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.