Master Ppt1

Pilot based channel estimation in OFDM system

Contents

Motivation for OFDM

OFDM System Architecture

Channel Estimation Techniques

Performance Analysis

Conclusion

Reference

1


Motivation

Disadvantages of FDMA

Bad Spectrum Usage

Disadvantages of TDMA

Multipath Delay spread problem

2

[1] Rappaport, T.,(Ed.) Wireless Communication: Principles and Practice. New Jersey: Prentice Hall,.1996.

[2] Taewon Hwang; Chenyang Yang; Gang Wu; Shaoqian Li; Li, G.Y., "OFDM and Its Wireless Applications: A Survey, " IEEE Trans.

Veh. Technol., vol.58, no.4, pp.1673,1694, May 2009.


Combating ISI

3

Guard interval cp > maxusing empty space as guard interval

at the beginning of each symbol

Data part of OFDM system Next OFDM system

Complete OFDM symbol

Data part of OFDM system Next OFDM system

Complete OFDM symbol

End of the symbol is pretended

to beginning guard interval

Still equals to Tcp

Using cyclic prefix OFDM symbol length

Tsym + TcpEfficency: Tsym /[Tsym+Tcp]

Fig. 1 Role of Guard Interval and Cyclic Prefix in Combating ISI and ICI [3]

[3] D. Matic, OFDM Synchronization and Wideband Power Measurements at 60 GHz for Future Wireless Broadband Multimedia

Communications, Ph.D. dissertation, Aalborg University, Denmark, September 2001.


OFDM: Use of Frequency Spectrum

4

Conventional FDM Orthogonal FDM

-3R/4

-R+R +R

BW=2RBW=2R

BW=2R

BW=3R/2

BW=2RBW=4R/3

N=1

N=2

SC BW=R

SC BW=2R/3

N=2

-R/4 R/4 3R/4

-R -2R/3 -R/3 R/3 +2R/3-R/3 R/3 R-R

-R +R

-R +R

-R +R

-R

Fig.2 Spectrum efficiency of OFDM compared to conventional FDM [4]

[4] Fazel, Khaled. "Performance of CDMA/OFDM for mobile communication system." Universal Personal Communications, 2nd Int.

Conf. on. vol. 2. IEEE, 1993.


Problem Statement

To evaluate the performance of the OFDM

communication system using Pilot based Channel

estimation algorithm for Addictive White Gaussian

Noise (AWGN) and Rayleigh Channel employing

International Telecommunication Union (ITU) specified

standard model.

5


6

Implement end to end base OFDM system and

analyze the output with AWGN channel.

Incorporating fading channel i.e. Rayleigh with

power delay profile between transmitter and

receiver

Adding channel estimation block at the receiver in

order to estimate the channel using pilot.

Applying different channel estimate algorithm at the

receiver and obtaining the performance criteria MSE and

Bit Error Rate (BER), and plot the scatter plot

Simulation Results

Flow of dissertation work


System Architecture

7

Fig. 3 OFDM multicarrier digital communication system [36]

[5] Khan, A.M.; Jeoti, V.; Zakariya, M.A., Improved pilot-based LS and MMSE channel estimation using DFT for DVB-T OFDM

systems, IEEE Symp on Wireless Technol and Applicat, pp.120-124, Sept. 2013.


Overview of System Architecture

8

( ) ( ) { }

1 ,..., 2 , 1 , 0 - =

=

N n

k X IDFT n x

( ) ( ) ( )

- =

- + - - = + =

1 ,..., 1 , 0 ,

1 ,..., 1 , ,

N n n x

N N n n N x n x

g g

f ( ) ( ) ( ) n w n h n x y f f + =

( ) ( ) 1 ,..., 1 , 0 - = = N n n y n y f ( ) ( ) { }

1 ,..., 2 , 1 , 0 - =

=

N k

n y DFT k Y

( ) ( ) ( ) ( )

1 ,..., 1 , 0

) (

- =

+ + =

N k

k W k I k H k X k Y ( ) ( ) ( )

1 ,..., 1 , 0 - = = N k k H

k Y k X

e

e

7 6

5 4

3 2 Guard Interval Channel

Guard Removal Output to Frequency Domain

Output Channel Estimation AWGN Channel Estimated

Channel

Input to Time domain 1


OFDM Overview

Divides high-speed serial information signal into multiple lower-speed sub-signals:

Transmits simultaneously at different frequencies in parallel.

Modulation ( BPSK, PSK,QPSK,16QAM, ).

Pilot sub-carriers used to prevent frequency and phase shift errors.

Usage of cyclic prefix for lower multi-path distortion

Controlled overlapping of bands in one channel

Max spectral efficiency (Nyquist rate)

Easy implementation using inverse FFTs

Easy time-freq. Synchronization

Modulate by switching between time and frequency domain

9


Channel Estimation Techniques.

1. Least Square

The least-square (LS) channel estimation method nds the

channel estimate in such a way that the following cost

function is minimized:

10

H

2

HX-Y)H( =J

pp

LS

p YXH1)( -=

T

pp

pp

p

p

p

p

NX

NY

X

Y

X

Y

)(

)(

)2(

)2(

)1(

)1(



2. MMSE

11

estimatechannelH

estimateactualHwhere

:asgivenissignalError

-

-

-= HHe

{ }

-=22 ~

HHEeE

asgiven is MSE



2. MMSE

12

asshown is MMSE of Expression

LSPH

NHHHHMMSE HXXRRH ,112 ))((

~ --+=



3.DFT based CE

13

Fig 4 Channel Estimation based on DFT []

[6] L. L. Yi Wang, Ping Zhang and Zemin Liu, "DFT-based channel estimation with symmetric extension for OFDMA systems," EURASIP

J. Wirel. Commun. Netw., vol. 2009, pp. 1-8, 2009.



3.DFT based CE

14

-=+

=otherwise

LnnznhnhDFT

0

1...2,1,0][][][

[7] L. L. Yi Wang, Ping Zhang and Zemin Liu, "DFT-based channel estimation with symmetric extension for OFDMA systems," EURASIP

J. Wirel. Commun. Netw., vol. 2009, pp. 1-8, 2009.


Comb type Pilot Arrangement

15

-=

==+=

1,...1

0)()1()(

F

p

FNlData

lmxmNXkX

[8] Hieh, MEng-Han and Wei, Che-Ho, Channel Estimation for OFDM Systems based on COMB-Type Pilot arrangement in frequency

selective fading channels. IEEE Trans. on Commun., vol.44, no.1.Feb 1998.


Performance of Pilot based OFDM

system in AWGN channel

16

0 5 10 15 20 25 30

-5

0

5

10LS-linear

Subcarrier Index

Pow

er[

dB

]

True Channel

LS-linear

Before Channel estimation

0 5 10 15 20 25 30

-5

0

5

10LS-linear with DFT

Subcarrier Index

Pow

er[

dB

]

True Channel

LS-linear with DFT


0 5 10 15 20 25 30

-5

0

5

10LS-spline

Subcarrier Index

Pow

er[

dB

]

True Channel

LS-spline


0 5 10 15 20 25 30

-5

0

5

10LS-spline with DFT

Subcarrier Index

Pow

er[

dB

]

True Channel

LS-spline with DFT


0 5 10 15 20 25 30

-5

0

5

10MMSE

Subcarrier Index

Pow

er[

dB

]

True Channel

MMSE


0 5 10 15 20 25 30

-5

0

5

10MMSE with DFT

Subcarrier Index

Pow

er[

dB

]

True Channel

MMSE with DFT


Fig.5 Illustration of performance improvement with MMSE




17

Fig.6 Illustration of performance improvement with DFT based channel estimation for MMSE




18

Fig.7 Received signal constellation diagram before and after channel estimation for AWGN channel




19

Table. 6.2 MSE for different channel estimation algorithm and SER

LS-Linear LS-spline MMSE BER Before/After CE

6.5879e-003 5.5941e-003 2.3241e-003 24/3

LS-Linear with

DFT

LS-spline with

DFT MMSE with DFT BER Before/After CE

1.4400e-003 1.0769e-003 7.3216e-004 24/1

Table. 6.3 MSE for different channel estimation algorithm with DFT and SER


Performance of Pilot based OFDM system in

Rayleigh channel for ITU OFFICE Model

Tap Relative delay (ns) Average power (dB) Doppler spectrum

1 0 0 Flat

2 50 -3 Flat

3 110 -10 Flat

4 170 -18 Flat

5 290 -26 Flat

6 310 -32 Flat

20

Table 1 ITU Indoor office [9], [10]

[9] ITU-R M.1225 International Telecommunication Union, Guidelines for evaluation of radio transmission technologies for

IMT-2000, 1997.

[10] Kun-Chien Hung; Lin, D.W., "Pilot-Based LMMSE Channel Estimation for OFDM Systems With PowerDelay Profile

Approximation," IEEE Trans. on Veh. Technol. , vol.59, no.1, pp.150,159, Jan. 2010.

1. ITU Indoor office




21

Fig.8 Multipath fading components and Impulse response of ITU Indoor office




22

Fig.9 Performance improvement with MMSE based CE for ITU OFFICE


Performance of Pilot based OFDM system in Rayleigh

channel for ITU OFFICE Model

23

Fig.10 Performance improvement with DFT based CE for ITU OFFICE



channel for ITU OFFICE Model

24

Fig.11 Received signal constellation diagram before and after CE for ITU OFFICE model



channel for ITU Pedestrian model


1 0 0 Classic

2 110 -9.7 Classic

3 190 -19.2 Classic

4 410 -22.8 Classic

25

Table 2 ITU Pedestrian [9], [10]


IMT-2000, 1997.



2. ITU Pedestrian




26

Fig.12 Multipath fading components and Impulse response of ITU pedestrian


27



Fig.13 Performance improvement with MMSE based CE for ITU PEDESTRIAN model




28

Fig.14 Performance improvement with MMSE based CE for ITU PEDESTRIAN




29

Fig.15 Received signal constellation diagram before and after CE for ITU PEDESTRIAN model



channel for ITU Vehicular model


1 0 0 Classic

2 310 -1 Classic

3 710 -9 Classic

4 1090 -10 Classic

5 1730 -15 Classic

6 2510 -20 Classic

30

Table 3 ITU vehicular (high antenna) [9], [10]


IMT-2000, 1997.



3. ITU Vehicular




31

Fig.16 Multipath fading components and Impulse response of ITU vehicular




32

Fig.17 Performance improvement with MMSE based CE for ITU VEHICULAR model




33

Fig.18 Performance improvement with DFT based CE for ITU VEHICULAR model




34

Fig.19 Received signal constellation diagram before and after CE for ITU VEHICULAR model


MSE for different channel estimation algorithm with and without DFT

and BER for ITU model

35

MSE Without DFT MSE With DFT BER

Before/After CE

Case Doppler

frequency

LS-linear LS-spline MMSE LS-linear LS-spline MMSE BER

Office 50 7.004e-02 6.998e-02 6.996e-02 6.996e-02 6.993e-02 6.995e-02 779/0

120 7.005e-02 7.004e-02 7.000e-02 6.997e-02 6.993e-02 6.997e-02 789/6

150 1.024e-01 1.021e-01 1.202e-01 1.201e-01 1.202e-01 1.202e-1 794/70

300 1.451e-01 1.453e-01 1.448e-01 1.446e-01 1.448e-01 1.446e-01 790/69

Pedestrian 50 1.495e-01 1.498e-01 1.497e-01 9.878e-02 9.878e-02 9.879e-02 771/0

120 1.499e-01 1.502e-01 1.499e-01 9.859e-02 9.860e-02 9.864e-02 776/23

150 1.117e-01 1.118e-01 1.117e-01 9.977e-01 9.977e-02 9.975e-02 770/33

300 1.117e-01 1.159e-01 1.148e-01 1.004e-01 1.004e-01 1.003e-01 465/67

Vehicular 50 7.470e-02 7.470e-02 7.469e-02 7.458e-02 7.457e-02 7.458e-02 770/0

120 7.466e-02 7.468e-02 7.461e-02 7.447e-02 7.446e-02 7.449e-02 787/19

150 1.148e-02 1.501e-01 1.479e-01 1.446e-01 1.145e-01 1.145e-01 801/57

300 1.330e-01 1.337e-01 1.325e-01 1.324e-01 1.325e-01 1.323e-01 803/282



channel for Extended ITU Pedestrian model


1 0 0 Classic

2 30 -1 Classic

3 70 -2 Classic

4 80 -3 Classic

5 110 -8 Classic

6 190 -17.2 Classic

7 410 -20.8 Classic

36

Table 4 Extended ITU outdoor to indoor and pedestrian-A model [9], [10]


IMT-2000, 1997.






37

Fig.20 Multipath fading components and Impulse response of Extended ITU outdoor to indoor and pedestrian-A model




38

Fig.21 Performance improvement with MMSE based CE for Extended ITU PEDESTRIAN




39

Fig.22 Performance improvement with DFT based CE for Ex ITU PEDESTRIAN model




40

Fig.23 Received signal constellation diagram before and after CE for Ex- ITU PEDESTRIAN model



channel for Extended ITU Urban model

Tap Relative delay

(ns)

Average power

(dB)

Doppler

spectrum

1 0 0 Classic

2 30 -1.5 Classic

3 150 -1.4 Classic

4 310 -3.6 Classic

5 370 -0.6 Classic

6 710 -9.1 Classic

7 1090 -7 Classic

8 1730 -12 Classic

9 2510 -16.9 Classic

41

Table 5 Extended ITU vehicular-A model [9], [10]


IMT-2000, 1997.






42

Fig.24 Multipath fading components and Impulse response of Extended ITU vehicular-A model




43

Fig.25 Performance improvement with MMSE based CE for Ex-ITU URBAN model




44

Fig.26 Performance improvement with DFT based CE for Ex-ITU URBAN model




45

Fig.27 Received signal constellation diagram before and after CE for Ex- ITU URBAN model



channel for Extended ITU Vehicular model


1 0 -1 Classic

2 50 -1 Classic

3 120 -1 Classic

4 200 0 Classic

5 230 0 Classic

6 500 0 Classic

7 1600 -3 Classic

8 2300 -5 Classic

9 5000 -7 Classic

46

Table 6 Extended ITU typical urban [9], [10]


IMT-2000, 1997.






47

Fig.28 Multipath fading components and Impulse response of Extended ITU typical Urban




48

Fig.29 Performance improvement with MMSE based CE for Ex-ITU VEHICULAR




49

Fig.30 Performance improvement with DFT based CE for Ex-ITU VEHICULAR model




50

Fig.31 Received signal constellation diagram before and after CE for Ex- ITU VEHICULAR model


MSE for different channel estimation algorithm with and without DFT

and BER for Extended ITU model

51

Without DFT With DFT BER

Before/After

CE

Case Doppler

frequency

LS-linear LS-spline MMSE LS-linear LS-spline MMSE BER

Pedestrian 50 7.766e-02 7.766e-02 7.762e-02 7.763e-02 7.762e-02 7.761e-02 768/0

120 7.815e-02 7.816e-02 7.805e-02 7.806e-02 7.805e-02 7.804e-02 720/5

150 8.277e-02 8.274e-02 8.247e-02 8.247e-02 8.242e-02 8.243e-02 725/257

Vehicular 50 1.288e-01 1.289e-01 1.288e-01 1.287e-01 1.288e-01 1.288e-01 812/21

120 1.298e-01 1.299e-01 1.297e-01 1.296e-01 1.298e-01 1.296e-01 802/140

150 8.357e-02 8.360e-02 8.346e-02 8.339e-02 8.337e-02 8.342e-02 805/257

300 8.566e-02 8.593e-02 8.508e-02 8.480e-02 8.480e-02 8.494e-02 807/286

Urban 50 1.424e-01 1.426e-01 1.426e-01 1.415e-01 1.417e-01 1.417e-01 808/6

120 1.428e-01 1.431e-01 1.428e-01 1.418e-01 1.421e-01 1.494e-01 804/15

150 8.472e-02 8.454e-02 8.481e-02 8.454e-02 8.417e-02 8.421e-02 804/332

300 7.783e-02 7.811e-02 7.715e-02 7.597e-02 7.597e-02 7.609e-02 805/281


Conclusion

52

OFDM System

Channel Estimation

LS or LMS estimation at pilot frequencies

Interpolation Techniques

Linear

Spline

Results:

Performance analysis of Pilot based channel estimation for AWGN and

Rayleigh channel

ITU and Ex-tended ITU model

As Doppler frequency increases MSE increases

With channel estimation SER and MSE rates are very low compared to

without channel estimation


References

[1] Rappaport, T.,(Ed.) Wireless Communication: Principles and Practice. New Jersey: Prentice Hall,.1996.

[2] Taewon Hwang; Chenyang Yang; Gang Wu; Shaoqian Li; Li, G.Y., "OFDM and Its Wireless Applications: A Survey, " IEEE

Trans. Veh. Technol., vol.58, no.4, pp.1673,1694, May 2009.

[3] D. Matic, OFDM Synchronization and Wideband Power Measurements at 60 GHz for Future Wireless Broadband

Multimedia Communications, Ph.D. dissertation, Aalborg University, Denmark, September 2001.

[4] Fazel, Khaled. "Performance of CDMA/OFDM for mobile communication system." Universal Personal Communications,

2nd Int. Conf. on. vol. 2. IEEE, 1993.

[5] Khan, A.M.; Jeoti, V.; Zakariya, M.A., Improved pilot-based LS and MMSE channel estimation using DFT for DVB-T

OFDM systems, IEEE Symp on Wireless Technol and Applicat, pp.120-124, Sept. 2013.

[6] L. L. Yi Wang, Ping Zhang and Zemin Liu, "DFT-based channel estimation with symmetric extension for OFDMA

systems," EURASIP J. Wirel. Commun. Netw., vol. 2009, pp. 1-8, 2009.

[7] A. M. Khan, Varun Jeoti, M. A. Zakariya, Improved pilotbased LS and MMSE channel estimation using DFT for DVB-T

OFDM System, IEEE Symposium on Wireless Technology and Applications, Kuching, Malaysia, September 22-25,2013

[8] L.J. Cimini, Jr., Analysis and simulation of a digital mobile channel using orthogonal frequency division multiplexing,"

IEEE Trans. on Commun.,vol. 33, pp. 665-675, July 1985.


IMT-2000, 1997.



53


Thank You

54

Master Ppt1

Documents

ofdm system ofdm

ofdm communication system

n n n y n y f

base ofdm system

ofdm synchronization

ofdm system problem

channel estimation algorithm

mmse channel estimation