Power Control in Wireless Ad Hoc Networks

1
Power Control in Wireless Ad Hoc Networks

• Crystal Jackson
• SURE 2005

2
Outline

• Motivation/Background
• System description
• Simulation model
• Results
• Conclusion and future work

3
Background/Motivation

• Ad hoc wireless network?
• Group of self configuring wireless nodes
• Lack infrastructure
• Power Control
• Important issue
• No arbiter
• Too much vs. too little
• Limited power supply

4
System Description

• Variable-length packets
• Data has variable lengths
• Larger packets require more energy
• Channel Protocol
• RTS/CTS/DATA/ACK rules

CTS
ACK
RTS
A
B
5
Signal vs. Interference
T1
R1
R2
T2
6
Signal vs. Interference
EINR (Energy to Interference plus Noise Ratio)
Received Energy
Spreading Factor
EINR gt ß ? Successful
Noise
Interference
and
where
path loss formula
7
Power Schemes

• Fixed Power
• Each node uses a fixed power, Pmax
• No calculations
• RTS, CTS and DATA
• Simple Adaptive Power
• Simple calculation
• RTS,CTS, and DATA

interference margin
depends on distance
8
Power Schemes
Distance 0.7

• Problem

Distance 2.0
16.7W
Distance 1.0
3.6W
19.5.0W
21.9W
15.0W
3.6W
16.7W
3.6W
9
Power Schemes

• Genie-Aided Adaptive Power
• Actual interference used from environment
• Iteratively adjust power based on interference
  values
• Power for each node converges to minimum
• Update power every time interference environment
  changes
• Only used for DATA transmissions

R1
T1
R2
T2


T1
RTS
DATA
R1
ACK
CTS
SAME


T2
RTS
DATA
R2
ACK
CTS
T.Elbatt and A. Ephremides, Joint Scheduling
and Power Control for Wireless Ad Hoc
Networks, IEEE Wireless Commun., vol.3, pp.
74-85, Jan. 2004.
10
Simulation Model

• Network Layer
• Queue
• First in First Out
• Maximum limit of 50 packets
• Routing
• Dijkstras algorithm to calculate routes with
  fewest relays
• Radius calculated using EINR equation
• Packet Generation
• Each node generates a packet in a slot with
  probability p
• Randomly selected destination for packet

11
Simulation Model
Diameter 2
Diameter 3
B
A
D
C
E
F
12
Results
Throughput for Fixed Power
Throughput for Simple Adaptive Power
13
Results
14
Results
Throughput for Genie-Aided Power
Throughput for Simple Adaptive Power
15
Results
16
Conclusion

• Simple Adaptive power better than fixed
• approximately same throughput
• 200 increase in throughput efficiency
• Genie-Aided Power
• significant improvement in throughput over Simple
  Adaptive
• Some increase in throughput efficiency
• Impractical to implement
• Future Work
• Improve Simple Adaptive scheme for better
  throughput

17
Acknowledgements

• Dr. Russell
• SURE Coordinators
• Dr. Noneaker
• Dr. Xu
• NSF