PERFORMANCE ANALYSIS ON MODULATION TECHNIQUES IN
WCDMA SYSTEM WITH DIFFERENT CHANNEL CONDITIONS
|
CONFIGURATIONS ON WCDMA SYSTEM
|
4.1 Introduction
We begin our research thesis on first reviewing the
high speed data rate modulation schemes, DSSS W-CDMA and fading effects on the
channels. Then, we develop a generic model of DSSS W-CDMA and is being simulated
by MatLab modulation schemes 16-QAM and QPSK. Both modulation techniques are
chosen in this thesis because there are the most important candidates to
deliver higher data rate for High Speed Downlink Packet Access (HSDPA), an
extension of 3G networks [25]. The simulation is done under AWGN noise and
multipath fading channel using MATLAB 7.1.
As it is shown in figure 4.1, the user data is
assumed to be Bernoulli distributed and can be represented as bn(t).
Each user data is then multiplied with independent or different PN code produced
by a PN generator using XOR logical operator. The multiplied signal of each user
is represented as sn(t) after the signal is modulated by either
16-QAM or QPSK. Each signal is added before it is subjected to the channel. At
the receiver, the signal sk(t) is demodulated before the user
data is separated from PN code by XOR logical operator. Finally, when the necessary
simulations are done, tables and graphs of BER as a function of SNR for various
parameters are plotted. Analysis, observations and results will be scaled on
plots based on the simulation results.
Rayleigh fading and AWGN noise (LOS) are selected to
symbolize fading effect in the channel because we want to make a comparison of
W-CDMA system models in two extreme channel conditions. There are many fading
effects that can be categorized as large-scale and small-scale fading. Rayleigh
fading represents the worst case of multipath fading where it represents
small-scale fading due to small changes in position with respect to time that
is Doppler Effect. On the other hand, AWGN represents the thermal noise
generated by electrical instruments.
4.2 Simulation Methodology
As computer based simulations are the most fitting,
powerful and proficient means to stand for the actual or real time scenarios of
mobile radio system. Thus, MATLAB 7.1 has been used to simulate W-CDMA model
based on associated parameters, theories and formulae. So we use the MatLab 7.4
for simulation using m files. Throughout this project, we set the bit rate of
384Kbps for the signal generator.
There will be three W-CDMA wireless cellular system
models that will be used in this research. The models are
1. W-CDMA system in AWGN channel
2. W-CDMA system in AWGN and Multipath Rayleigh
Fading.
3. Multi-user W-CDMA system in AWGN and Multipath
Rayleigh Fading (static and mobile)
4.3 Generation of Spreading Code
In CDMA, the choice of code sequenc1e is very
important in respect to multiuser and multipath interference encountered by the
signal in the channel. To combat these interferences, the code has to have the
following properties:
1. Each
code sequence generated from a set of code-generation functions must be
periodic with a constant length.
2. Each
code sequence generated from a set of code-generation functions must be easy to
distinguish from its shifted code.
3. Each
code sequence generated from a set of code-generation functions must be easy to
distinguish from other code sequences.
The
first and second requirements are important with respect to the multipath
propagation effects that occur in mobile outdoor and indoor radio environments.
However, the third requirement is important with respect to the multiple access
capability of communication systems. Thus, to ensure a distinction level of
codes for requirements 1 and 2, an autocorrelation function and a
cross-correlation function are used respectively
Autocorr.m is the MATLAB function that shows
the autocorrelation function. The argument of this function is the number of
periods of the code for which the autocorrelation function is to be obtained.
Autocorrelation function is used to measure the distinction level and it is
defined as follows:
For instance, to obtain an autocorrelation function
of a code, the following command can be typed in the command window.
X(t)=[1,
1, 1, -1, -1, 1, -1]
>> X=[1, 1, 1, -1, -1, 1, -1];
>> 
=autocorr(X);
=autocorr(X);
In this case, three-stage M-sequence with a code
length of 7 is used. The autocorrelation function is 7 at t=k.T(k=1, 2, 3,…)
and -1 at other points. The value of correlation function R can be obtained by
typing RXX.
On the other hand, cross-corr.m is used to calculate
the value of cross-correlation function between two distinct codes X(t) and
Y(t).
The arguments of this function are the name of the
sequence and the number of periods of the code for which the autocorrelation
function is to be obtained. The following function will be typed to calculate
the cross-correlation function of codes X(t) and Y(t).
>> X=[1, 1, 1, -1, -1, 1,
-1];
>> Y=[1, -1, 1, -1, 1, 1, -1, 1];
>> 
=crosscorr(X,Y);
=crosscorr(X,Y);
Also, in this case, three-stage M-sequence and a
random sequence with a code length of 7 will be used. To calculate the
cross-correlation function,
is typed.
is typed.
Thus, the spreading code can be calculated by using
these autocorrelation and cross-correlation functions.
4.4 Code
Generation by LFSR (Linear Feedback Shift Register)
In this research feedback shift register will be used
to generate code sequences in WCDMA. A shift register contains a number of
cells (numbered 1 to r) and each cell is a storage unit that, under the control
of a clock pulse, moves its contents to its output while reading its new
contents from its input. In a standard configuration of a feedback register,
the input of cell m will be a function of the output of cell m-1
and the output of cell r (the last cell of the shift register) forms the
desired code sequence.
In linear
feedback shift registers (linear FSRs), the function combining the outputs of
cell m-1 and cell r with the input of cell m is linear. Figure shows a single
linear binary shift register, which can generate a sequence from generation
polynomial h(x) = x5 + x2 + 1. In general, the configuration of a linear binary
shift register of n sections is described by a generator polynomial,
which is a binary polynomial of degree n. n, in this case, is the
number of sections of the shift register.
4.5 Generation
of M-Sequence
M-sequence is a sequence generated by a single LSR
where a sequence of possible period, (Nc=2n-1), is
generated by an n-stage binary shift register with linear feedback. To generate
an M-sequence, the generator polynomial must be a generation polynomial of
degree n. Thus, the periodic autocorrelation function of an M-sequence
is given by
If n 
0 mod 4, there exist pairs of maximum-length
sequence with a three-valued cross-correlation function, where the two values
are {-t(n), t(n)-2} with
0 mod 4, there exist pairs of maximum-length
sequence with a three-valued cross-correlation function, where the two values
are {-t(n), t(n)-2} with
The m file is given as mseq.m. The number of
registers, the initial values of the registers and the position of the feedback
taps are given as argument in mseq.m. For instance, suppose the number of
register is 3, the initial values of the registers are [1, 1, 1] and the position
of the feedback tap is in the first and third taps. The generation polynomial
can be expressed as h(x) = x3+ x+1. This
configuration can be visualized using shift register as it is shown in figure.
M-sequence can be generated by using the following command.
>> m1=mseq(3, [1,3],
[1,1,1])
As a result, a three-stage M-sequence [1, 1, 1, 0, 1,
0, 0] is generated as a vector. A fourth argument, which denotes the number of
output, is available in mesq.m. For a given number of N output, N
one-chip shifted M-sequence. For example, another three stage M-sequence is
generated by the following command:
>> m2=mseq(3, [2, 3], [1, 1, 1], 3)
This command yields and output of
Ans=
1 1 1 0 1 0 1 0
0 1 1 1 0 1 0 1
1 0 1 1 1 0 1 0
The shifting of the number of chips given by the
users for the vector or matrix is performed by the function in file shift.m.
The characteristics of the M-sequences can be
evaluated by using functions autocorr.m and crosscorr.m. The following commands
are used to convert the generated code sequences consisting 0 and 1 to code
sequences consisting -1 and 1.
>> m1=m2*2-1;
>> m2=m2*2-1;
The correlation function of three-stage M-sequence m1
can be calculated by typing the following command.
>> autocorr(m1);
The autocorrelation value obtained is [7, -1, -1, -1,
-1, -1, -1]. Thus it satisfies equation (). Next, the following command is used
to find the cross-correlation function between m1 and m2(1,:).
>>
crosscorr(m1,m2(1,:1));
[3, -1,
3, -1, -1, -5, 3] is the cross-correlation value obtained from this command.
This result takes three values namely [-1, -t(n), t(n)-2] where t(n)=5 taken
from equation (). Thus, m1 and m2(1,:) have the characteristics of a preferred
pair.
|
|||
 |
|||
Figure 4.2: Three-stage
M-sequence
4.6 Configuration of Transmitter and Receiver
In this section, the system is configured based on
synchronous WCDMA system. Each user employs their own sequence to spread the
information data. In this downlink transmission, the information data are
modulated by the modulation scheme. Then, the modulated data are spread by code
sequence that is M-sequence. The spreaded data of all users in the system are
transmitted to the mobile users at the same time. The mobile user detects the
information data of each user by correlating the received signal with a code
sequence allocated to each user. The performance of the WCDMA system is studied
based on QPSK and 16-QAM modulation techniques that will be used in this simulation.
The main simulation file is dscdma.m. The parameters
used in the simulation are defined as follows.
Table 4.1: The
parameter for the main file
|
Parameter Name
|
Value
|
|
Symbol rate, sr
|
192000
|
|
Number of modulation levels,
ml
|
2
|
|
Bit rate, br
|
sr * m1
|
|
Number of symbols, nd
|
100
|
|
Eb/No, ebn0
|
10
|
|
Number of filter taps, irfn
|
21
|
|
Number of oversamples, IPOINT
|
8
|
|
Roll-off factor, alfs
|
0.5
|
The coefficients of the filter that evaluates the
performance of QPSK and 16-QAM are defined as follows.
[xh] = hrollfcoef(irfn, IPOINT,
sr, alfs,1); % T Filter Function
[xh2]= hrollfcoef(irfn, IPOINT,
sr, alfs,0); % R Filter Function
In synchronous WCDMA, the number of code sequences
that can be allocated to different users is equal to the number of code lengths.
Therefore, the length of the code sequence must be larger than of code lengths.
Thus, the length of the code sequence must be larger than the number of users.
To generate a code, the number of registers, the position of the feedback tap
and the initial value of the registers has to be specified. Thus, the following
parameters are used.
Table 4.2: The parameter for the code sequence
|
Parameter Name
|
Value
|
|
Number of users
|
1
|
|
M-sequence
|
1
|
|
Number of stage
|
3
|
|
Position of taps for 1st
|
[1 3]
|
|
Position of taps for 2nd
|
[2 3]
|
|
Initial value of register for 1st
|
[1 1 1]
|
|
Initial value of register for 2nd
|
[1 1 1]
|
By using these parameters, a spread code is generated
and the generated code is stored as variable code. Code is a matrix with a
sequence of the number of users multiplied by the length of the code sequence.
Then, the following commands are used to convert generated code sequence
consisting 0 and 1 into a sequence of -1 and 1.
Code =
code * 2 – 1;
Clen =
length(code);
Subsequently,
the parameters for the fading simulator are defined. When rfade is declared as
0 (LOS), the file that evaluates the BER performance in the AWGN channel. On
the other hand, when rfade is 1(NLOS), the simulation evaluates the BER
performance in a multipath Rayleigh fading environment.
Table 4.3: The
parameter for fading simulator
|
Parameter
|
Value
|
|
Rayleigh
fading, rfade
|
0
|
|
Delay
time, itau
|
[0,8]
|
|
Attenuation
level,dlvl1
|
[0.0,40.0]
|
|
Number
of wave to generate fading,n0
|
[6
,7]
|
|
Initial
phase of delayed wave,th1
|
[0.0,
0.0]
|
|
Set
fading counter, itnd1
|
[3001,4004]
|
|
Number of direct waves
+delayed waves, now1
|
2
|
|
Frequency
resolution, tstp
|
1/sr/IPOINT/clen
|
|
Doppler frequency (HZ),fd
|
160
|
|
Flat
Rayleigh environment, flat
|
1
|
Consequently, the number of simulation loops is set.
The variables that count the number of transmitted data bits and the number of
errors are initiated.
Table 4.4: The parameter for simulation loop
|
Parameter
|
Value
|
|
Simulation number of times,
nloop
|
1000
|
|
Number of errors, noe
|
0
|
|
Number of data, nod
|
0
|
The transmitted data in the in-phase channel and
quadrature phase modulated by QPSK or 16-QAM are multiplied by the code
sequence used to spread the transmitted data. The spread data are then
oversampled and filtered by a roll-off filter and transmitted to a
communication channel. The format used to input these new functions does not
depend on the vector or matrix. The files that perform these simulations are
compoversamp2.m and compconv2.m.
Data = rand(user,nd*m1) 0.5;
[ich, qch] =
qpskmod(data,user,nd,m1);
% QPSK modulation
[ich1,qch1] =
spread(ich,qch,code); % Spreading
[ich2,qch2] =
compoversamp2(ich1,qch1,IPOINT); %
Oversampling
[ich3,qch3] =
compconv2(ich2,qch2,xh);
% T filter
It follows with the synthesis of transmitted signals
from users.
If user = = 1 % Number of users is 1
ich4 = ich3;
qch4 = qch3;
else % Number of user is
plural
ich4 = sum(ich3);
qch4= sum(qch3);
end
Then, the synthesized signal is contaminated in a
Rayleigh fading channel.
If rfade = = 0 % in AWGN
Ich5 = ich4;
qch5 = qch4;
else % Rayleigh fading channel
[ich5,qch5]=sefade(ich4,qch4,itau,dlvl1,th1,n0,itnd1,..
now1,length(ich4),tstp,fd,flat);
itnd1 = itnd1 + itndel; % fading counter
end
At the receiver, AWGN is added to the received data
as it is represented in a simulation file comb2.m. Next, the contaminated
signal is filtered by using root cosine roll-off filter.
spow = sum(rot90(ich3.^2 +
qch3.^2)) / nd; %attenuation
Calculation
attn = sqrt(0.5 * spow * sr /
br * 10^(-ebn0/10));
[ich6,qch6] =
comb2(ich5,qch5,attn); %
Add AWGN
[ich7,qch7] =
compconv2(ich6,qch6,xh2); %
filter
sampl = irfn * IPOINT + 1;
ich8 =
ich7(:,sampl:IPOINT:IPOINT*nd*clen+sampl-1);
% Resampling
qch8 =
qch7(:,sampl:IPOINT:IPOINT*nd*clen+sampl-1);
Now the resample data are the synthesized data of all
the users. By correlating the synthesized data with the spread code used at the
transmitter, the transmitted data of all the users are detected. The
correlation is done by despread.m.
[ich9 qch9] =
despread(ich8,qch8,code); % dispreading
Then, the correlated data is demodulated by a
modulation technique. The total number of errors for all the users is
calculated. Eventually, the BER is calculated.
noe2 =
sum(sum(abs(data-demodata))); % QPSK demodulation
nod2 = user * nd * ml;
noe = noe + noe2;
nod = nod + nod2;
To simulate WCDMA system in multipath fading channel
with Doppler shift, similar procedures are used. The Doppler shifts (Hz)
are based on mobile terminal velocity of 60kmph, 90kmph and 120kmph.
4.7 Steps Taken to Realize the Simulation in main.m
file
The simulations for QPSK and 16-QAM modulation
techniques are done by simulating the value of EbNo at a fixed interval. For
example, if the range of EbNo is from 0 to 10 with interval of 1, the value of
BER will be obtained for EbNo at 1 interval. This means the simulation to get
the value of BERs has to be done 11 times. The range of EbNo is determined by
the behavior of the BER at that EbNo’s range.
To realize the simulation of WCDMA in AWGN channel,
the value of rfade is initialize to 0. Otherwise, it can be assigned to 1. When
rfade=1, the channel of WCDMA system is subjected to AWGN and multipath fading
channel. The Doppler shift, on the other hand, is defined in fd. It represents
the value of Doppler shift in Hertz (Hz).
Furthermore, the simulation of 16-QAM can be achieved
by swapping the functions of modulator and demodulator from qpskmod and
qpskdemod to qammod and qamdemod respectively.
4.8 Limitation
DS-CDMA
is the main system model to study the performance of modulation techniques in
multipath channel. There will be no error correction scheme (channel coding)
used in this research. Also, there will be no equalization as well as
interleaving employed in the W-CDMA system model. The receiver is assumed not a
RAKE receiver neither MIMO receiver. The channel is subjected to AWGN noise and
Rayleigh fading only.
Furthermore,
the BER in LOS for this model is based on the Simplified Improved Gaussian
Approximation (SIGA). On the other hand, BER for Rayleigh fading is based on
either synchronous or asynchronous transmissions. For asynchronous
transmission, the assumption is that the Multi Access Interference (MAI) on the
flat Rayleigh fading channel has a Gaussian first-order distribution. However,
characteristic function, Ф, is used in asynchronous transmission to determine
the total MAI, I, and therefore the BER can be computed based on these
variables.
Other Sites : Try Job Online Sharing Idea's Online Tag Your Thoughts readylancer getmicroworker SEOCLERKS AFFILIATE STORE getworkers Social Networking Sites Tips : Get Too Much Socail microlancers cheapworkers Scammy Emails readylancers readyworkers getminiworker sociallancer GET MORE TWITTER FOLLOWER FREE Online Services Online Tips and Trics Video Editing Tips Get Yahoo Answer and Boost Traffic Lets Get Totally Social Scam Parcel Notification Job Alert VIDEO EDITING Godaddy Coupon or Promo Codes Totally Online seocerks affiliate ilancers Scammy Emails Information And Communication Technology Search Scam Email Coupon Codes or Promotional Codes Online SEO Social Media Marketing Graphics Desgin Tips Facebook All About Twitter Youtube Linkedin Online Ads Adsense Freelancing Covid 19 Online Education PHP
Crypto Currency
SQL Backlink All About Torrents Livestream Affiliate All About Weight Loss health Online World Digital Marketing Crowd Fund Real Estate Hotel Business Weather Sports Freelancing Trics and Tips Online Online Online Online Online Scam Email Online Scam Email Google Google Online Google Google
LOSE BELLY FAT EASILY
BURN BELLY FAT
REDUCE BELLY FAT
HOW TO LOSE BELLY FAT
free dating site Online Job Site
Online Ideas
Information about Tungipara
Sharing Idea's Online
Tag Thoughts
Keto Diet Weight Loss
Best Replacement Windows
Energy Efficient Doors
Energy Efficient Windows
Pure Energy Window Company
Door Replacements
Best Window Installation in Michigan
Window Replacement Novi
Warehouse Staffing
Warehouse Staff Agency
>Warehouse recruitment agency
Warehouse Jobs
Warehouse Agency
Storage Firm
Depot Company
Distribution Agency
Inventory Service
Logistics Bureau
Stockroom Enterprise
Fulfillment Organization
Supply Chain Company
Warehousing Corporation
Freight Hub Agency
Hospitality Staffing
Hospitality Staff Agency
Hospitality recruitment agency
Hospitality Jobs
Hospitality Agency
Hospitality Firm
Hospitality Staffing Agency
Hospitality Employment Firm
Hospitality Recruitment Service
Hospitality recruitment Company
Hospitality Industry Jobs
Hospitality Careers
Hospitality Manpower Agency
Hospitality Employee Placement Service
Hospitality Staff Recruitment Agency
Receptionist Staffing
Receptionist Staff Agency
Receptionist recruitment agency
Receptionist Jobs
Receptionist Agency
Front Desk Staffing Agency
Reception Services Firm
Administrative Assistant Placement Service
Office Reception Personnel Agency
Office Receptionist Placement Company
Administrative Assistant Hiring Agency
Front Desk Employee Agency
Reception Personnel Agency
Reception Personnel Firm
Administrative Assistant Recruitment Service
Front Desk Staffing Firm
No comments:
Post a Comment