UCBC

UCSC Broadband Communications (UCBC) Research Group Hamid R. SadjadpourApril 2004

Space-Time Signal Processing for Wireless Ad-hoc Networks

UCBC

Wireless (Ad-hoc) Networks: Challenges

• Group of nodes communicate with the base station or each other over a common wireless channel.• Interference is one of the limiting factors in such environment.• Fading is another limiting factor for reliable communications.• Future wireless applications require high throughput for such networks.

UCBC

Wireless (Ad-hoc) Networks: Challenges

UCBC

Wireless (Ad-hoc) Networks: Solutions

• The use of multiple antennas at both transmitter and receiver.• Foschini & Gans (98), Teletar (99): Multiple Tx/Rx antennas can increase capacity of wireless channels significantly. • Space-time signal processing is the solution to achieve high data rates.• Many practical communication devices can not carry multiple antennas or at least the same number of antennas as base station.

UCBC

Talk summary

• Part I: Packet transmission in wireless ad-hoc networks.• Part II: Space-time processing for uplink of wireless communications systems. • Part III: Future research in space-time signal processing for mobile wireless networks.

UCBC

Packet Transmission in Wireless Networks

• Gupta & Kumar (2000): Capacity per source-destination of fixed ad-hoc wireless networks goes to zero! • Grossglauser & Tse (2002): Capacity per source-destination of mobile ad-hoc wireless networks goes to a constant using multiuser diversity.• Problem: Infinite delay even with finite number of nodes in the network.

tT etF 1)(Cumulative density

delay function with uniform distribution

of nodes

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

SIR q n

if r r

q n if r r i e the cell boundaryn

r

o

r o

lim ( )

'

( ) ' , . . , .' , ,

' , ,

2

2

2

21 0

1

2

2

1

q n q n

if r r and

if r r and

if r r and

if r r and

if r r and

r r

o

o

o

o

o

' , , ' ,( ) ( )

'

. '

. '

. '

. '

1 01

2

1 4 6 71

3

1 3 3 31

4

1 2 7 01

5

1 2 3 21

6

UCBC

Packet Transmission in Wireless Networks

SIR nr

nr r

rd

nr

r r

rd

nr r

r

ro

o

o o

o

' ( )

ln( ' s in ) ' co s

ln( ' s in ) ' co s

ln( ' s in ) ' co s

1

2

1

1

2

1

21

2

2

0

2

2

0

1

2

d

0

1

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

UCBC

Packet Transmission in Wireless Networks

UCBC

STC in Wireless Networks: Challenges

• To achieve high data rates we need multiple antennas at both ends.• Many mobile units are not capable of carrying multiple antennas. In some cases, they only have a single transmit antenna. • Is there any solution when there is a single transmit antenna? Cooperation diversity.• Can we do it without the need for multiple transmit antennas or using cooperation diversity?

Yes!!

UCBC

STC in Wireless Networks

Tx1

Tx2

Rx1

Rx2

1,1h

2,1h1,2h

1,2h

tN

i

jts

itji

jt nEcthr

1, )(

UCBC

STC in Wireless Networks

Performance Comparison between MIMO and the PA

0.001

0.01

0.1

1

10

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

SNR, dB

Blo

ck E

rro

r R

ate MIMO(2x2)

Virtual_optimum(t)

Virtual_Cnormal(t)

Virtual_optimal

Virtual_Cnormal

MIMO(2x1)

Space-time Convolutional codes for 4 states, Rayleigh fading channel

UCBC

STC in Wireless Networks

Performance comparison between MIMO and the PA

0.0001

0.001

0.01

0.1

1

5 6 7 8 9 10 11 12 13 14 15 16

SNR, dB

Blo

ck

Erro

r R

ate

MIMO

Virtual_optimum(t)

Virtual_Cnormal(t)

Virtual_optimal

Virtual_Cnormal

Space-time Convolutional codes for 16 states, Rayleigh fading channel

UCBC

STC in Wireless Networks

Performance Comparison between MIMO and the PA

0.0001

0.001

0.01

0.1

1

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

SNR, dB

Blo

ck E

rro

r R

ate

MIMO

Virtual_optimum(t)

Virtual_Cnormal(t)

Virtual_optimal

Virtual_Cnormal

STC for 4 states, Rician fading channel, k=0 dB

UCBC

STC in Wireless Networks

Performance Comparison between MIMO and the PA

0.0001

0.001

0.01

0.1

1

3 4 5 6 7 8 9 10 11 12 13 14

SNR, dB

Blo

ck E

rro

r R

ate

MIMO

Virtual_optimum(t)

Virtual_Cnormal(t)

Virtual_optimal

Virtual_Cnormal

STC for 16 states, Rician fading channel, k=0 dB

UCBC

STC in Wireless Networks

Performance Comparison between MIMO and the PA

0.00001

0.0001

0.001

0.01

0.1

1

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

SNR, dB

Blo

ck E

rro

r R

ate

MIMO

Virtual_optimum(t)

Virtual_Cnormal(t)

Virtual_optimal

Virtual_Cnormal

STC for 4 states, Rician fading channel, k=6 dB

UCBC

STC in Wireless Networks

Performance Comparison between MIMO and the PA

0.001

0.01

0.1

1

3 4 5 6 7 8 9 10 11 12 13 14

SNR, dB

Blo

ck E

rro

r R

ate

MIMO

Virtual_optimum(t)

Virtual_Cnormal(t)

Virtual_optimal

Virtual_Cnormal

STC for 16 states, Rician fading channel, k=6 dB

UCBC

STC in Wireless Networks

Performace Comparison of MIMO with the PA

0.0001

0.001

0.01

0.1

1

2 3 4 5 6 7 8 9 10 11 12 13 14 15

SNR, dB

Blo

ck E

rro

r R

ate

MIMO(fading)

Virtual(fading)

MIMO(-10)

Virtual(-10)

MIMO(0)

Virtual(0)

MIMO(5)

Virtual(5)

MIMO(10)

Virtual(10)

MIMO(20)

Virtual(20)

UCBC

Current Activities at UCBC

• ML decoding of linear block code for application to Turbo block codes.• Adaptive coding and modulation for OFDM systems. • Coding and Signal processing for fiber optic channels (PMD compensation).•Tradeoff studies between capacity and delay in various networks.