Short Course: Wireless Communications : Lecture 3. Professor Andrea Goldsmith. UCSD March 2223 La Jolla, CA. Lecture 2 Summary. Capacity of Flat Fading Channels. Four cases Nothing known Fading statistics known Fade value known at receiver Fade value known at receiver and transmitter
Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.
Wireless Communications: Lecture 3
Professor Andrea Goldsmith
UCSD
March 2223
La Jolla, CA
Bc
Frequency Selective Fading Channels1/H(f)2
f
M(g) Points
log2 M(g) Bits
To Channel
M(g)QAM
Modulator
Power: S(g)
Point
Selector
Uncoded
Data Bits
Delay
g(t)
g(t)
16QAM
4QAM
BSPK
VariableRate VariablePower MQAMGoal: Optimize S(g) and M(g) to maximize EM(g)
gk
g
Optimal Adaptive SchemeEquals Shannon capacity with
an effective power loss of K.
g
cos(2pf0t)
cos(2pfNt)
x
x
MCM and OFDMR/N bps
QAM
Modulator
R bps
Serial
To
Parallel
Converter
R/N bps
QAM
Modulator
Tc
Spread SpectrumS(f)
s(t)
sc(t)
Sc(f)
S(f)*Sc(f)
1/Tb
1/Tc
Tb=KTc
2
Lecture 3
Channel or MAC):
Many Transmitters
to One Receiver.
Downlink (Broadcast Channel or BC):
One Transmitter
to Many Receivers.
x
x
x
x
h1(t)
h21(t)
h22(t)
h3(t)
Multiuser Channels:Uplink and DownlinkR3
R2
R1
Uplink and Downlink typically duplexed in time or frequency
Code Space
Code Space
Time
Time
Time
Frequency
Frequency
Frequency
Bandwidth Sharing7C29822.033Cimini9/97
a2
a1

Signal 1
=
A/D
Signal 1
Demod
A/D
A/D
A/D
A/D
Iterative
Multiuser
Detection
Signal 2
Signal 2
Demod

=
Why Not Ubiquitous Today?
Power and A/D Precision
RANDOM ACCESS TECHNIQUES
7C29822.038Cimini9/97
Capacity: The set of simultaneously achievable rates {R1,…,Rn}
R3
R2
R3
R2
R1
R1
H1(w)
H2(w)
Broadcast Channels with ISIw1k
xk
w2k
Nondegraded
broadcast channel
MIMO MAC capacity easy to find
MIMO BC channel capacity obtained using dirty paper coding and duality with MIMO MAC
In licensed bands
and unlicensed bands
Wifi, BT, UWB,…
Cellular, Wimax
Due to its scarcity, spectrum is reused
STATION
Cellular System Design8C32810.44Cimini7/98
Increases BER, reduces capacity
Multiuser detection can
completely remove interference
How should cellular
systems be designed?
Coop
MIMO
Femto
Relay
Will gains in practice be
big or incremental; in
capacity or coverage?
DAS
In practice, all techniques have roughly the same capacity
BASE
STATION
A=.25D2p =
fc=2 GHz
101
100
D=4R
Average Area Spectral Efficiency
[Bps/Hz/Km2]
D=6R
D=8R
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Cell Radius R [Km]
STATION
Dynamic Resource AllocationAllocate resources as user and network conditions changeS
D
Application metric: f(C,D,E):
(C*,D*,E*)=arg max f(C,D,E)
(C*,D*,E*)
Is a capacity region all we need to design networks?Yes, if the application and network design can be decoupled
Capacity
Delay
Energy
3
5
4
1
Ad Hoc Network Achievable Rate Regionsvectors achieved
by time division
Capacity region
is convex hull of
all rate matrices
Achievable RatesCapacity region is 30dimensional
(a): Single hop, no simultaneous
transmissions.
(b): Multihop, no simultaneous
transmissions.
(c): Multihop, simultaneous
transmissions.
(d): Adding power control
(e): Successive interference
cancellation, no power
control.
Multiple
hops
SIC
Spatial
reuse
Extensions:
 Capacity vs. network size
 Capacity vs. topology
 Fading and mobility
 Multihop cellular
Terminal
Exposed
Terminal
1
2
3
4
5
Medium Access ControlDestination
Source
“A Performance Comparison of MultiHop Wireless Ad Hoc Network
Routing Protocols”: Broch, Maltz, Johnson, Hu, Jetcheva, 1998.
Interference: Friend or Foe?
Friend
Especially in a network setting
TX1
X1
Y4=X1+X2+X3+Z4
relay
Y3=X1+X2+Z3
X3= f(Y3)
Y5=X1+X2+X3+Z5
X2
TX2
RX2
Generalized RelayingAnalog network coding
P3
P1
Ps
D
S
P2
P4
Noisy/Compressed
Output feedback
CSI
Acknowledgements
Network/traffic information
Something else
CrossLayer Design
Error Prone
DiversityMultiplexingDelay Tradeoffs for MIMO Multihop Networks with ARQARQ
ARQ
Beamforming
H2
H1
Low Pe
Block 1
ARQ round 1
Block 1
ARQ round 2
Block 1
ARQ round 3
Block 2
ARQ round 2
Block 2
ARQ round 1
Receiver has enough
Information to decode
Block 1
ARQ round 1
Block 2
ARQ round 1
Block 2
ARQ round 2
Block 1
round 3
Block 1
ARQ round 2
Receiver has enough
Information to decode
stopping times: VBL ARQ has
the smaller outage regions among
multihop ARQ protocols
Substantial gains in throughput, efficiency, and endtoend performance from crosslayer design
B
A
Lossresilientsource codingand packetization
Application layer
Ratedistortion preamble
Congestiondistortionoptimized
scheduling
Transport layer
Congestiondistortionoptimized
routing
Traffic flows
Network layer
Capacity assignmentfor multiple service classes
Link capacities
MAC layer
Link state information
Adaptive
link layertechniques
Joint with T. Yoo, E. Setton,
X. Zhu, and B. Girod
Link layer
Fundamental Limits
of Wireless Systems
(DARPA Challenge Program)
Network Metrics
C
B
A
NetworkFundamental Limits
Capacity
Delay
D
Outage
Crosslayer Design and
Endtoend Performance
Capacity
(C*,D*,R*)
Delay
Robustness
Application Metrics
Network
Optimization
Dynamic
Programming
Game
Theory
Network Utility
Maximization
Distributed
Optimization
State Space
Reduction
Mechanism Design
Stackelberg Games
Nash Equilibrium
Wireless NUM
Multiperiod NUM
Distributed
Algorithms
*Much prior work is for wired/static networks
U1(r1)
U2(r2)
Un(rn)
Network Utility MaximizationRi
Rj
flow k
routing
Fixed link capacity
Optimization is Centralized
Knowledge
and
Complexity
IP
NCR
CR
CR
NCR
RX1
CR
RX2
NCR
Enhance capacity via cognitive relays
Cognitive relays overhear the source messages
Cognitive relays then cooperate with the transmitter in the transmission of the source messages
Cognitive Relay 1
data
Source
Cognitive Relay 2
CrossLayer Tradeoffs under Energy Constraints
Transient Energy
Circuit
MultiMode OperationTransmit, Sleep, and Transientwhere a is the amplifier efficiency and
0.1
Red: hub node
Green: relay/source
0.085
2
4
1
3
0.115
0.185
(15,0)
(0,0)
(5,0)
(10,0)
0.515
• Optimal routing uses single and multiple hops
• Link adaptation yields additional 70% energy savings
Wireless Internet access
Nth generation Cellular
Wireless Ad Hoc Networks
Sensor Networks
Wireless Entertainment
Smart Homes/Spaces
Automated Highways
All this and more…
Applications have hard delay constraints, rate requirements,
and energy constraints that must be met
These requirements are collectively called QoS
Automated Vehicles
 Cars
 UAVs
 Insect flyers
 Different design principles
 Controllers must be robust and adaptive to random delay/loss.
 Networks must be designed with control as the design objective.