Scribd
Upload
55 views52 pages

Fast Fourier Transform

The document discusses OFDM implementation including both offline and real-time parts. It provides an overview of OFDM basics, advantages and disadvantages, block diagram and implementation using MATLAB for offline and Modelsim for real-time. Key concepts like orthogonality, QAM, FFT and twiddle factors are also explained.

Uploaded by

احمد ابراهيم
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
55 views52 pages

Fast Fourier Transform

The document discusses OFDM implementation including both offline and real-time parts. It provides an overview of OFDM basics, advantages and disadvantages, block diagram and implementation using MATLAB for offline and Modelsim for real-time. Key concepts like orthogonality, QAM, FFT and twiddle factors are also explained.

Uploaded by

احمد ابراهيم
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 52

welcome

Real-Time and
Offline
Implementation
of
OFDM system
Outline
• Project overview
• Introduction
OFDM basics
How OFDM works ?
• OFDM Adv & disadv
• OFDM block diagram
• Implementation
Project overview 

Offline part
•Using MATLAB

Real-time part
•Using modelsim
Project overview 

Offline part
Project overview 

Real-Time part
OFDM basics 

Basic principle is splitting a high rate data stream •


into a number of low rate data streams and
transmit them simultaneously over many carriers
Frequency band is divided into many sub channels •
(multi carrier modulation)
Data transmitted over many low rate sub carriers •
Each two adjacent subcarriers are orthogonal and •
separately modulated
Orthogonality causes BW saving •
A guard time is added to each symbol to combat the •
channel delay spread
OFDM Adv. 

Resistant to multipath fading 


Resistant to ISI 
Easy to implement with FFT/IFFT 
Permits higher data rate as 
compared to FDM
Increase security and BW efficiency 
OFDM disadv. 

High sensitivity to ICI 


The problem of synchronization 
symbol synchronization
frequency synchronization
Need IFFT units at Tx & FFT at receiver 
average power ratio-to-Large peak 
Need to linear amplification with large dynamic 
range to keep the orthogonality of The OFDM
signal and introduced in band radiation
Block Diagram Of OFDM System
Orthogonality 
Orthogonality is a property that allows multiple •
information signals to be transmitted perfectly
over a common channel and detected without
interference
Orthogonality condition: 𝑓𝑘 − 𝑓𝑙 •
𝑛
= ,𝑛 𝑖𝑠 𝑖𝑛𝑡𝑒𝑔𝑒𝑟 𝑛𝑢𝑚𝑏𝑒𝑟
Ts
Simulation In Matlab
Overview of OFDM 
Simulation
This project is to illustrate the
process of generating an OFDM
modulated signal, addition of
AWGN, recover the OFDM signal, its
demodulation and estimation of bit
error rate.
OFDM system configuration & 
Parameters
At the beginning of this simulation MATLAB
program it’s needed to initialize all
required OFDM parameters and program variables
to
start the simulation.

Some variables are entered by the user. The rest


are either fixed or derived from the user-input and
fixed variables. The user input variables include:

Input file (Image) 


IFFT size – an integer of a power of two 
Number of carriers 
 OFDM
Transmitter

The transmitter part


contains serial to parallel
and vice versa, 16-QAM,
IFFT, Noise addition
sections. The OFDM is the
modulation scheme
having multicarrier
transmission
techniques here the
available spectrum is
divided into many carriers
each one being
modulated at a low rate
data stream.
Input Image
QAM
y = qammod(x,M) returns a baseband quadrature
amplitude modulated (QAM) signal given input
signal x and modulation order M.
Serial to Parallel

IFFT
OFDM Transmitted Signal
The channel is modeled as a simple AWGN channel. Since the
channel is considered to be an AWGN channel, there is no need
for the frequency domain equalizer in the OFDM receiver
(Frequency domain equalizer will be helpful only if the channel
introduces multipath fading). Since our channel is an AWGN
channel, the frequency domain equalizer block in the above
diagram can be removed.
OFDM Receiver 
The receiver part contains channel
estimation to remove noise effect, serial
to parallel and vice versa, 16-QAM de-
mapping, FFT.
Results 
Rx Image at SNR=0dB •
Rx Image at SNR=10 •
Rx Image at SNR= 20dB •
OFDM

The principles of OFDM have been in existence for several


decades.
Exactly The idea, was proposed in mid-1960s
However, in recent years these techniques
came into practice in modern communications system.

In 4G wireless communication systems


This make us wonder why it take this long time to be
practically used in wireless appications such as 4G
OFDM
Let us simflify the answer as possiple as we can
in early times OFDM was implemented using
an extremely complex structure of analog
oscillators and filters to generate orthogonal
subcarriers.
It was un efficient method Furthermore
it,’s complexity
Instead of applying filter banks, block of
oscillators and mixers at the transmitter and
receiver
for each subcarrier at discrete frequencies
We will use DFT
DFT and FFT

DFT needs N^2 multiplication


and N^2 – N additions

but by the apperance of FFT become easy to get


orthogonal carriers
plus number of operations will be reduced due to the
usage of twiddle factors and radix as it will be explained

𝑁
FFT needs log 2 𝑁 for multiplications
2
And N log 2 𝑁 for addition
HDL
 We said emplementation this mean hardware
hardware design so we need hardware language !!

What meant by HDL
 Benefits of HDL.
 The major benefit of the language is fast design and
better verification.
 Another major advantage is related to complex designs,
which can be managed and verified easily.
 HDL types
 Verilog Vs VHDL
HDL

For example to show the difference between two languages

we present AND Gate in both VHDL and Verilog

AND Gate in Verilog

module and1(c,a,b);
output c;
input a,b;
assign c=a&b;
endmodule
VHDL
The VHDL for a two input AND gate is shown
below:
For simulation we have two programs.
1-Modelsim
2-Xilinx
QAM
• Introduction :
I. Quadrature amplitude modulation (QAM) is a form of
modulation which is widely used because it offers high-rate,
high bandwidth efficiency systems.
II. QAM is a signal in which two carriers shifted in phase by 90
degree are modulated and the resultant output consist of both
amplitude and phase variations, so QAM is a combination of
both amplitude and phase modulation.
- The I and Q signals can be represented by the equations
below:
I = A cos(Ψ) and Q = A sin(Ψ)
It can be seen that the I and Q components are
represented as cosine and sine. This is
because the two signals are 90° out of
phase with one another.
Block Digram

Modulation
techniqes

ASK FSK PSK

QAM
constellation diagram
It shows the different position of the states, •
when the order of QAM increases, the
number of points on the QAM constellation
diagram increase
2-QAM

4-QAM 16-QAM
QAM
why we used QAM not only PSK?

• The answer is that the distance between


constellation points in M-QAM is larger than
in M-PSK, making detection easier, despite
amplitude changes.
• For example, if you compare QPSK and
QAM, they are identical systems in terms of
constellation points. But if you compare 16
PSK with 16 QAM, you see that the distance
between the constellation points in the QAM
system is greater.
• This happens because, the QAM system
offers two degrees of freedom(phase and
amplitude), while the QPSK system offers
only one(phase).
PSK&QAM
constellation diagram
Implementation of QAM

Block digram:
clk

d7 ….. d0

reg
s/p latch

reset

en

mapping
Implementation of QAM

if(latch[7:6]==2'b00)begin
QAM1[0][3]=1;
QAM1[1][3]=1;
10 00
end else if(latch[7:6]==2'b01)begin
QAM1[0][3]=1;
QAM1[1][3]=-1;
end else if(latch[7:6]==2'b11)begin
QAM1[0][3]=-1;
QAM1[1][3]=-1;
end else if(latch[7:6]==2'b10)begin
QAM1[0][3]=-1; 01
QAM1[1][3]=1; 11
end
Results of QAM in modelsim
with verilog
Fast fourier transform

Before going to discuss FFT & IFFT , it’s


good to know their operation .
FFT & IFFT are derived from DFT.
we use FFT & IFFT instead of DFT as
computation of the function is faster for
implementation in DSP and done
simultaneously . So saving time.
TWIDDLE FACTOR

 DFT Function :
N 1
X(k) =  x
n 0
( n )e  j 2 nk / N

WNnk
N 1
X(k)=
 x
n 0
( n )W nk
N

where WNnk is the twiddle factor .


 The factor is calculated and put in a table to make
computation easier and can run simultaneously .
 The twiddle factor depends on the number of point use.
 During IFFT , it doesn’t to recalculate as it can
refere to the table , so saving time .
TWIDDLE FACTOR VALUES FOR FFT
FFT SIGNAL FLOW GRAPH
This figure shows:
 there are three stages and in each stage there
are two operation .
 the upward arrow will execute addition
operation while downward arrow will execute
subtraction operation .
 the subtracted value is multiplied with
twiddle factor value before being processed
into next stage.
 this process is known as butterfly process.
Calculation real and imaginary
numbers
Comperation between fft function and
the radix
Result of radix 2 with verilog
Result of radix 2 by matlab
The final result of verilog code
Thank You

You might also like