Transcript HYBRID FSO/RF LINKS AND NETWORKS WITH DIVERSITY CONTROL

HYBRID FSO/RF LINKS AND
NETWORKS WITH DIVERSITY
CONTROL
Christopher C. Davis
The Maryland Optics Group
Department of Electrical and Computer Engineering
University of Maryland, College Park, MD 20742
February 17
Wireless Communications
The Maryland Optics Group
ACKNOWLEDGEMENTS
• Dr. Stuart D. Milner – Department of Civil and
Environmental Engineering
• Dr. Igor Smolyaninov, Department of Electrical and
Computer Engineering
• Dr Quirino Balzano, Department of Electrical and
Computer Engineering
• Professor Kyuman Cho (Sogang University, Seoul,
KOREA)
• Pam Clark, ITT
• Linda Wasiczko, Sugianto Trisno, Jaime Llorca, TzungHsien Ho, Heba El-Erian, Aniket Desai, Clint Edwards,
(graduate students)
• AFOSR, DARPA,NSA, ARL, Army CECOM
February 17
Wireless Communications
The Maryland Optics Group
WI-FI
• The current “hot topic”
• Its growing popularity will cause its demise
– Spectral overcrowding
– Lack of security
– Interference with other users and equipment
– Remember CB radio?
• But… if you are mobile you can’t be
connected by wires
February 17
Wireless Communications
The Maryland Optics Group
Modified from a TeraBeam picture
Dynamic, Reconfigurable
Hybrid FSO/RF Wireless Networks
February 17
Wireless Communications
The Maryland Optics Group
Hybrid FSO/RF Wireless Networks – WHY?
• RF wireless networks
– Broadcast RF networks are not scaleable
– RF cannot provide very high data rates
– RF is not physically secure
• High probability of detection/intercept
– Not badly affected by fog and snow, affected by rain
• Optical wireless networks
– Very high data rates
• 2.5Gb/s commercially available
• 1Tb/s demonstrated
– Almost zero probability of detection/intercept
– Affected by fog and snow
February 17
Wireless Communications
The Maryland Optics Group
Hybrid FSO/RF Wireless Networks – WHY?
• Deal with the non-acceptance of optical wireless
alone
• High availability (>99.99%)
• Much higher goodput than RF alone
• Last/First Mile Solution
• FSO is not regulated by the FCC
– must be eyesafe
• For greatest flexibility need unlicensed RF band
• Installed optical fiber – up to $1M/mile
February 17
Wireless Communications
The Maryland Optics Group
A Hybrid FSO/RF Link Handles Weather
A Hybrid FSO/RF Network Involves Disparate Data Rates
AVERAGE DATA TRANSFER RATE
OF HYBRID FSO/RF LINK
AVERAGE DATA RATE (Gb/s)
3
FSO 2.5Gb/s
2
1
RF 10Mb/s
0
0
10
20
30
40
50
60
70
80
90
100
FSO LINK AVAILABILITY (%)
February 17
Wireless Communications
The Maryland Optics Group
Challenges and Developments
• FSO is available commercially
– has not been widely accepted
– most systems do not do pointing, acquisition,
and tracking (PAT)
– most systems are not FSO/RF Hybrids
• FSO/RF Hybrid networks are in the R&D
stage
• High performance PAT must be developed
February 17
Wireless Communications
The Maryland Optics Group
Challenges and Developments (2)
• Many applications of FSO/RF networks
involve dynamic situations
– Reconfigurability (topology control) is required
– Diversity of links (transmitter and receivers)
– Changeover algorithms
– Network optimization
• DoD applications
February 17
Wireless Communications
The Maryland Optics Group
February 17
Wireless Communications
The Maryland Optics Group
DYNAMIC AND VOLATILE ATMOSPHERIC
AND PLATFORM EFFECTS
February 17
Wireless Communications
The Maryland Optics Group
OPTICAL WIRELESS
TRANSCEIVER
February 17
Wireless Communications
The Maryland Optics Group
OMNIDIRECTIONAL OPTICAL
WIRELESS TRANSCEIVER
February 17
Wireless Communications
The Maryland Optics Group
Topology Control
in Optical Wireless Networks
Network Layer
Topology Control
Link Layer
•Autonomous Backbone
Reconfiguration
•Pointing, Acquisition and
Tracking
Physical Layer
February 17
Wireless Communications
The Maryland Optics Group
Pointing, Acquisition, and Tracking
in Optical Wireless Networks
• Allows wireless links to be established and
maintained between moving platforms
• Maintains alignment of optical wireless
links
• Required for autonomous reconfiguration
and topology control in optical wireless
networks
February 17
Wireless Communications
The Maryland Optics Group
Agile Optical Wireless Transceiver
and Motorized Platform



Data rate: 155Mb/s
High speed (800K
steps per second),
resolution and pointing
accuracy up to
0.00072° per step
Fish-eye lens (180°)
used to identify and
track neighbor nodes
(beacons)
February 17
Wireless Communications
The Maryland Optics Group
Bi-Static Transceiver Design
Mono-static
Advantages: Reduces the complexity of PAT process
Disadvantages: Power isolation problem (TX/RX feedback)
Bi-static
Advantages: No power isolation problem
Disadvantages:
1. Extra alignment process required to obtain parallel axes
2. Potential misalignment in short-distance application
February 17
Wireless Communications
The Maryland Optics Group
Link Failures between 2 Transceivers
For large
application
distance
For short
application
distance
February 17
Wireless Communications
The Maryland Optics Group
PAT Process
Transceiver
Axis
Alignment
Step
Select the desired target
from the CCD image
System
Scanning
Step
Using TCP/IP socket to
check link availability
Link Table
Update
Record the current [θ,φ]
into the link table
Acquisition
Process
Object still exists
Tracking
Process
Motion Prediction Analysis
(Track beacon)
Tracking
Process
Object disappears
February 17
Wireless Communications
The Maryland Optics Group
Experimental Setup
1. Study the performance of the link with
respect to link closure latency for different
motor parameters
2. To investigate the effects of larger FOV of
our system
February 17
Wireless Communications
The Maryland Optics Group
FEATURES OF OUR CURRENT
OPTICAL WIRELESS SYSTEMS
•
•
•
•
•
•
•
Bistatic TX/RX systems
1.3m and 1.55m transmitters
CPC and lens based receivers
Fast aspheric lens receivers
Cassegrain and Fresnel lens receivers
Rugged alignment stages
Topology control
February 17
Wireless Communications
The Maryland Optics Group
OUR NEW CONCEPTS
AND THEIR IMPACT
• Maximally efficient use of high data rate FSO and RF
communication modes
• Network and link recovery everywhere through
communication mode diversity and autonomous Physical
and logical reconfigurability
• Reduced GTT due to instantaneous network recovery
• Physical reconfigurability assures > 99% availability
– Higher optical availability increases MDR
• Seamless diversity control between optical and RF
communication
• Internet-like software fully portable to DoD systems
• Network software is independent of terminal design
specifics
February 17
Wireless Communications
The Maryland Optics Group
INNOVATION
• Intelligent Aperture Diversity and Media Controller
– “Smart” identification of RF/FSO availability at each RX/TX
– Dynamic allocation of FSO/RF
• Autonomous physical and logical reconfiguration
– “Make before break” dissemination of topologies using high availability
RF control channel
• Enhanced TCP/IP protocol suite for Hybrid FSO/RF Networks
– Multi-Protocol Label Switching (Traffic Engineering) exploits media
diversity
– Proxy software provides instantaneous reaction to physical change in
topology
– Autonomous reconfigurability integrated with TCP/IP suite
• Comprehensive network modeling and simulation
– Advanced atmospheric propagation modeling (turbulence, aerosols,
obscuration)
– Discrete Event Simulation for Hybrid Networks to aid implementation
planning
February 17
Wireless Communications
The Maryland Optics Group
BACKUP SLIDES
February 17
Wireless Communications
The Maryland Optics Group
TX
TX
RX
RX
Bistatic optical wireless link
February 17
Wireless Communications
The Maryland Optics Group
The DARPA ORCLE PROGRAM
(formerly THOR Program)
• Long range (up to 100km) high altitude
(10km) laser communication links
• Rytov variance is  ln2 I  1.23Cn2k 7 / 6 L11 / 6
• 2lnI Ranges from 10 to 100
• Small Cn2, but long range makes this a strong
turbulence situation
• May be strong boundary layer turbulence at
transmitter and receivers
February 17
Wireless Communications
The Maryland Optics Group
Many Link Physics
and Engineering Issues
• Turbulence
– Variations with height
• Obscuration
– Optical depth
– Spatial distribution
• Aerosols
• Aperture averaging
• Transceiver optimization
February 17
Wireless Communications
The Maryland Optics Group
RYTOV VARIANCE FOR A 100km LINK
102
1.3micrometer laser
7
6
5
4
3
Rytov Variance
2
101
7
6
5
4
3
2
100
7
6
5
4
3
2
1.00e-18
2
3
4
5 6
1.00e-17
2
3
4
5 6
1.00e-16
2
3
4
5 6
1.00e-15
Cn2 m-2/3
February 17
Wireless Communications
The Maryland Optics Group