Implementation of SISO-OFDM Transmission using MATLAB on DSP Processor

www.ijsetr.com
ISSN 2319-8885
Vol.05,Issue.04,
February-2016,
Pages:0676-0682
Copyright @ 2016 IJSETR. All rights reserved.
Implementationof SISO-OFDM Transmission using MATLAB on DSP Processor
S. V. N. SATTYA KIRAN
1
, S. P. V. SUBBA RAO
2
1
PG Scholar, Dept of ECE, Sreenidhi Institute of Science & Technology, Hyderabad, TS, India, E-mail: svnskiran@gmail.com.
2
Professor, Dept of ECE, Sreenidhi Institute of Science & Technology, Hyderabad, TS, India.
Abstract: The design and deployment of a real-time simplex wireless SISO-OFDM transmission setup on 5GHz employing on
TMS320C6713 DSP Starter Kits (DSKs) as the baseband transmitter and receiver is proposed. The baseband system level design
is done using Matlab/Simulink and the deployment on the TMS320C6713 DSKs is performed by means of IDE link via automatic
code generation directly from Simulink blocks, offering a flexible way from system level design and simulation to actual
implementation and verification. Orthogonal Frequency Division Multiplexing (OFDM) technology has been adopted by a variety
of standards. This technology was chosen due to its robustness at high data rates in a frequency selective multipath channel. In
orthogonal frequency-division multiplexing (OFDM), multiple user symbols are transmitted simultaneously over orthogonal
subcarriers which form an OFDM symbol. Compared to other modulation methods, OFDM symbols have a relatively long time
duration, whereas each subcarrier has a narrow bandwidth. The bandwidth of each subcarrier is small enough to assume a flat(non
selective) fading in a moderately frequency-selective system. These subcarriers have overlap in time and frequency domains,
nonetheless, the signal waveforms are designed to be orthogonal, due to a cyclic extension of each OFDM symbol in the time
domain. A practical implementation involves the inverse fast Fourier transform (IFFT) at the transmitter and the fast Fourier
transform (FFT) at the receiver. The narrowband nature of subcarriers makes the signal robust against frequency selectivity
caused by a multipath delay spread. However, OFDM is relatively sensitive to its dual, namely, time selectivity, which is due to
rapid time variations. Time variations corrupt the orthogonality of the OFDM subcarrier waveforms so that inter carrier
interference (ICI) occurs. Orthogonal frequency division multiplexing (OFDM) techniques has widely been considered to be a
very promising strategy to enhance data rate, capacity, and quality for broadband wireless systems.
Keywords: SISO-OFDM, MATLAB, Simulink ,Code Composer Studio(CCS), DSP Processor, FFT,IDFT.
I. INTRODUCTION
The rapid growths in telecommunication market in the last
decades are in need of new techniques that can provide high
data rates. Wireless communications has seen enormous
growth in the past few years. Wideband wireless OFDM
communication systems are much popular because of their
high spectral efficiency and the capability of OFDM to
transmit high data rates over Broadband radio channels with
frequency selective fading. Such systems are also popular due
to the inherent error robustness in a multipath environment
and hence has been chosen for several next generation
broadband wireless local area network (WLAN) standards
like IEEE802.11a,IEEE802.11g,and European HIPERLAN/2,
and terrestrial digital audio broadcasting (DAB) and digital
video broadcasting (DVB-T).OFDM is a special case of
multicarrier data transmission technique in which a single
data stream is transmitted over multiple subcarriers of narrow
bands. The subcarriers are closely spaced and are modulated
with low data rates. These subcarriers are orthogonal to each
other. Due to rapid increase in the transistor density, it is
better to keep the functionality completely in a programmable
Digital Signal Processor (DSP) which allows for much faster
changes and upgrades. OFDM needs to compute several
hundred or thousand points of Fast Fourier Transforms (FFT)
within a few tens of microseconds in order to achieve high
speed data transmission in narrow bands. Initially
TMS320C5402 DSP Starter Kit (DSK5402) was used as the
combination of C and Assembly programming languages.
After that the main programming language was assembly.
Since the assembly programming is very difficult, ready to
use assembly code segments are required. So instead of
working on C or Assembly languages, it is better to work with
Matlab and Simulink. First, the use of Simulink enables the
creation of models from the Simulink’s large collection of
pre-written block-sets and then the C code is generated with
the help of Real Time Workshop and Embedded Target for TI
C6000 DSP. Here, both Matlab and Simulink are effectively
used to develop an OFDM system including transmitter and
receiver. The complete OFDM system is simulated in the
Simulink environment, which can then be downloaded on to a
DSP processor. OFDM is leading the engineers into a new era
of digital transmission and is becoming the chosen
modulation technique worldwide.
II. IMPLEMENTATION
Here, the signal processing is description that is we will
begin with a concept of what should be programmed. The
next step is to model the concept with blocks from Simulink’s
large collection of prewritten block sets. Basically, a block

S. V. N. SATTYA KIRAN, S. P. V. SUBBA RAO
International Journal of Scientific Engineering and Technology Research
Volume.05, IssueNo.04, February-2016, Pages: 0676-0682
diagram that models the concept is built using Simulink. If a
specific block required is not included in Simulink’s block
sets, we may choose to write the own blocks from scratch
using Matlab. The Simulink uses a block based approach
to algorithm design and implementation. At this point the
design is still not designed to operate on any specifically
hardware. To do that, we are using the C6x Target and Real
Time Workshop to generate (or build) ANSI C code intended
for the TI C6713 DSK. The C6x Target will then
automatically take the generated ANSI C code from the
SIMULINK block diagram. In the next step, the TI CCS tools
such as the compiler, assembler and a linker are used to
compile the specific machine code and load the targeted
machine code to the TI C6713 DSK hardware. The
experienced designer may choose to directly write the ANSI
C code instead of building the SIMULINK model. The Code
Composer Studio communicates with the DSP board through
an USB host interface. The real time processed data can then
be finally observed using an oscilloscope.
III. OFDM SYSTEM USING DSP PRINCIPLES
OFDM is a special case of multicarrier data transmission
technique in which a single data stream is transmitted over
multiple subcarriers of narrow bands. The subcarriers
are closely spaced and are modulated with low data rates.
They are made orthogonal to one another by appropriately
choosing the frequency spacing between them. The
orthogonality allows simultaneous transmission on a lot of
sub-carriers in a tight frequency space without interference
from each other. The block diagram of an OFDM system is
shown in fig.1. The random data generator generates a bit
stream. This data stream is processed using the 16-QAM
modulator to map the data into complex signals. These
symbols are passed to an IFFT block to perform the IFFT
operation and the time domain signals are obtained.
The time domain modulated OFDM symbol from IFFT block
is converted back to a serial stream and a guard interval in the
form of cyclic prefix is added to each OFDM symbol. This
data is now passed through an AWGN channel. The reverse
operation is done at the receiver to obtain the original data
back. That is, the cyclic prefix is removed and processed in
the FFT block and finally passed through demodulator
for data recovery.
Fig.1. Block diagram of OFDM system.
A. Modulation
In the modulation technique, the signal wave is transformed
in order to send it over the communication channel to
minimize the effect of noise. This is done so that the received
data can be demodulated to give back the original data. In an
OFDM system, the high data rate information is divided into
small packets of data which are placed orthogonal to each
other. This is achieved by modulating the data by a desirable
modulation technique such as Quadrature Amplitude
Modulation.
B. Communication Channel
This is an AWGN channel through which the data is
transferred. The noise present in this channel affects the
signal and causes distortion in its data content.
C. Demodulation
Demodulation is the technique by which the original data is
recovered from the modulated signal which is received at the
receiver end. Here, the received data is first passed through a
low pass filter and the Guard Interval is removed. After the
data is converted back to parallel stream, the FFT operation is
performed. A demodulator is used to get back the original
signal. By considering the un modulated signal data and the
data at the receiving end, the bit error rate and the signal to
noise ratio is calculated. This is achieved by de modulating
the data by a desirable modulation technique such as
Quadrature de amplitude modulation.
IV. OFDM SIMULATION
The OFDM model has been developed in Simulink using
16- level QAM as shown in Fig.2. The blocks included in
this model are random integer generator, Integer to Bit
Converter, Bit to integer converter, general 16 level QAM
modulator, selecting rows and obtaining complex conjugate,
IFFT(Inverse Fast Fourier Transform), FFT (Fast Fourier
Transform frame status Conversion general 16 Level De-
Modulator And a time scope to observe the simulation results.
Fig.2. Simulink model of a complete OFDM system.

Implementationof SISO-OFDM Transmissionusing MATLAB on DSP Processor
V. DIGITAL SIGNAL PROCESSING
DSP is one of the fastest growing fields of technology and
computer science in the world. In today's world almost
everyone uses DSPs in their everyday life but, unlike
PCusers, almost no one knows that he/she is using DSPs.
DSPs are special purpose microprocessors used in all kind of
electronic products, from mobile phones, modems and CD
players to the automotive industry; medical imaging systems
to the electronic battlefield and from dish washers to
satellites. There are many reasons why we process these
analog signals in the digital world. Traditional signal
processing was achieved by using analogue components such
as resistors, capacitors and inductors. However, the inherent
tolerance associated with these components, temperature and
voltage changes and mechanical vibrations can dramatically
affect the effectiveness of analogue circuitry. On the other
hand, DSP is inherently stable, reliable and repeatable. With
DSP it is easy to chance, correct or updates applications.
Additionally, DSP reduces noise susceptibility, chip count,
development time, cost and power consumption.
A. TMS320C6713 DSP Description
The TMS320C67 DSPs (including the TMS320C6713
device) compose the floating point DSP generation in the
TMS320C6000 DSP platform. The TMS320C6713 (C6713)
device is based on the high-performance, Very Long
Instruction Word (VLIW) architecture developed by Texas
Instruments, making this DSP an excellent choice for
multichannel and multifunction applications. The features of
the C6713, 225 MHz device delivering up to1800 MIPs. This
DSP generation is designed for applications that require high
precision accuracy. The C6713 is based on the
TMS320C6000 DSP platform designed to fit the needs of
high-performing high- precision applications such as pro-
audio, medical and diagnostic.
Fig.3. TMS320C6713 board architecture.
Other hardware features of the C6713 board include:
 Embedded JTAG support via USB.
 High-quality 24-bit stereo codec.
 Four 3.5mm audio jacks for microphone, line in,
speaker and line out.
 512K words of Flash and 8 MB SDRAM.
 Expansion port connector for plug-in modules.
 On-board standard IEEE JTAG interface.
 +5V universal power supply.
The DSP environment used in this paper is the C6713. Fig.3
shows the architecture of the C6713.
B. DSP Implementation
In the past, the small systems implemented on DSP by
writing its algorithm by C/C++ or assembly, using a text
programming languages in writing, but long and complicated
algorithms makes the process exhaustive and difficult to
debug, modify, and learn. The following sections describe the
simulation and implementation of OFDM transceiver with on
a DSP platform through a graphical programming language,
which is SIMULINK. This makes every part in the system
architecture very clear and easier to understand, follow,
modify and debug.
VI. DSP SOFTWARE IMPLEMENTATION
In, the C6713 board, like a personal computer, needs to
load software to establish its behaviour and function. This
software’s can be designed in a variety of approaches. The
designer will begin with a concept of what they want to
program. The next step is to model the concept with blocks
from Simulink’s large collection of pre written block sets.
Basically, a block diagram that models the concept is built
using Simulink as shown in Fig.4. If a specific block
required is not included in Simulink’s block sets, the
designer may choose to write their own blocks from
scratch using MATLAB. At this At this point the design is
still not designed to operate on any specified hardware. To do
that, the designer uses the C6x Target and RTW to generate
(or build) ANSI C code intended for the C6713. The C6x
Target will then automatically take the generated ANSI C
code and uses the Code Composer Studio (CCS) tools to
compile specific machine code and finally loads the targeted
machine code to the C6713 hardware. For the experienced
designer, they may choose to directly write the ANSI C code.
Fig.4. Software design flow.

A. Simulation Model
OFDM transmitter with variable symbol timing and carrier
frequency error is designed. The model implements a 64
carrier OFDM waveform with a 16 point cyclic prefix. The
OFDM symbol has 16 pilots, and uses 4 Quadrature
Amplitude (QAM) for the 48 remaining data streams. Once
the implementation of the model in SIMULINK is achieved,
and already tested that it works properly, the next step is
targeting the model on the DSP board. In this paper, the
implementation of the model on a TMS320C6713 DSP board
from Texas Instrument is carried out. Having the
implementation of the model in Simulink and after testing that
this is working properly, the next step is targeting the model
on the DSP board. For accomplishing this assignment, the
following tools have been used:
 Code Composer Studio v 3.1
 Embedded Target for TI c6000 DSP
B. The Code Composer Studio
CCS is the necessary software support tools, which provides
an integrated development environment (IDE), bringing
together the C compiler, assembler, linker, debugger, and so
on. It has graphical capabilities and supports real-time
debugging. It provides an easy-to-use software tool to build
and debug programs. The C compiler compiles a C source
program to produce an assembly source file. The assembler
assembles the source file to produce a machine language
object file, combines object files and object libraries as input
to produce an executable file by the linker with extension
".out". This executable file can be loaded and run directly on
the C6713 processor.
C. Real-Time Data Exchange
DSP programs can be transfer data between the DSP device
and a host computer without affecting the DSP application
using RTDX. There are components on the DSP device
(target) and the host computer that allow this communication
to happen. Real-Time Data Exchange (RTDX) provides real-
time, continuous visibility into the way target applications
operate in the real world. RTDX allows system developers to
transfer data between target devices and a host without
interfering with the target application. Key statistics and
performance can be monitored in real time. Through the Joint
Team Action Group (JTAG), communication with on-chip
emulation support occurs to control and monitor program
execution. The C6713 DSK board includes a JTAG interface
through the USB port.
D. Target for c6000 DSP
Embedded Target for the TMS320C6000 DSP Platform
integrates Simulink® and MATLAB®with Texas Instruments
eXpressDSP™ tools. The software collection lets you
develop and validate digital signal processing designs from
concept through code. The Embedded Target for C6000
DSP consists of the C6000 target that automates rapid
prototyping on your C6000 hardware targets. The target uses
C code generated by Real-Time Workshop® and your
development tools to build an executable file for your targeted
processor. The Real-Time Workshop build process loads the
targeted machine code to your board and runs the executable
file on the digital signal processor. Using the Embedded
Target for C6000 DSP and Real-Time Workshop, you can
create executable code for the following boards:
 C6416 DSP from Texas Instruments
 C6701 Evaluation Module from Texas Instruments
 C6711 DSP from Texas Instruments
 C6713 DSP from Texas Instruments (board used in this
paper work)
 TMDX326040. A daughter card for the C6711 DSK
VII. DEPLOYMENT AND EXPERIMENTAL
DEMONSTRATION
The Simulink models for the transmitter and receiver are
hierarchically organized into subsystems. That is, processes
such as modulation, synchronization, estimation, sampling
clock offset estimation etc., are organized into discrete
modules. Hence, a clear intuitive link between theory and
practice is illustrated. As mentioned, implementation is done
at the TMS320C6713 DSK, through automatic code
generation from these Simulink modules. The TMS320C6713
DSK board revision C, serves as a hardware reference design
for the TMS320C6713 Digital Signal Processor (DSP) . The
DSP is a 32 bit floating-point processor operating at 225MHz.
The board has 8MB of synchronous DRAM and 512KB of
non-volatile Flash memory. The board ships with an
embedded AIC23 stereo codec with sampling frequencies up
to 96 KHz. In this work, we use a sampling frequency of
48 KHz. The first 10 subcarriers(1875Hz) are set to null
(unused) due to codec, mixer and amplifier bandwidths.
Technical specifications are shown in Table 1. Also, to reduce
the Codec’s frame rate, blocks symbols are transmitted/
received and then un-buffered so as processing to be
performed over single transmission symbols.
TABLE I: DSP Parameters
VIII. SIMULATION AND DSP BOARD RESULTS
The simulation and DSP results are taken for CFO=0 Hz,
20 Hz, 100 Hz, 300 Hz, 500 Hz, while STO = 0 sample, 1
sample, 4 samples and SNR=10 dB. Then the Bit Error Rate
(BER) and Mean Square Error (MSE) are plotted against
SNR.

A. Symbol Timing Offset and Carrier Frequency Offset
Effect
These are some of simulation and DSP cases results as
shown in Figs.5 to 13.
Case 1: There is no timing and frequency error CFO = 0,
STO= 0, Channel = No Noise
(a)Data(MATLAB)
(b)Data(C6713)
Fig.5. MATLAB and DSP data Constellation provided
CFO = 0, STO = 0 and No Noise.
Case 2: In this case STO =1 sample, CFO=0 and Channel
=No Noise.
(a) Data (MATLAB)
(b)Data (C6713)
CFO = 0, STO = 1 sample and No Noise.
Case 3: In this case STO = 0, CFO = 100 Hz, and
Channel=No Noise.
(a)Data (MATLAB)
(b) Data (C6713)
CFO=20 Hz, STO = 0 and Channel = No Noise.

Case4: In this case STO=0, CFO=100Hz and Channel
=AWGN SNR = 10.
(a) Data (MATLAB)
(b) Data (C6713)
Fig.8.MATLAB and DSP data Constellation provided
CFO = 100 Hz, STO = 0 and Channel = AWGN SNR = 10.
After simulate the model on different values of STO and
CFO and use Monte Carlo Simulink, plot the Bit Error Rate
and SNR relation for these different values.
Fig.9. Illustrates the Effect of Symbol Timing Offset.
Fig.10. Illustrates the Effect of Carrier Frequency Offset.
From MATLAB results, the Mean Square Error with respect
to SNR for different values of STO and CFO had been plotted
after record the input data and the output data, calculate the
mean square error for these results. Fig.11 and Fig.12
illustrate the effect of timing and frequency offsets.
Fig.11. Illustrates the Effect of Symbol Timing Offset.
Fig.12. Illustrates the Effect of Carrier Frequency Offset.

Fig.13. Code Composer Studio and hardware set up with
DSK 6713.
IX. CONCLUSION
Real time broadcast-like SISO-OFDM wireless transmission
on 5GHz employing TMS320C6713 DSP boards was
demonstrated. Baseband design was based on Simulink and
deployment was done on the C6713 DSP’s via automatic code
generation. Experimental measurements confirmed the
system’s specifications and typical results verified its
functionality and showed its potential for demonstrating
contemporary wireless communication concepts to students.
Thus the purpose of this simulator has achieved. It makes the
study of OFDM processing very easy.
X. REFERENCES
[1]P. Remlein, T. Cogalan and T. Gucluoglu, “OFDM with
Transmit and Receive Antenna Selection Based on Subcarrier
Groups”2010.
[2]Aida Zaier and Ridha Bouallègue, “A Full Performance
Analysis Of Channel Estimation Methods For Time Varying
OFDM Systems”2011.
[3]Anas A. Abu Tabaneh1, Abdulmonem H.Shaheen.
“MATLAB Simulation for Fixed Gain Amplify and Forward
MIMO Relaying System using OSTBC under Flat Fading
Rayleigh Channel”2011.
[4]Michael Dreschmann, Joachim Meyer Michael H¨ubner,
“Time and frequency synchronization for ultra-high speed
OFDM systems”2011.
[5]Navid Daryasafar and Ali Akbar Hashemi, “A Study Of
Channel Estimation Techniques With Carrier-Frequency
Offset Estimation In SISO-OFDM Systems”2012.
[6]Vishal Sharma and Sushank Chaudhary, “Implementation
of Hybrid OFDM-FSO Transmission System”2012.
[7]Nagarjuna.T and Ganesh.S , “Design And Development Of
Hybrid Adaptation Techniques In Mimo Ofdm System For
4G Wireless Networks”2013.
[8]Cong Luong Nguyen, Anissa Mokraoui, Pierre Duhamel,
Nguyen Linh-Trung. “Improved Time and Frequency
Synchronization Algorithm for 802.11a Wireless Standard
based on the SIGNAL Field”2013.
[9]P. Venkateswarlu, R. Nagendra. “Channel Estimation
Techniques in MIMO-OFDM LTE Systems”2014.
[10]Edem Thesis ,” OFDM Modulation Technique And ROF
System, Design-Principle Of Operations.
[11]Vol.Yong Soo Cho, Jaekon Kim And Won Young Yang,
“MIMO-OFDM Wireless Communication With MATLAB”.
Vol.

Implementation of SISO-OFDM Transmission using MATLAB on DSP Processor

More Related Content

Similar to Implementation of SISO-OFDM Transmission using MATLAB on DSP Processor (20)

More from mohan676910 (6)

Recently uploaded (20)

Implementation of SISO-OFDM Transmission using MATLAB on DSP Processor