Cognitive Radio
Jeff Reed
reedjh@vt.edu
reedjh@crtwireless.com
(540) 231 2972
James Neel
James.neel@crtwireless.com
(540) 230-6012
www.crtwireless.com
General Dynamics
April 9, 2007
CRT
 Cognitive Radio Technologies, 2007
Cognitive Radio
1
Technologies
Jeffrey H. Reed
•
Director, Wireless @ Virginia Tech
•
Willis G. Worcester Professor, Deputy
Director, Mobile and Portable Radio
Research Group (MPRG)
Authored book, Software Radio: A
Modern Approach to Radio Engineering
IEEE Fellow for Software Radio,
Communications Signal Processing and
Education
Industry Achievement Award from the
SDR Forum
Highly published. Co-authored – 2 books,
edited – 7 books.
Previous and Ongoing CR projects from
•
•
•
•
•
– ETRI, ONR, ARO, Tektronix
•
Email: reedjh@vt.edu
2
 Cognitive Radio Technologies, 2007
James Neel
• President, Cognitive Radio Technologies,
LLC
• PhD, Virginia Tech 2006
• Textbook chapters on:
– Cognitive Network Analysis in
– Data Converters in Software Radio: A
Modern Approach to Radio Engineering
– SDR Case Studies in Software Radio: A
Modern Approach to Radio Engineering
– UWB Simulation Methodologies in An
Introduction to Ultra Wideband
Communication Systems
• SDR Forum Paper Awards for 2002, 2004
papers on analyzing/designing cognitive
radio networks
• Email: james.neel@crtwireless.com
 Cognitive Radio Technologies, 2007
CRT
Cognitive Radio
Technologies3
Overview of Presentation
Material (1/2)
Presenter
Material
Reed
1.5 hrs
0830-1000
1.Introducing Cognitive Radio
1.1 What is a Cognitive Radio?
1.2 Relationship between CR and SDR
1.3 Typical Commercial CR Applications
1.4 How does CR Relate to WANN and future military networks?
1.5 Overview of Implementation Approaches
1.6 Overview of Networking Approaches
2. Implementing a Cognitive Radio
2.1Architectural Approaches
Break
~20min
1000-1020
Neel
~ 1.5 hrs
1020-11:50
Break
2.2 Observing the Environment
2.2.1 Autonomous Sensing
2.2.2 Collaborative Sensing
2.2.3 Radio Environment Maps and Observation Databases
2.3 Recognizing Patterns
2.3.1 Neural Nets
2.3.2 Hidden Markov Model
2.3.3 Ontological Reasoning
2.4 Making Decisions
2.4.1 Common Heuristic Approaches
2.4.2 Case-based Reasoning
4
 Cognitive Radio Technologies, 2007
Overview of Presentation
Material (2/2)
Presenter
Material
Lunch ~ 40min
1150-1230
Lunch Break
Reed
~ 1 hr
1230-1330
2.4 Helping a Machine Learn
2.5 Representing Information
2.6 Current Implementations including VT’s Prototypes
Neel
~ 1.0 hrs
1330-1430
3. Networking Cognitive Radios
3.1 The Interactive Problem
3.2 The Role of Policy in Networked Cognitive Radios
Break ~ 20min
1430-1450
Break
Neel
~ 0.5 hrs
1450-1520
Reed
~ 0.6 hrs
1520-1600
3.3 Approaches to Designing Well-behaved Cognitive Radio Networks
3.4 Emerging Standards
4. Summary and Conclusions
4.1 Outstanding Research Issues
4.2 The Opportunities
4.3 Speculation on How the Future May Evolve
 Cognitive Radio Technologies, 2007
5
What is a Cognitive Radio?
Concepts, Definitions
6
 Cognitive Radio Technologies, 2007
Cognitive Radio: Basic Idea
–
–
–
–
Intelligent, autonomous control of the radio
An ability to sense the environment
Goal driven operation
Processes for learning about
environmental parameters
– Awareness of its environment
• Signals
• Channels
– Awareness of capabilities of the radio
– An ability to negotiate waveforms with
other radios
 Cognitive Radio Technologies, 2007
Waveform Software
Control Plane
• Software radios permit network or
user to control the operation of a
software radio
• Cognitive radios enhance the control
process by adding
Software Arch
Services
OS
Board APIs
Board package
(RF, processors)
7
Cognitive Radio Capability
Matrix
Can sense
Environment
Transmitter
Receiver
“Aware”
Environment
Goal Driven


Haykin







IEEE 1900.1





IEEE USA






ITU-R






Mitola







NTIA







SDRF CRWG





SDRF SIG
VT CRWG







 Cognitive
 Radio Technologies,

2007
No interference
Autonomous

Negotiate
Waveforms
“Aware”
Capabilities
Learn the
Environment
Adapts
(Intelligently)

Definer
FCC














8
Why So Many Definitions?
• People want cognitive radio to be something
completely different
– Wary of setting the hype bar too low
– Cognitive radio evolves existing capabilities
– Like software radio, benefit comes from the paradigm shift in
designing radios
• Focus lost on implementation
– Wary of setting the hype bar too high
– Cognitive is a very value-laden term in the AI community
– Will the radio be conscious?
• Too much focus on applications
– Core capability: radio adapts in response changing operating
conditions based on observations and/or experience
– Conceptually, cognitive radio is a magic box
9
 Cognitive Radio Technologies, 2007
Used cognitive radio
definition
• A cognitive radio is a radio whose control processes
permit the radio to leverage situational knowledge
and intelligent processing to autonomously adapt
towards some goal.
• Intelligence as defined by [American Heritage_00] as
“The capacity to acquire and apply knowledge,
especially toward a purposeful goal.”
– To eliminate some of the mess, I would love to just call
cognitive radio, “intelligent” radio, i.e.,
– a radio with the capacity to acquire and apply knowledge
especially toward a purposeful goal
10
 Cognitive Radio Technologies, 2007
Levels of Cognitive Radio
Functionality
Level
Capability
Comments
0
Pre-programmed
A software radio
1
Goal Driven
Chooses Waveform According to Goal. Requires
Environment Awareness.
2
Context Awareness
Knowledge of What the User is Trying to Do
3
Radio Aware
Knowledge of Radio and Network Components,
Environment Models
4
Capable of Planning
Analyze Situation (Level 2& 3) to Determine Goals
(QoS, power), Follows Prescribed Plans
5
Conducts Negotiations
Settle on a Plan with Another Radio
6
Learns Environment
Autonomously Determines Structure of
Environment
7
Adapts Plans
Generates New Goals
8
Adapts Protocols
Proposes and Negotiates New Protocols
Adapted From Table 4-1Mitola, “Cognitive Radio: An Integrated Agent Architecture for Software Defined Radio,” PhD Dissertation
Royal Institute of Technology, Sweden, May 2000.
 Cognitive Radio Technologies, 2007
11
Cognition Cycle
0
1
2
3
4
5
6
7
8
Level
SDR
Goal Driven
Context Aware
Radio Aware
Planning
Negotiating
Learns Environment
Adapts Plans
Adapts Protocols
Infer from Context
Infer from Radio Model
Orient
Establish Priority
Pre-process
Parse Stimuli
Observe
User Driven
Autonomous
(Buttons)
Immediate
Select Alternate
Generate
Normal
Goals
Normal
Urgent
Plan
Learn
New
States
Decide
Determine “Best”
Plan
Determine
“Best”
Generate “Best”
Waveform
Waveform
Allocate ResourcesKnown
Initiate Processes
12
Negotiate
Negotiate
Protocols
States
Act
Outside
World
 Cognitive Radio Technologies, 2007
Adapted From Mitola, “Cognitive Radio for Flexible Mobile Multimedia Communications ”, IEEE Mobile Multimedia Conference, 1999, pp 3-10.
Conceptual Operation
Cognition cycle [Mitola_99]
OODA Loop: (continuously)
Infer from Context
• Observe outside world
Orient Infer from Radio Model
• Orient to infer meaning of
Establish Priority
observations
Normal
Pre-process
Select Alternate
Goals
Parse Stimuli
• Adjust waveform as
Urgent
Immediate
Plan
needed to achieve goal
• Implement processes
needed to change
Learn
Observe New
waveform
States
Decide
Other processes: (as
needed)
States
User Driven
Generate “Best”
• Adjust goals (Plan) Autonomous (Buttons)
Waveform
• Learn about the outside
Act
world, needs of user,… Outside
Allocate Resources
13
Initiate Processes
World
 Cognitive Radio Technologies, 2007
Negotiate Protocols
Relationship Between SDR
and CR
Cognitive radio
is a revolutionary
evolution of
software radio
14
 Cognitive Radio Technologies, 2007
Cognitive Radio & SDR
• SDR’s impact on the wireless world is difficult to predict
– “But what…is it good for?”
• Engineer at the Advanced Computing Systems Division of
IBM, 1968, commenting on the microchip
• Some believe SDR is not necessary for cognitive radio
– Cognition is a function of higher-layer application
• Cognitive radio without SDR is limited
– Underlying radio should be highly adaptive
• Wide QoS range
• Better suited to deal with new standards
– Resistance to obsolescence
• Better suited for cross-layer optimization
15
 Cognitive Radio Technologies, 2007
How is a Software Radio Different
from Other Radios? - Application
Conventional
Radio
• Supports a fixed
number of
systems
• Reconfigurability
decided at the
time of design
• May support
multiple services,
but chosen at the
time of design
Software Radio
• Dynamically
support multiple
variable systems,
protocols and
interfaces
• Interface with
diverse systems
• Provide a wide
range of services
with variable QoS
Cognitive Radio
• Can create new
waveforms on its
own
• Can negotiate new
interfaces
• Adjusts operations
to meet the QoS
required by the
application for the
signal environment
16
 Cognitive Radio Technologies, 2007
How is a Software Radio Different
from Other Radios?- Design
Conventional
Radio
• Traditional RF
Design
• Traditional
Baseband Design
Software Radio
• Conventional
Radio +
• Software
Architecture
• Reconfigurability
• Provisions for
easy upgrades
Cognitive Radio
•
•
•
•
•
SDR +
Intelligence
Awareness
Learning
Observations
17
 Cognitive Radio Technologies, 2007
How is a Software Radio Different
from Other Radios? - Upgrade Cycle
Conventional
Radio
• Cannot be made
“future proof”
• Typically radios
are not
upgradeable
Software Radio
• Ideally software
radios could be
“future proof”
• Many different
external upgrade
mechanisms
Cognitive Radio
• SDR upgrade
mechanisms
• Internal upgrades
• Collaborative
upgrades
– Over-the-Air
(OTA)
18
 Cognitive Radio Technologies, 2007
Typical Cognitive Radio
Applications
What does
cognitive
radio enable?
19
 Cognitive Radio Technologies, 2007
Bandwidth isn’t scarce,
it’s underutilized
Measurements averaged
over six locations:
1. Riverbend Park, Great
2.
3.
4.
5.
6.
Falls, VA,
Tysons Corner, VA,
NSF Roof, Arlington, VA,
New York City, NY
NRAO, Greenbank, WV,
SSC Roof, Vienna, VA
~25% occupancy at
peak
20
Modified from Figure 1 in Published August 15, 2005 M. McHenry
in “NSF Spectrum Occupancy Measurements Project Summary”, Aug 15,
 Cognitive Radio Technologies, 2007
2005. Available online: http://www.sharedspectrum.com/?section=nsf_measurements
Conceptual example of
opportunistic spectrum utilization
Primary Signals
Random
Access
TDMA
21
 Cognitive Radio Technologies, 2007
Opportunistic Signals
Cognitive radio permits the
deployment of cheaper radios
• RF components are expensive
• Cheaper analog implies more
spurs and out-of-band
emissions
• Processing is cheap and getting
cheaper
• Cognitive radios will adapt
around spurs (just another
interference source) or teach
the radio to reduce the spurs
• Better radios results in still more
available spectrum as the need
arises.
• Likely able to exploit SDR
 Cognitive Radio Technologies, 2007
22
Improved Link Reliability
• Cognitive radio is aware of
areas with a bad signal
• Can learn the location of the
bad signal
– Has “insight”
• Radio takes action to
compensate for loss of signal
– Actions available:
• Power, bandwidth, coding,
channel, form an ad-hoc
network
Signal Quality
Good
Transitional
Poor
– Radio learns best course of
action from situation
 Can aid cellular system
 Inform system & other radios of identified gaps
 Cognitive Radio Technologies, 2007
23
Automated Interoperability
• Basic SDR idea
– Use a SDR as a gateway to
translate between different
radios
• Problems
– Which devices are present?
– Which links to support?
– With SDR some network
administrator must answer
these questions
• Basic CR idea
– Let the cognitive radio observe
and learn from its environment
in an automated fashion.
24
 Cognitive Radio Technologies, 2007
Spectrum Trading
• Underutilized spectrum
can be sold to support a
high demand service
– Currently done in Britain
– Permitted in US among
public safety users
• Currently has a very long
time scale (months)
• Faster spectrum trading
could permit for significant
increases in available
bandwidth
– How to recognize need and
availability of additional
spectrum?
– Environment + context
awareness + memory Cognitive Radio Technologies, 2007
25
Collaborative Radio
• A radio that leverages the
services of other radios to
further its goals or the
goals of the networks.
• Cognitive radio enables
the collaboration process
– Identify potential
collaborators
– Implies observations
processes
• Classes of collaboration
– Distributed processing
– Distributed sensing
26
 Cognitive Radio Technologies, 2007
Cooperative Antenna Arrays
• Concept:
– Leverage other radios to
effect an antenna array
Cooperative MIMO
First Hop
Second Hop
• Applications:
– Extended vehicular coverage
– Backbone comm. for mesh
networks
Source Cluster Relay cluster Destination Cluster
– Range extension with cheaper
devices
Transmit Diversity
• Issues:
– Timing, mobility
– Coordination
– Overhead
destination
 Cognitive Radio Technologies, 2007
source
27
Other Opportunities for
Collaborative Radio (1/3)
• Distributed processing
– Exploit different
capabilities on different
devices
• Maybe even for waveform
processing
– Bring extra
computational power to
bear on critical problems
• Useful for most
collaborative problems
• Collaborative sensing
– Extend detection range by
including observations of
other radios
• Hidden node mitigation
– Improve estimation statistics
by incorporating more
independent observations
– Immediate applicability in
802.22, likely useful in future
adaptive standards
28
 Cognitive Radio Technologies, 2007
Other Opportunities for
Collaborative Radio (2/3)
• Improved localization
– Application of
collaborative sensing
– Security
– Friend finders
• Reduced contention
MACs
– Collaborative
scheduling algorithms
to reduce collisions
– Perhaps of most value
to 802.11
• Some scheduling
included in 802.11e
29
 Cognitive Radio Technologies, 2007
Other Opportunities for
Collaborative Radio (3/3)
• Distributed mapping
• Theft detection
– Gather information relevant to
specific locations from mobiles
and arrange into useful maps
– Coverage maps
• Collect and integrate signal
strength information from mobiles
• If holes are identified and fixed,
should be a service differentiator
– Congestion maps
– Devices can learn which
other devices they tend to
operate in proximity of and
unexpected combinations
could serve as a security
flag (like flagging
unexpected uses of credit
cards)
– Examples:
• Density of mobiles should
correlate with traffic (as in
automobile) congestion
• Customers may be willing to pay
for real time traffic information
 Cognitive Radio Technologies, 2007
• Car components that expect
to see certain mobiles in the
car
• Laptops that expect to
operate with specific
mobiles nearby
30
Cognitive Radio and Military
Networks
How is the military
planning on using
cognitive radio?
31
 Cognitive Radio Technologies, 2007
Drivers in Commercial and
Military Networks
•
Many of the same commercial
applications also apply to military
networks
–
–
–
–
–
•
Opportunistic spectrum utilization
Improved link reliability
Automated interoperability
Cheaper radios
Collaborative networks
Military has much greater need for
advanced networking techniques
– MANETs and infrastructure-less
networks
– Disruption tolerant
– Dynamic distribution of services
– Energy constrained devices
•
Goal is to intelligently adapt device,
link, and network parameters to
help achieve mission objectives
32
 Cognitive Radio Technologies,
From:2007
P. Marshall,
“WNaN Adaptive Network Development (WAND)
BAA 07-07 Proposers’ Day”, Feb 27, 2007
Wireless Network after Next
(WNaN)
Program Organization
Reliability through frequency and path diversity
Intelligent agent cross-layer optimization
33
 Cognitive Radio Technologies, 2007
Figures from: P. Marshall, “WNaN Adaptive Network Development (WAND) BAA 07-07 Proposers’ Day”, Feb 27, 2007
DARPA’s WNAN Program
• Objectives
– Reduced cost via intelligent
adaptation
– Greater node density
– Gains in throughput/scalability
WNaN Protocol Stack
Optimizing
Topology
• Leveraged programs
– Control Based MANET – low
Network
overhead protocols
– Microsystems Technology Office
– RFMEMS, Hermit, ASP
MAC
– xG – opportunistic use of
spectrum
– Mobile Network MIMO - MIMO
Physical
Wideband Network Waveform
– Connectionless Networks –
rapid link acquisition
– Disruption Tolerant Networks
(DTN) – network layer protocols
 Cognitive Radio Technologies, 2007
Legend
CBMANET
WNaN
CBMANET
WNaN
CBMANET
MIMO (MNM)
xG
COTS
MEMS (MTO)
Other
programs
WNaN34
program
Overview of Implementation
Approaches
How does the
radio become
cognitive?
35
 Cognitive Radio Technologies, 2007
Implementation Classes
• Weak cognitive radio
– Radio’s adaptations
determined by hard coded
algorithms and informed by
observations
– Many may not consider this
to be cognitive (see
discussion related to Fig 6
in 1900.1 draft)

• Strong cognitive radio
– Radio’s adaptations
determined by conscious
reasoning
– Closest approximation is
the ontology reasoning
cognitive radios
In general, strong cognitive radios have potential to achieve
both much better and much worse behavior in a network, but
may not be realizable. Cognitive Radio Technologies, 2007
36
Brilliant Algorithms and
Cognitive Engines
• Most research focuses on
development of
algorithms for:
–
–
–
–
–
Observation
Decision processes
Learning
Policy
Context Awareness
• Some complete OODA
loop algorithms
• In general different
algorithms will perform
better in different
situations
• Cognitive engine can be
viewed as a software
architecture
• Provides structure for
incorporating and
interfacing different
algorithms
• Mechanism for sharing
information across
algorithms
• No current
implementation standard
37
 Cognitive Radio Technologies, 2007
Observation Sources
Information is about
How the cognitive
radio gets the
information?
Other opportunities to
get information
Environment
(physical quantities, position,
situations)
•Measures
temperature, light
level, humidity, …
• Receives GPS signals to
determine position
• Parses short-range wireless
broadcasts in buildings or
urban areas for mapped
environment
• Observes the network for e.g.
weather forecast, reported
traffic jams, …etc.
Spectrum
(communication
opportunities)
• Passively "listens" to the
spectrum
• Performs channel quality
estimation
• Spectrum information is
provided by the network
• Spectrum information is
shared by other cognitive
radios
User
• Observes user's applications,
incoming/ outgoing data
streams
• Performs speech analysis
 Cognitive Radio Technologies, 2007
38
Orientation Processes
• Gives radio significance of observations
– Does multipath profile correspond to a known
location?
– Really just hypotheses testing
• Algorithms
–
–
–
–
–
Data mining
Hidden Markov Models
Neural Nets
Fuzzy Logic
Ontological Reasoning
39
 Cognitive Radio Technologies, 2007
Decision Processes
• Purpose: Map what radio believes about
network state to an adaptation
• Guided by radio goal and constrained by policy
– May be supplemented with model of real world
• Common algorithms (mostly heuristics)
–
–
–
–
Genetic algorithms
Simulated annealing
Local search
Case based reasoning
40
 Cognitive Radio Technologies, 2007
Learning Processes
• Informs radio when situation is not like one its seen before or if
situation does not correspond to any known situation
• Logically, just an extension to the orientation process with
– a “none of the above” option
– Increase number of hypotheses after “none of the above”
– Refine hypotheses and models
• Algorithms:
–
–
–
–
–
–
–
Data mining
Hidden Markov Models
Neural Nets
Fuzzy Logic
Ontological Reasoning
Case based learning
Bayesian learning
• Other proposed learning tasks
– New actions, new decision rules, new channel models, new goals, new
internal algorithms
41
 Cognitive Radio Technologies, 2007
Knowledge Representation
• Issue:
– How are radios “aware” of
their environment and how
do they learn from each
other?
• Technical refinement:
– “Thinking” implies some
language for thought.
• Proposed languages:
– Radio Knowledge
Representation Language
– XML
– Web-based Ontology
Language (OWL)
42
 Cognitive Radio Technologies, 2007
Overview of Cognitive
Networking
What happens when
they leave the lab?
43
 Cognitive Radio Technologies, 2007
The Interaction Problem
Outside
World
• Outside world is determined by the interaction
of numerous cognitive radios
44
• Adaptations spawn adaptations
 Cognitive Radio Technologies, 2007
Potential Problems with
Networked Cognitive Radios
Distributed
•
•
•
•
•
Centralized
Infinite recursions
Instability (chaos)
Vicious cycles
Adaptation collisions
Equitable distribution of
resources
• Byzantine failure
• Information distribution
•
•
•
•
Signaling Overhead
Complexity
Responsiveness
Single point of failure
45
 Cognitive Radio Technologies, 2007
Implications
• Best of All Possible Worlds
– Low complexity distributed algorithms with low anarchy factors
• Reality implies mix of methods
– Hodgepodge of mixed solutions
• Policy – bounds the price of anarchy
• Utility adjustments – align distributed solution with centralized
solution
• Market methods – sometimes distributed, sometimes centralized
• Punishment – sometimes centralized, sometimes distributed,
sometimes both
• Radio environment maps –”centralized” information for distributed
decision processes
– Fully distributed
• Potential game design – really, the Panglossian solution, but only
applies to particular problems
46
 Cognitive Radio Technologies, 2007
Cognitive Networks
• Rather than having
intelligence reside in a
single device, intelligence
can reside in the network
• Effectively the same as a
centralized approach
• Gives greater scope to the
available adaptations
– Topology, routing
– Conceptually permits
adaptation of core and edge
devices
• Can be combined with
cognitive radio for mix of
capabilities
• Focus of E2R program
R. Thomas et al., “Cognitive networks: adaptation and learning to achieve
end-to-end performance objectives,” IEEE Communications Magazine, Dec.
2006
 Cognitive Radio Technologies, 2007
47
Emerging Commercial
Implementations
• Dynamic Frequency
Selection
– 802.11h
– 802.11y
– 802.11 for TV bands?
• Distributed
Collaboration
– 802.16h
• Collaborative Sensing
– 802.22
• Radio Resource
Maps
– 802.16h
– 802.11y
• Policy radios
– 802.11e
– 802.11j
48
 Cognitive Radio Technologies, 2007
Summary
• Cognitive radio evolves the
software radio concept to permit
intelligent autonomous
adaptation of radio parameters
– Significant variation in definitions
of “cognitive radio”
– Question of how “cognitive” the
radio is
• Numerous new applications
enabled
– Opportunistic spectrum
utilization, collaborative radio,
link reliability, advanced network
structures
• Differing implementation
approaches
• Many objectives will require
development of a cognitive
language
• In a network, adaptations of
cognitive radios interact
– Interaction can be mitigated
with policy, punishment, cost
adjustments, centralization or
potential games
• Commercial implementations
starting to appear
– 802.22, 802.11h,y, 802.16h
– And may have been around for
a while (cordless phones with
DFS)
– Many applications
implementable with simple
algorithms
– Greater flexibility achievable with
 Cognitive Radio Technologies, 2007
a cognitive engine approach
49