by

Manu Prakash
Roll No: 801061015

Under Supervision of Dr. Amit Kumar Kohli Assistant Professor

Electronics and Communication Engineering Department Thapar University.

CONTENTS
Introduction
Fading & Diversity Diversity & OFDM

Introduction to Alamouti scheme

Alamouti STBC Assumption for Alamouti scheme STBC-OFDM

Space Codes Space-Time Codes in OFDM Space-Frequency Codes in OFDM Simulation Results Concluding Remarks Future Scope Applications
2

Introduction
Fading & Diversity Wireless communication systems encounter multipath fading in any environment where there is multipath propagation and some moving objects. The fading channel could be modelled with Nakagami-m distribution if the fading is severe (Rayleigh distribution is special case of Nakagami-m, where m=1) Wireless broadband systems offer Spatial Diversity to combat fading, which can be properly exploited by a proper coding and transmission scheme. The maximum diversity can be realized using the Space-Time Block Codes (STBC) proposed by Alamouti.
3

Introduction
Diversity & OFDM
STBC was first designed assuming a narrowband wireless

system, i.e. a flat fading channel.

Therefore STBC is used in conjunction with OFDM since,
OFDM converts the frequency selective channel into a set

of independent, parallel and flat-frequency sub-channels.

Introduction to Alamouti Code

Alamouti-STBC Following code is an example of Space-Time Block code (STBC)

Fig. 4: Diversity scheme using STBC

5

Introduction to Alamouti Code

Assumption for Alamouti scheme

Assuming that channel response remains constant across

two consecutive symbols, we can write

But in fast fading channels it is not the case. Therefore OFDM is used with STBC.

Introduction to Alamouti Code

STBC-OFDM OFDM converts a frequency selective channel into a collection of (relatively) flat fading channels(h0 & h1 constant); Therefore it satisfies the Alamouti code requirement; Hence STBC-OFDM is introduced.
Spectral efficiency is increased by allowing frequency

overlapping of the different carriers (compared to FDMA system).

Space-Time Block Codes in OFDM

-X*2 X1 Consecutive symbols

C2T2 Space 1
Space 2
4 3

C2T1 3
4

C1T2
2 1

C1T1 1
2
8

CiTj -> subcarrier no. i, time slot j

Space-Freq. Block Codes in OFDM

-X*2 X1 Adjacent Subcarriers

C2T2 Space 1
Space 2
4 3

C1T2 3
4

C2T1
2 1

C1T1 1
2
9

CiTj -> subcarrier no. i, time slot j

Space-Frequency Block Codes in OFDM

To avoid the problem of fast channel variations in

time, the symbols of an orthogonal design can be transmitted on neighboring subcarriers of the same OFDM symbol rather than on the same subcarrier of subsequent OFDM symbols.
Space

frequency block code has the better performance compared to space- time block coding in highly varying environments.

10

SFBC-OFDM implementation

11

1 =

1 1

=0

1 =

1 2 [1 2 + 1 2 + 1 ] =0 2

1 2

(n) (n)

+ + X + X

x1(n)
0.5

(arg)*

X ++

X -

x2(n)

Similarly for second antenna 1 1 2 () = =0 2

2 =

1 2 [2 2 + 2 2 + 1 ] =0 2

Now, applying Alamouti scheme to adjacent carriers we have X1(n) = X1(0), X1(1), X1(2) . X1(N-2), X1(N-1)
X2(n) = X1 (1), X1 0 , X1 (3),. . . . . . X1 1 , X1 ( 2)

(n) (n)

+ + X + X

x1(n)
0.5

(arg)*

X ++

X -

x2(n)

Now, applying Alamouti scheme to adjacent carriers we have X1(n) = X1(0), X1(1), X1(2) . X1(N-2), X1(N-1)
X2(n) = X1 (1), X1 0 , X1 (3),. . . . . . X1 1 , X1 ( 2)
1 2 [X1 (2 + 1) + X1 (2)] =0 2

2 =

2 =

1 2

0 +

1 =

1 2

(n) (n)

+ + X + X

x1(n)
0.5

(arg)*

X ++

X -

x2(n)

Simulation Results
Following parameters are assumed for simulation purpose
Parameter Modulation Technique Bits per symbol SNR FFT size Number of used subcarriers FFT Sampling frequency Value QPSK 2 0 to 20dB 64 52 20MHz

Subcarrier spacing
Used subcarrier index Cylcic prefix duration, Tcp Data symbol duration, Td Total Symbol duration, Ts

312.5kHz
{-26 to -1, +1 to +26} 0.8us 3.2us 4us

15

Simulation Results
SFBC-OFDM performance for different fading channels
10
-1

SFBC-OFDM Performance for different Channels Nakagami-m=1.25; Nakagami-m=1; Nakagami-m=0.75;

10

-2

Symbol Error Rate

10

-3

10

-4

10

-5

10

-6

10 SNR(dB)

12

14

16

18

20

16

Simulation Results
STBC-OFDM performance for different fading channels
10
-1

STBC-OFDM for m=0.75,1,1.25

Nakagami-m=1.25 Nakagami-m=0.75 Nakagami-m=1.0 10

-2

Smbol Error Rate

10

-3

10

-4

10

-5

10 SNR(dB)

12

14

16

18

20

17

Simulation Results
STBC-OFDM & SFBC-OFDM performance comparison for fast fading channel
10
-1

STBC-OFDM, SFBC-OFDM Comparison for m=0.75

STBC-OFDM for m=0.75 SFBC-OFDM for m=0.75 10

-2

Symbol Error Rate

10

-3

10

-4

10

-5

10

-6

10 SNR(dB)

12

14

16

18

20

18

Simulation Results
STBC-OFDM & SFBC-OFDM performance comparison for Rayleigh fading channel
10
-1

STBC-OFDM, SFBC-OFDM Comparison STBC-OFDM for Nakagami-m=1 SFBC-OFDM for Nakagami-m=1

10

-2

Symbol Error Rate

10

-3

10

-4

10

-5

10

-6

10 SNR(dB)

12

14

16

18

20

19

Simulation Results
STBC-OFDM & SFBC-OFDM performance comparison for Slow fading channel
10
-1

STBC,SFBC,OFDM Comparison STBC-OFDM for m=1.25 SFBC-OFDM for m=1.25

10

-2

Symbol Error Rate

10

-3

10

-4

10

-5

10

-6

10 SNR(dB)

12

14

16

18

20

20

Concluding Remarks
We consider a strategy which consists of coding across

OFDM tones and therefore called space-frequency coding. The SFBC system codes across two antennas and over two adjacent subcarriers instead of two consecutive symbol intervals (STBC) and thus becomes robust against fast fading distortion Simulation results show that in mobile environments where fading is relative slow such as Rayleigh (special case of Nakagami-m where m=1,) SFBC-OFDM does not give significant performance over STBC-OFDM
21

Concluding Remarks
more over it is observed that STBC-OFDM mitigate slow

fading where m > 1 more efficiently than SFBC-OFDM.

Further, in case when channel is modeled as fast fading

(Nakagami-m, m < 1) simulation results show that SFBCOFDM outperforms STBC-OFDM.

When compared for different fading channels SFBC-OFDM

gives lowest SER for slow fading (m > 1) over entire SNR range (0-20dB). For poor SNR, more precisely SNR ranging from 0 to 11 dB SFBC-OFDM performs better under fast fading channel (m < 1) then under Raleigh channel (m = 1.)
22

Future Scope
In this thesis work only QPSK modulation technique is dealt

with, the effect of usage of M-QAM on our performance comparison is not addressed here and is a subject for future research. done for doubly selective fading channel.

We have considered time selective channel research can also be The elements of the orthogonal design are distributed in time

over adjacent OFDM symbols in STBC-OFDM which do not give good performance in fast fading as channel is no longer constant for two consecutive symbols. In similar way OC in SFBC-OFDM are distributed over neighboring subcarriers which causes problems in severe frequency-selective channels. Therefore a space-time-frequency mapping, which might be a good compromise, needs to be investigated
23

24