Introduction to 4G

Communication System

1
 Motivation

• High bit-rate wireless applications in a multipath radio

environment.

• OFDM can enable such applications without a high

complexity receiver.

• OFDM is part of WLAN, DVB, and BWA standards

and is a strong candidate for some of the 4G wireless
technologies.
2
 Multipath Transmission

• Fading due to constructive and destructive addition of

multipath signals.

• Channel delay spread can cause ISI.

• Flat fading occurs when the symbol period is large compared

to the delay spread.

• Frequency selective fading and ISI go together.

3
 Delay Spread

• Power delay profile conveys the multipath delay spread

effects of the channel.

• RMS delay spread quantifies the severity of the ISI

phenomenon.

• The ratio of RMS delay spread to the data symbol period

determines the severity of the ISI.

4
 A Solution for ISI channels

• Conversion of a high-data rate stream into several low-rate

streams.

• Parallel streams are modulated onto orthogonal carriers.

• Data symbols modulated on these carriers can be recovered

without mutual interference.

• Overlap of the modulated carriers in the frequency domain -

different from FDM.

5
 OFDM

• OFDM is a multicarrier block transmission system.

• Block of ‘N’ symbols are grouped and sent parallely.

• No interference among the data symbols

sent in a block.

6
 OFDM Mathematics
N 1
s (t )   X k e j 2 f k t t os]
k 0
Orthogonality Condition
T

 g (t ).g (t )dt  0
*
1 2
0
In our case
T
j 2 f p t  j 2 f q t
e
0
.e dt  0

For p q Where fk=k/T

7
Transmitted Spectrum

8
 OFDM terminology

• Orthogonal carriers referred to as subcarriers {fi,i=0,....N-1}.

• OFDM symbol period {Tos=N x Ts}.

• Subcarrier spacing f = 1/Tos.

9
 OFDM and FFT

• Samples of the multicarrier signal can be obtained using

the IFFT of the data symbols - a key issue.

• FFT can be used at the receiver to obtain the data symbols.

• No need for ‘N’ oscillators,filters etc.

• Popularity of OFDM is due to the use of IFFT/FFT which

have efficient implementations.

10
 OFDM Signal
 N 1
s (t )   ( X
n  k  0
n ,k g k (t  nTos ))

e j 2 fk t t os]
g k (t )  
0 Otherwise
k
fk  K=0,..........N-1
Tos

11
By sampling the low pass equivalent signal at a rate N times
higher than the OFDM symbol rate 1/Tos, OFDM frame
can be expressed as:
N 1
m
Fn (m)   X n ,k g k (t  nTos ) t  (n  )Tos m = 0....N-1
k 0 N
 N 1 m

  N .IDFT  X n,k 
j 2 k
Fn (m)    X n,k e N

 k 0 

12
 Interpretation of IFFT&FFT

• IFFT at the transmitter & FFT at the receiver

• Data symbols modulate the spectrum and the time

domain symbols are obtained using the IFFT.

• Time domain symbols are then sent on the

channel.

• FFT at the receiver to obtain the data.

13
 Interference between OFDM Symbols

• Transmitted Signal

OS1 OS2 OS3

• Due to delay spread ISI occurs

Delay Spread

IOSI

• Solution could be guard interval between OFDM symbols

14
 Cyclic Prefix

• Zeros used in the guard time can alleviate interference

between OFDM symbols (IOSI problem).

• Orthogonality of carriers is lost when multipath channels

are involved.

• Cyclic prefix can restore the orthogonality.

15
 Cyclic Prefix

• Convert a linear convolution channel into a circular

convolution channel.

• This restores the orthogonality at the receiver.

• Energy is wasted in the cyclic prefix samples.

16
 Cyclic Prefix Illustration

Tg Tos

OS 1 OS 2

Cyclic Prefix

OS1,OS2 - OFDM Symbols

Tg - Guard Time Interval

Ts - Data Symbol Period

Tos - OFDM Symbol Period - N * Ts

17
 OFDM Transmitter

X0 x0
Parallel
Serial
Input to
Add
to IFFT Serial
Symbols Parallel CP
and
add CP
XN-1 xN-1

RF Section DAC Windowing

18
 OFDM Receiver

x0 X0
ADC Parallel
and Serial to to Serial Output
FFT
Remove Parallel and Symbols
CP Decoder

xN-1 XN-1

19
 Synchronization

• Timing and frequency offset can influence performance.

• Frequency offset can influence orthogonality of subcarriers.

• Loss of orthogonality leads to Inter Carrier Interference.

20
 Peak to Average Ratio

• Multicarrier signals have high PAR as compared to single

carrier systems.

• PAR increases with the number of subcarriers.

• Affects power amplifier design and usage.

21
Peak to Average Power Ratio

22
 The IEEE 802.11a Standard

• Belongs to the IEEE 802.11 system of

specifications for wireless LANs.
• 802.11 covers both MAC and PHY layers.
• Five different PHY layers.
• 802.11a belongs to the High Speed WLAN category
with peak data rate of 54Mbps
• PHY Layer very similar to ETSI’s HIPERLAN
Type 2

23
 Key Physical Layer Things

• Use of OFDM for transmission.

• Multiple data rate modes supported using

modulation and coding/puncturing.

24
Multiple Data Rates/Modes

25
 OFDM Parameters
• Useful Symbol Duration - 3.2s

• Guard Interval Duration - 0.8s

• FFT Size - 64

• Number of Data Subcarriers - 48

• Number of Pilot Subcarriers - 4

• Subcarrier Spacing - 312.5 kHz

26
 OFDM Transmitter

BPSK/
QPSK/
Input Convolution 64QAM/
Scrambler Interleaver
Bits Encoder 16QAM
Constellation
Mapping

IFFT OFDM
DAC
and Symbol
Add CP Construction

27
 Transmitter Features

• 1/2 rate convolution encoder combined with puncturing

to obtain different coding rates

• Interleaving of bits within an OFDM symbol.

• Variable number of bits within an OFDM symbol.

• Sampling period-50ns-64 data samples,16 samples for the

cyclic prefix.

• Windowing operation for pulse shaping.

28
 Data Subcarriers

• DC subcarrier (0th) not used since it can cause

problems in the DAC

• -32 to -27 and 28 to 32 not used.(Guard band on

both extremes)

• Null subcarriers help in reducing out of band

power

29
 Receiver
• Synchronization
• Channel Estimation and Equalization
• FFT (OFDM demodulation)
• Demapping
• De-Interleaver
• Viterbi Decoder
• De-Scrambling
30
 802.11a Receiver

Channel
Received Synchro- Estimation
FFT Demapping
Samples nization And
Equalization

Viterbi
Descrambler Deinterleaver
Data Decoder

31
Frequency offset estimation continued….
• Implementing the self correlation scheme for short
preamble sequence,
M M
1 f 1
4
M  j 2 est 4

 
2
J yc* (n) yc (n  )  e 4
yc (n)
n 0 4 n 0

so that;
f est 1 J*
 arg ....................................................(1)
4 2 J
M  Number of samples in the short preamble.

32