Sample rate:
f
s
=1
=t
s
= 25MHz
II. Transmitter Design
Trasmitter parts of the MATLAB script are consists of
`White random binary source generator', `Serial to Parallel
converter', `Zero Insertion' and `Pulse Shaping Filter'.
0 500 1000 1500 2000 2500 3000 3500 4000 4500 500
40
20
0
20
40
60
80
00
Spectrum of Input binary data
Fig. 1. Spectrum of Input random binary data
Spectrum of random input binaydataisshown as gure
y
MATLAB is a registered trademark of The MathWorks, Inc.
Report submitted May 1, 1999 (L
A
T
E
X2
"
used).
Chang-Gene Woo (0991156), Sung-Wook Jung (0982072) and
Hwan-Mok Jung (0992075) are the members of the ASIC Lab.,
Kyungpo ok National University.
1. And Spectrum of I-, Q-channel data with data after zero
insertion is plotted in gure 2.
0 1000 2000 3000 4000 5000
0
50
00
Spectrum of I−channel data
0 1000 2000 3000 4000 500
0
50
100
Spectrum of Q−channel data
0 0.5 1 1.5 2 2.5
x 10
4
20
0
20
40
60
80
00
20
40
Spectrum of I−channel data after zero insertion
0 0.5 1 1.5 2 2.5
x 10
4
−20
0
20
40
60
80
100
120
140
Spectrum of Q−channel data after zero insertio
Fig. 2. Spectrum of I-, Q-channel data,
The design of raised cosine lter is mainly focused on
, the roll-o factor also known as the normalized excess
bandwidth which is given by
=
f
f
x
and can take any
value between 0 and 1. According to the sp ecication
given, the
is 2
=
5. But we dene the roll-o factor(rf )
as 0.1 for the sucient margin. After the `Mapping' and
`Filtering', the data are IF modulated by I, Q mixer respec-
tively. I-channel data are in-phase mixed with IF carrier
and Q-channel data are quadrature mixed.
200 220 240 260 280 300
−0.2
−0.1
0
0.1
0.2
0.3
0.4
0.5
Time domain response of pulse shaping filter
0 100 200 300 400 500
−350
−300
−250
−200
−150
−100
−50
0
50
Transfer function of pulse shaping filter
0 0.5 1 1.5 2 2.5
x 10
4
250
200
150
100
−50
0
50
100
150
Spectrum of I−channel after pulse shaping filter
0 0.5 1 1.5 2 2.5
x 10
4
−250
−200
−150
−100
−50
0
50
100
150
Spectrum of Q−channel after pulse shaping filte
Fig. 3. Time domain response of pulse shaping lter,
III. Receiver Design
Since QPSK systems require a fully coherent or dieren-
tially coherent detection scheme, assuming that the amount