Int. J. Communications, Network and System Sciences, 2010, 3, 441-445
doi:10.4236/ijcns.2010.35058 Published Online May 2010 (http://www.SciRP.org/journal/ijcns/)
The Performance Improvement of BASK System for
Giga-Bit MODEM Using the Fuzzy System
Ki-Hwan Eom1, Kyo-Hwan Hyun1, Kyung-Kwon Jung2
1Department of Electronic Engineering, Dongguk University, Seoul, Korea
2Department of Electronic Engineering, Hanlim University, Chuncheun, Korea
Received August 21, 2009; revised December 15, 2009; accepted March 17, 2010
Abstract
In this paper we propose an automatic bandwidth control method for the performance improvement of Bi-
nary Amplitude Shift Keying (BASK) system for Giga-bit Modem in millimeter band. To improve the per-
formance of the BASK system with a fixed bandwidth, the proposed method is to adjust a bandwidth of low
pass filter in receiver using the fuzzy system. The BASK system consists of a high speed shutter of the
transmitter and a counter and a repeater of receiver. The repeater consists of four stage converters, and a
converter is constructed with a low pass filter and a limiter. The inputs to the fuzzy system are the reminder
and integral of remainder of counter, and output is a bandwidth. We used a Viterbi algorithm to find the op-
timum detection from output of the counter. Simulation results show that the proposed system improves the
performance compared to the fixed bandwidth.
Keywords: BASK, Giga-Bit MODEM, Bandwidth, Low Pass Filter, Fuzzy System
1. Introduction
The 60 GHz band still being free and unlicensed, a large
bandwidth, for example of the order of 1 GHz, and easily
be used. In digital modulation of the 60 GHz band, a
problem is ISI (Inter Symbol Interference) [1]. Digital
base band signals often are rectangular pulse train. When
rectangular pulses are passed through a band limited
channel, the pulses will spread in time, and the pulse for
each symbol will smear into the time intervals of suc-
ceeding symbols. This causes ISI and leads to an in-
creased probability of the receiver making an error in
detecting a symbol. There are many methods to minimize
ISI as likelihood sequence estimation, whitened matched
filters and decision-feedback equalization [2]. One ap-
proach to minimizing ISI is to use pulse shaping tech-
niques. The most popular pulse shaping filter used in
mobile communications is the raised cosine filter. How-
ever, the maximum value of the RF waveform and raised
cosine filtered pulses do not always match [2-4]. Also, in
Heterodyne method, IF process is given gain of receiver,
but an increase in analog conversion steps, the more the
price will also increase. The BASK system consists of a
high speed shutter and a mixer of the transmitter, and a
counter and a repeater of the receiver for solves these
problems. The high speed shutter of the transmitter is
introduced for pulse shaping, which can minimize ISI.
Using repeater for improve SNR and make rectangular
pulse train. The repeater consists of few stage converters.
A converter is constructed with a low pass filter and a
limiter.
In this paper propose an automatic bandwidth control
method for performance improvement of BASK system.
Propose method is that adjust a bandwidth of low pass
filter in receiver using the fuzzy logic system. The fuzzy
logic system is normally used to formulate human know-
ledge, but here we create the membership functions and
the fuzzy rule base by means of the simulation results.
The inputs of the fuzzy logic system are the reminder
and integral of remainder of counter, and output is band-
width. We use 8 bit counter and Viterbi algorithm with
soft decision. Rule base inference was accomplished
using the max-min inference procedure. Defuzzification
of the bandwidth output was achieved the center of grav-
ity computation. In order to verify the effectiveness of
the proposed method, simulations were performed by
fixed bandwidth and BER.
2. BASK System with a Fixed Bandwidth
In millimeter wave band, BASK system with a fixed

K.-H. EOM ET AL. 442
bandwidth of Giga-bit MODEM without IF process us-
ing high speed shutter for pulse shaping of input signal
and minimize ISI in the transmitter, and using repeater
for improve SNR and make rectangular pulse train in
receiver. Figure 1 shows the block diagram of BASK
system with a fixed bandwidth.
In the transmitter, RCS is raised cosine signal genera-
tor. The transmitter uses a high speed shutter that can
truncate the side lobe of the raised cosine filter. A shutter
performs switching window. The output of a shutter is
given by
sin(/ )
(), 1
()
0,0
s
s
tT
tn
ht t
n







(1)
Where ()t

is a gain for the symbol period,

is
the roll off factor, t is the time, S T is the symbol pe-
riod, and n is the state of the symbol. A shutter func-
tion is to make a constant envelope.
The receiver uses a repeater without IF (Intermediate
Frequency) that consists of two stage converters. A con-
verter is constructed with the LPF and the limiter. Design
parameters of converters are bandwidth of the LPF
(BLPF) and stiffness of the limiter (SL: Stiffest Limiter).
The theoretical solution is given by
()( ())
ii
ytSL xtG (2)
Where ()
i
x
tG is the input of the limiter, ()
i
yt is
the output of the converter, SL is a transfer function of
the limiter. The block diagram of a converter is shown in
Figure 2.
Pulse train
Viterbi
Encode
r
Shutter Mixer LPF
Antenna
Amp
RCS A
c
cos(2 )
i
f
t

(a) The transmitter
Antenna
BDF Amp Rectifle
r
LPF Amp
DC cut
filte
r
Pulse
train
Viterbi
Decoder Counter
nth
Converte
r
1th
Converte
r
U
n
U
1
Repeater
(b) The receiver
Figure 1. The block diagram of BASK system with a fixed
bandwidth.
The repeater can improve signal-to-noise ratio (SNR),
and make rectangular pulse train.
3. Proposed Method
The block diagram of proposed automatic bandwidth
control is shown in Figure 3.
The proposed method is that adjust the bandwidth of
low pass filter in receiver using a fuzzy logic system.
The output of counter in receiver depends on the pattern
sequence deeply, so we need the controls for the ranges
of bandwidth to improve the performance of the system.
The inputs to the fuzzy logic system are the remainder
and integral of remainder of counter, and output is a
bandwidth. In order to create the membership functions
and fuzzy rule base, we simulated on reminder and inte-
gral of reminder of counter. The simulation results of the
reminder and integral reminder of 8 bit counter is shown
in Figure 4.
In Figure 4, we can study that the sum of reminder
jumps if a big reminder happens in negative or positive.
Therefore we apply the bandwidth control using the
fuzzy logic system due to such situations.
The inputs are fuzzified according to the input mem-
bership functions and output membership functions in
Figures 5 and 6.
The fuzzy rule-base consists of a total of 15 rules. The
LPF
Gain
output
i
nput
σ1
xi(t)Gxi(t)
yi(t)
Figure 2. The block diagram of a converter.
Antenna
BDF Amp Rectifle
r
LPF Amp
DC cut
filter
Pulse train
Viterbi
Decoder Counter
nth
Converte
r
1th
Converte
r
U
n
U
1
Repeater
F y look
Figure 3. The block diagram of proposed bandwidth con-
trol system.

K.-H. EOM ET AL. 443
Remainder (x-8
*
Round[x/8])
4
2
-2
-4
Integral of remainder
40
20
-20
-40
50 100 150 200 250 300
(a) Narrow bandwidth
Remainder (x-8
*
Round[x/8])
4
2
-2
-4
Integral of remainder
40
20
-20
-40
50 100 150 200 250 300
(b) Optimal bandwidth
Remainder (x-8
*
Round[x/8])
4
2
-2
-4
Integral of remainder
40
20
-20
-40
50 100 150 200 250 300
(b) Broad bandwidth
Figure 4. Simulation of counter for bandwidth control.
(a) Remainder
(b) Integral of remainder
Figure 5. The membership function of fuzzy input.
Figure 6. The membership function of fuzzy output.
input/output fuzzy relation is chosen on the basis of the
simulation results as shown in Table 1.
In Table 1, R and IR are remainder and integral of
remainder. Linguistic Variables are NB (Negative Big),
NM (Negative Medium), NS (Negative Small), N (Nega-
tive), Z (Zero), P (Positive), PS (Positive Small), PM
(Positive Medium) and PB (Positive Big).

K.-H. EOM ET AL. 444
Table 1. Fuzzy rules.
IR
R NB NS Z PS PB
N PB PM PS Z NS
Z PM PS Z NS NM
P PS Z NS NM NB
Rule base inference was accomplished using the max-
min inference procedure. Defuzzification of the band-
width output was achieved the center of gravity compu-
tation [5].
4. Simulation
In order to verify the effectiveness of the proposed met-
hod, Simulations were performed using MATLAB. The
carrier frequency was 60 GHz and message data rate was
1 Gbps. In order to improve SNR, it is better to change
angle of limiter as θ1 < θ2 < θ3 < θ4, and these parameters
are not required exact value. Viterbi algorithm parame-
ters are constrain length k = 7, coding rate = 1/2, and
generator polynominals for octal codes are 171, 133 [6].
Figure 7 shows the average BER for the signal prior
to repeater and the signal posterior to repeater using Vi-
terbi algorithm.
In Figure 7, SNR of non-shutter, the non-repeater, and
the repeater is 31 dB, 30 dB, and 22 dB respectively
when the BER is 10-3.
Figure 8 shows the average BER for the fixed band-
width and automatically controlled bandwidth by fuzzy
logic system.
In Figure 8, the proposed automatic bandwidth con-
trol method by fuzzy logic system is improved the SNR
BER performance
Carrier to noise ratio E
s
/N
0
/
dB
0 5 10 15 20 25 30
Average bit error rate
repeater
non-repeater
non-shutter
10
0
10
-1
10
-2
10
-3
10
-4
Figure 7. BER performance for the repeater.
Carrier to noise ratio E
s
/N
0
/dB
0 5 10
15 20 25
Average bit error rate
BW control
BW fix
10
0
10
-1
10
-2
10
-3
10
-4
Figure 8. Simulation of BER performance.
about 8 dB at BER of 10-3 against the case of fixed
bandwidth.
5. Conclusions
In this paper proposed a method for improving the per-
formance of the BASK system for automatically tuning
the bandwidth of LPF. The BASK system was con-
structed a high speed shutter of transmitter and a repeater
of receiver. The shutter was introduced for pulse shaping
to improve the intersymbol interference and the repeater
consists of few stage converters, and a converter was
constructed with a low pass filter and a limiter. Proposed
method was using fuzzy logic system. Fuzzy inputs were
remainder and integral of remainder of counter. Output
was bandwidth. In order to verify the effectiveness of the
proposed method, simulations were performed by fixed
bandwidth and BER. The simulation results are summa-
rized as follows:
• Fuzzy System has 2 inputs, 1 output, 15 the number
of fuzzy rules. So that can be configured simply.
• SNR of non-shutter, the non-repeater, and the re-
peater is 31 dB, 30 dB, 22 dB, respectively at BER of
10-3.
• The proposed method is improved the SNR about 8
dB at BER of 10-3 against the case of fixed bandwidth.
6. References
[1] V. R. M. Thyagarajan, R. H. M. Hafez and D. D. Fal-
coner, “Broadband Indoor Wireless Communication in
(20 ~ 60) GHz Band: Signal Strength Considerations,”
Universal Personal Communication, Vol. 2, October
1993, pp. 894-899.
[2] T. S. Rappaport, “Wireless Communications,” 2nd Edi-

K.-H. EOM ET AL. 445
tion, Prentice Hall, New Jersey, 2002.
[3] E. Lindskog and A. Paulraj, “A Transmit Diversity
Scheme for Channels with Intersymbol Inference,” Pro-
ceedings of IEEE International Conference on Commu-
nications, New Orleans, Vol. 1, June 2000, pp. 307-311.
[4] S. Haykin, “Communication Systems,” 4th Edition, John
Wiley Inc., Canada, 2000.
[5] R. Johnston, “Fuzzy Logic Control,” GEC Journal of
Research, Vol. 11, No. 2, 1994, pp. 99-109.
[6] M. Hosemann, R. Habendorf and G. P. Fettweis, “Hard-
ware-Software Codesign of A 14.4 Mbit - 64 State - Vi-
terbi Decoder for An Application-Specific Digital Signal
Processor,” Proceedings of IEEE Workshop on Signal
Processing Systems 2003, Seoul, 27-29 August 2003, pp.
45-50.