Download Cognitive Radio and Wireless Trends

Analysis and Design of Cognitive
Radio Networks
and Distributed Radio Resource
Management Algorithms
James Neel
[email protected]
April 25-26, 2007
CRT
1
Cognitive Radio
© Cognitive Radio Technologies, 2007
Technologies
CRT Information
Small business officially incorporated in Feb 2007 to
commercialize cognitive radio research
Email: [email protected]
[email protected]
Website: crtwireless.com
Tel:
540-230-6012
Mailing Address:
Cognitive Radio Technologies, LLC
147 Mill Ridge Rd, Suite 119
Lynchburg, VA 24502
2
© Cognitive Radio Technologies, 2007
140
CRT’s Strengths
Frequency
Adjustments
Channel
120
100
80
60
40
0
50
100
150
200
250
300
0
50
100
150
200
250
300
0
50
100
150
iteration
200
250
300
-60
–
–
Infrastructure, mesh, and ad-hoc
networks
DFS, TPC, AIA, beamforming,
routing, topology formation
Statistics from yet to be published DFS
algorithm for ad-hoc networks
P2P 802.11a links in 0.5kmx0.5km area
(discussed in tomorrow’s slides)
I (f) (dBm)
-70
i
-80
-90
-55
-60
(f) (dBm)

Analysis of networked cognitive
Average Link
Interference
radio algorithms (game theory)
Design of low complexity, low
overhead (scalable), convergent
and stable cognitive radio
Net Interference
algorithms
-65
-70
-75
-80
-45
-50
Steady-state Interference levels (dBm)

-55
-60
-65
Typical Worst Case Without Algorithm
Average Without Algorithm
Typical Worst Case With Algorithm
Average With Algorithm
Colission Threshold
-70
-75
-80
-85
-90
3
© Cognitive Radio Technologies, 2007
0
10
20
30
40
50
60
Number Links
70
80
90
100
Selected Publications (why selected)

(First paper to propose application of game theory to cognitive
radio networks)
–

(First paper to use game theory to examine convergence
properties of cognitive radio networks)
–

J. Neel, J. Reed, and R. Gilles, “Convergence of Cognitive Radio
Networks,” Wireless Communications and Networking Conference
2004, March 21-25, 2004, vol. 4, pp. 2250-2255.
(Outstanding Paper Award)
–

J. Neel, R. Buehrer, J. Reed, and R. Gilles, “Game Theoretic Analysis
of a Network of Cognitive Radios,” Midwest Symposium on Circuits
and Systems 2002
J. Neel, J. Reed, R. Gilles, “Game Models for Cognitive Radio
Analysis,” SDR Forum 2004 Technical Conference, Nov. 2004, paper
# 01.5-05.
(Outstanding Paper Award)
–
J. Neel, J. Reed, R. Gilles, “The Role of Game Theory in the Analysis
of Software Radio Networks,” SDR Forum Technical Conference, San
Diego Nov. 11-12, 2002, paper # 04.3-001.
4
© Cognitive Radio Technologies, 2007
Selected Publications (why selected)

(Textbook chapter closely related to today’s presentation)
–

(Shorter upcoming tutorial across the pond)
–

J. Neel. J. Reed, A. MacKenzie, Cognitive Radio Network
Performance Analysis in Cognitive Radio Technology, B. Fette,
ed., Elsevier August 2006.
J. Neel, “Game Theory in the Analysis and Design of Cognitive
Radio Networks,” DySPAN07 April 17, 2007,.
(Stuff beyond Cognitive Radio Networks)
–
–
–
W. Tranter, J. Neel, and C. Anderson, Simulation of Ultra
Wideband Communication Systems, in An Introduction to Ultra
Wideband Communication Systems, Prentice Hall 2005.
J. Neel and J. Reed. Case Studies in Software Radio Design, in
Jeffrey H. Reed. Software Radios: A Modern Approach to
Radio Engineering, Prentice Hall 2002.
J. Reed, J. Neel and S. Sachindar, Analog to Digital and Digital to
Analog Conversion, in Jeffrey H. Reed, Software Radios: A
Modern Approach to Radio Engineering, Prentice Hall 2002.
5
© Cognitive Radio Technologies, 2007
Tutorial Background

Most material from my three week defense
–
–
–

Other material from training short course I gave in
summer 2003
–

Very understanding committee
Dissertation online @
http://scholar.lib.vt.edu/theses/available/etd-12082006141855/
Original defense slides @
http://www.mprg.org/people/gametheory/Meetings.shtml
http://www.mprg.org/people/gametheory/Class.shtml
Some material is new
–
Consider presentation subject to existing NDAs
6
© Cognitive Radio Technologies, 2007
Research in a nutshell

Hypothesis: Applying game theory and game models
(potential and supermodular) to the analysis of
cognitive radio interactions
–
–
–
–
–
–
Provides a natural method for modeling cognitive radio
interactions
Significantly speeds up and simplifies the analysis process
(can be performed at the undergraduate level – Senior EE)
Permits analysis without well defined decision processes (only
the goals are needed)
Can be supplemented with traditional analysis techniques
Can provides valuable insights into how to design cognitive
radio decision processes
Has wide applicability
7
© Cognitive Radio Technologies, 2007
Research in a nutshell

Focus areas:
–
–
–
Formalizing connection between game theory and
cognitive radio
Collecting relevant game model analytic results
Filling in the gaps in the models



–
–
Model identification (potential games)
Convergence
Stability
Formalizing application methodology
Developing applications
8
© Cognitive Radio Technologies, 2007
Presentation Overview
Cognitive Radio
Basic Functionality
Applications
Classifications
Modeling Cognitive Radio Behavior
Motivating Problems
Analysis Objectives
Basic Models
y
“Traditional” Analysis Insights
Evolution Functions
Contraction Mappings
x
9
© Cognitive Radio Technologies, 2007
Presentation Overview
Game Theory
Normal Form games
Mixed Strategies
Extensive Form Games
Repeated Games
Special Game Models
Supermodular games
Potential games
Algorithms
Ad-hoc power control
Sensor network formation
Interference Reducing Networks
Infrastructure DFS
Ad-hoc DFS
Joint algorithms
10
© Cognitive Radio Technologies, 2007
Approximate Timeline
Day 1 (April 25th)
Day 2 (April 26th)
08:00-08.30 Introduction
08:30-09:30 Cognitive Radio
08:00-10:00 Special Models
09:30-10:30 Models of CR Behavior
10:45-12:00 Traditional Analysis
10:15-12:00 Algorithms
13:00-15:00 Game Theory
13:15-17:00 Game Theory
13:00-17:00 Discussion
11
© Cognitive Radio Technologies, 2007
Cognitive Radio Concepts
How does a radio come
to be “cognitive”?
12
© Cognitive Radio Technologies, 2007
Cognitive Radio: Basic Idea

Software radios permit network or
user to control the operation of a
software radio
Cognitive radios enhance the control
process by adding
–
–
–
–
–
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
13
© Cognitive Radio Technologies, 2007
Waveform Software
Control Plane

Software Arch
Services
OS
Board APIs
Board package
(RF, processors)
Defining Cognitive Radio is Surprisingly
Difficult

[Mitola 1999] coined the term to refer to
–

“A radio that employs model based reasoning to achieve a
specified level of competence in radio-related domains.”
[Haykin 2005]
–
“An intelligent wireless communication system that is aware of its
surrounding environment (i.e., outside world), and uses the
methodology of understanding-by-building to learn from the
environment and adapt its internal states to statistical variations in
the incoming RF stimuli by making corresponding changes in
certain operating parameters (e.g., transmit-power, carrierfrequency, and modulation strategy) in real-time, with two primary
objectives in mind:


highly reliable communications whenever and wherever needed;
efficient utilization of the radio spectrum.
14
© Cognitive Radio Technologies, 2007
Regulatory

FCC
–

NTIA
–

“A radio that can change its transmitter parameters based
on interaction with the environment in which it operates.”
“A radio or system that senses its operational
electromagnetic environment and can dynamically and
autonomously adjust its radio operating parameters to
modify system operation, such as maximize throughput,
mitigate interference, facilitate interoperability, and access
secondary markets.”
ITU (Wp8A)
–
“A radio or system that senses and is aware of its
operational environment and can dynamically and
autonomously adjust its radio operating parameters
accordingly.”
15
© Cognitive Radio Technologies, 2007
SDR Forum Definitions

Cognitive Radio Working Group
–

“A radio that has, in some sense, (1) awareness of changes
in its environment and (2) in response to these changes
adapts its operating characteristics in some way to improve
its performance or to minimize a loss in performance.”
Cognitive Radio Special Interest Group
–
“An adaptive, multi-dimensionally aware, autonomous radio
(system) that learns from its experiences to reason, plan,
and decide future actions to meet user needs.”
16
© Cognitive Radio Technologies, 2007
IEEE

IEEE USA
–

“A radio frequency transmitter/receiver that is designed to
intelligently detect whether a particular segment of the radio
spectrum is currently in use, and to jump into (and out of, as
necessary) the temporarily-unused spectrum very rapidly, without
interfering with the transmissions of other authorized users.”
IEEE 1900.1
–
“A type of radio that can sense and autonomously reason about its
environment and adapt accordingly. This radio could employ
knowledge representation, automated reasoning and machine
learning mechanisms in establishing, conducting, or terminating
communication or networking functions with other radios. Cognitive
radios can be trained to dynamically and autonomously adjust its
operating parameters.”
17
© Cognitive Radio Technologies, 2007
Virginia Tech Cognitive Radio Working
Group

“An adaptive radio that is capable of the following:
a) awareness of its environment and its own
capabilities,
b) goal driven autonomous operation,
c) understanding or learning how its actions impact
its goal,
d) recalling and correlating past actions,
environments, and performance.”
18
© Cognitive Radio Technologies, 2007
Cognitive Radio Capability Matrix
Transmitter
Receiver
“Aware”
Environment
Goal Driven
Learn the
Environment


Haykin








IEEE 1900.1





IEEE USA






ITU-R






Mitola







NTIA







SDRF CRWG





SDRF SIG






19
VT CRWG






© Cognitive Radio Technologies, 2007
No interference
Can sense
Environment

Negotiate
Waveforms
Autonomous

“Aware”
Capabilities
Adapts
(Intelligently)
FCC
Definer











Why So Many Definitions?

People want cognitive radio to be something
completely different
–
–
–

Focus lost on implementation
–
–
–

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
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
20
© 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
21
© Cognitive Radio Technologies, 2007
Levels of Cognitive Radio Functionality
Level
22
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
Cognition Cycle
Level
0 SDR
1 Goal Driven
2 Context Aware
3 Radio Aware
4 Planning
5 Negotiating
6 Learns Environment
7 Adapts Plans
8 Adapts Protocols
Infer from Context
Orient
Establish Priority
Pre-process
Parse Stimuli
Observe
User Driven
Autonomous
(Buttons)
Outside
World
23
Infer from Radio Model
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
Negotiate
Negotiate
Protocols
States
Act
Adapted From Mitola, “Cognitive Radio for Flexible Mobile Multimedia
Communications
© Cognitive Radio
Technologies, 2007 ”, 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
 Adjust waveform as needed Parse Stimuli
Urgent
Immediate
Plan
to achieve goal
 Implement processes needed to
Learn
change waveform
Observe New
Other processes: (as needed)
States
Decide
 Adjust goals (Plan)
States
 Learn about the outside world,
User Driven
Generate “Best”
(Buttons)
Autonomous
Waveform
needs of user,…
24
Outside
World
© Cognitive Radio Technologies, 2007
Act
Allocate Resources
Initiate Processes
Negotiate Protocols
Policy-Based Radio
mask



A radio whose
operation/ adaptations
are governed by a set of
rules
Almost necessarily
coupled with cognitive
radio
Allows flexibility for
setting spectral policy to
satisfy regional
considerations
25
© Cognitive Radio Technologies, 2007
frequency
Policies
Cognitive Radio Applications
26
© Cognitive Radio Technologies, 2007
Strong Artificial Intelligence

27
Concept: Make a machine aware (conscious) of
its environment and self aware
• A complete failure (probably a good thing)
© Cognitive Radio Technologies, 2007
Weak Artificial Intelligence


Concept: Develop powerful (but limited) algorithms
that intelligently respond to sensory stimuli
Applications
–
–
–
–
–
Machine Translation
Voice Recognition
Intrusion Detection
Computer Vision
Music Composition
28
© Cognitive Radio Technologies, 2007
Implementation Classes

Weak cognitive radio

Strong cognitive radio
– Radio’s adaptations
Radio’s adaptations
determined by conscious
determined by hard
reasoning engine
coded algorithms and
– Closest approximation is
informed by observations
the ontology reasoning
– Many may not consider
cognitive radios
this to be cognitive (see
discussion related to Fig
6 in 1900.1 draft)
In general, strong cognitive radios have potential to achieve
both much better and much worse behavior in a network.
–

29
© Cognitive Radio Technologies, 2007
Weak/Procedural Cognitive Radios


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)
–

A function of the fuzzy definition
Implementations:
–
–
–
–
–
–
CWT Genetic Algorithm Radio
MPRG Neural Net Radio
Multi-dimensional hill climbing DoD LTS (Clancy)
Grambling Genetic Algorithm (Grambling)
Simulated Annealing/GA (Twente University)
Existing RRM Algorithms?
30
© Cognitive Radio Technologies, 2007
Strong/Ontological Radios


Radio’s adaptations determined
by some reasoning engine which
is guided by its ontological
knowledge base (which is
informed by observations)
Proposed Implementations:
–
–
–
CR One Model based reasoning
(Mitola)
Prolog reasoning engine (Kokar)
Policy reasoning (DARPA xG)
31
© Cognitive Radio Technologies, 2007
Intelligent 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
32
© Cognitive Radio Technologies, 2007
Example Architecture from CWT
Radio
CE-Radio Interface
Observation
Orientation
Performance API
Action
Hardware/platform API
Radio-domain cognition
WMS
Waveform
Recognizer
Radio
Performance
Monitor
Channel
Identifier
User Domain
User data security
System/Network security
Policy Domain
Search
Space
Config
Chob
Cognitive System Controller
User Model
Uob
Evolver
|(Simulated Meters) – (Actual Meters)|
Policy Model
Reg
Decision Maker
User preference
Local service facility
Security
WSGA
Security
CE-user interface
User preference
Local service facility
Radio
Resource
Monitor
DCH  max{S CH  U CH }
DU  max{SU  U U }
Actual Meters
Simulated
Meters
Initial Chromosomes
WSGA Parameters
Objectives and weights
Learning
X86/Unix
Terminal
Knowledge Base
33
Decision
Short Term Memory
Long Term Memory
WSGA Parameter Set
Regulatory Information
Models
Cognitive
System Module
© Cognitive Radio Technologies, 2007
System Chromosome
DFS in 802.16h
Decision,
Action
Service in function
Channel Availability
Check on next channel


Drafts of 802.16h
defined a generic
DFS algorithm
which implements
observation,
decision, action,
and learning
processes
Very simple
implementation
Choose
Different Channel
Observation
Available?
No
Yes
Observation
In service monitoring
of operating channel
No
Decision,
Action
Detection?
Select and change to
new available channel
in a defined time with a
max. transmission time
Learning
Yes
Stop Transmission
Start Channel Exclusion timer
Log of Channel
Availability
Channel unavailable for
Channel Exclusion time
Yes
Available?
34
Modified from Figure h1 IEEE 802.16h-06/010 Draft IEEE Standard for Local and
metropolitan area networks Part 16: Air Interface for Fixed Broadband Wireless Access
Systems Amendment for Improved Coexistence Mechanisms for License-Exempt
Operation, 2006-03-29
© Cognitive Radio Technologies, 2007
Background In service
monitoring (on nonoperational channels)
No
802.11j – Policy Based Radio
2.4 GHz


35
Explicitly opened up
Japanese spectrum
for 5 GHz operation
Part of larger effort to
force equipment to
operate based on
geographic region,
i.e., the local policy
Lower
Upper
U.S.
2.402
2.48
Europe
Japan
Spain
France
2.402
2.473
2.447
2.448
2.48
2.495
2.473
2.482
5 GHz
US
UNII Low 5.15 – 5.25 (4) 50 mW
UNII Middle 5.25 – 5.35 (4) 250 mW
UNII Upper 5.725-5.825 (4) 1 W
5.47 – 5.725 GHz released in Nov 2003
Europe
5.15-5.35 200 mW
5.47-5.725 1 W
Japan
4.9-5.091
5.15-5.25 (10 mW/MHz) unlicensed
© Cognitive Radio Technologies, 2007
802.11e – Almost Cognitive


Enhances QoS for Voice over Wireless IP (aka Voice
over WiFi ) and streaming multimedia
Anticipated changes
–
Enhanced Distributed Coordination Function (EDCF)

–
Shorter random backoffs for higher priority traffic
Hybrid coordination function (orientation)



Defines traffic classes
In contention free periods, access point controls medium
access (observation)
Stations report to access info on queue size. (Distributed
sensing)
36
© Cognitive Radio Technologies, 2007
802.11h – Unintentionally Cognitive

Dynamic Frequency Selection
(DFS)
–
Avoid radars

–

Listens and discontinues use of
a channel if a radar is present
Uniform channel utilization
Transmit Power Control (TPC)
–
–
–
–
Interference reduction
Range control
Power consumption Savings
Bounded by local regulatory
conditions
37
© Cognitive Radio Technologies, 2007
802.11h: A simple cognitive radio
Observe
Must estimate channel characteristics (TPC)
–
Must measure spectrum (DFS)
Orientation
Orient
a)
Radar present?
b)
In band with satellite??
c)
Bad channel?
d)
Other WLANs?
Observe
–
Decision
–
–
–
Implement decision
Learn
–
Learn
Change frequency
Change power
Nothing
Action
Decide
Act
Outside
World
Not in standard, but most implementations should learn the environment to
address intermittent signals
38
© Cognitive Radio Technologies, 2007
802.16h



Improved Coexistence
Mechanisms for LicenseExempt Operation
Basically, a cognitive radio
standard
Incorporates many of the
hot topics in cognitive
radio
–
–
–
–

Token based negotiation
Interference avoidance
Network collaboration
RRM databases
Coexistence with non
802.16h systems
–
Regular quiet times for
other systems to transmit
From: M. Goldhamer, “Main concepts of IEEE P802.16h / D1,” Document Number:
IEEE C802.16h-06/121r1, November 13-16, 2006.
39
© Cognitive Radio Technologies, 2007
General Cognitive Radio Policies in
802.16h





Must detect and avoid radar and other higher priority
systems
All BS synchronized to a GPS clock
All BS maintain a radio environment map (not their
name)
BS form an interference community to resolve
interference differences
All BS attempt to find unoccupied channels first
before negotiating for free spectrum
–
Separation in frequency, then separation in time
40
© Cognitive Radio Technologies, 2007
802.11h (“Weak” CR on hardware radios
– defined shortly)


Idea: Upgrade control processes to
permit use bands 802.11a devices to
operate as secondary users to radar
and satellites
Dynamic Frequency Selection (DFS)
–
Avoid radars

–

Listens and discontinues use of a
channel if a radar is present
Uniform channel utilization
Transmit Power Control (TPC)
–
–
–
–
Interference reduction
Range control
Power consumption Savings
Bounded by local regulatory
conditions
41
© Cognitive Radio Technologies, 2007
IEEE 802.22 – Planned Cognition

Wireless Regional Area Networks
(WRAN)
–
–
–

802.22 specifications
–

Devices
–
–

Master/Slave relation
–
–
42

–
Base Station (BS)
Customer Premise Equipment
(CPE)
BS is master
CPE slave
Aimed at bringing broadband access in
rural and remote areas
Takes advantage of better propagation
characteristics at VHF and low-UHF
Takes advantage of unused TV
channels that exist in these sparsely
populated areas (Opportunistic
spectrum usage)
–
–
–
–
TDD OFDMA PHY
DFS, sectorization, TPC
Policies and procedures for operation in
the VHF/UHF TV Bands between 54
MHz and 862 MHz
Target spectral efficiency: 3 bps/Hz
Point-to-multipoint system
100 km coverage radius
Max Transmit CPE 4W
© Cognitive Radio Technologies, 2007
802.22: Cognitive Aspects

Observation
–
–
–
–
–

Orientation
–

Infer type of signals that are present
Decision
–

Aided by distributed sensing (subscriber units return data to base)
Digital TV: -116 dBm over a 6 MHz channel
Analog TV: -94 dBm at the peak of the NTSC (National Television
System Committee) picture carrier
Wireless microphone: -107 dBm in a 200 kHz bandwidth.
Possibly aided by spectrum usage tables
Frequencies, modulations, power levels, antenna choice (omni and
directional)
Policies
–
–
4 W Effective Isotropic Radiated Power (EIRP)
Spectral masks, channel vacation times
43
© Cognitive Radio Technologies, 2007
The Interaction Problem
Outside
World

44

Outside world is determined by the interaction of
numerous cognitive radios
Adaptations spawn adaptations
© Cognitive Radio Technologies, 2007
Dynamic Spectrum Access Pitfall

Suppose
–

2
Without loss of
generality
–
–

g31>g21; g12>g32 ;
g23>g13
g31, g12, g23 = 1
g21, g32, g13 = 0.5
Infinite Loop!
–
4,5,1,3,2,6,4,…
3
1
Interference Characterization
Chan.
(0,0,0)
(0,0,1)
(0,1,0)
(0,1,1)
(1,0,0)
(1,0,1)
(1,1,0)
(1,1,1)
Interf.
(1.5,1.5,1.5)
(0.5,1,0)
(1,0,0.5)
(0,0.5,1)
(0,0.5,1)
(1,0,0.5)
(0.5,1,0)
(1.5,1.5,1.5)
4
5
45
0
1
2
3
© Cognitive Radio Technologies, 2007
6
7
Implications


In one out every four deployments, the
example system will enter into an infinite
loop
As network scales, probability of entering
an infinite loop goes to 1:
–
–

2 channels p  loop   1   3 / 4 
k channels p  loop   1  1  2 k 1
n C3


n Ck 1
Even for apparently simple algorithms,
ensuring convergence and stability will be
nontrivial
46
© Cognitive Radio Technologies, 2007
Locally optimal decisions that lead to globally
undesirable networks



47
Scenario: Distributed
SINR maximizing power
control in a single cluster
For each link, it is
desirable to increase
transmit power in
response to increased
interference
Steady state of network is
all nodes transmitting at
maximum power
Power
SINR
Insufficient to consider only a
single link, must consider
interaction
© 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
48
© 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
49
© Cognitive Radio Technologies, 2007
1.
focus
2.
3.
4.
5.
Steady state characterization
Steady state optimality
Convergence
Stability/Noise
Scalability
(Radio 2’s available actions)
Network Analysis Objectives
NE3
NE3
a2
NE2
NE1
NE1
a1
a1
(Radio 1’s available actions)
a3
Optimality
Stability/Noise
Convergence
Scalability
Steady
State Characterization
Are
How
these
donumber
system
initial
outcomes
variations/noise
desirable?
impact
the
impact
system
the system?
steady state?
As
of
devices
increases,
Is
itthe
possible
toconditions
predict
behavior
in
the
system?
Do
What
these
the
processes
steady
outcomes
will outcomes
maximize
lead
change
to steady
with
the
system
statevariations/noise?
conditions?
target parameters?
How
is
thestates
system
impacted?
How
many
different
are small
possible?
Is
How
convergence
long
does itaffected
take
to by
reach
system
thestates
steady
variations/noise?
state?optimal?
Do
previously
optimal
steady
remain
50
© Cognitive Radio Technologies, 2007
Modeling
51
© Cognitive Radio Technologies, 2007
General Comments on Analyzing
Cognition Cycle
0.
1.
2.
3.
4.
5.
6.
7.
No - not a CR
OK
Focus of this work
OK
OK
Probably
Ok
Ok (might even simplify)
No – unconstrained
problem
8. No – unconstrained
problem
52
© Cognitive Radio Technologies, 2007
Level
0 SDR
1 Goal Driven
2 Context Aware
3 Radio Aware
4 Planning
5 Negotiating
6 Learns Environment
7 Adapts Plans
8 Adapts Protocols
Why focus on OODA loop, i.e., why
exclude other levels?


OODA loop is implemented
now (possibly just ODA loop
as little work on context
awareness)
Changing plans
–
–

Over short intervals plans
don’t change
Messy in the general case
(work could easily
accommodate better
response equivalent goals)

Learning environment
–
–

Creation of new actions,
new goals, new decision
rules makes analysis
impossible
–
Negotiating
–
–
Could be analyzed, but
protocols fuzzy
General case left for future
work
Implies improving
observations/orientation.
Over short intervals can be
assumed away
Left for future work
–
Akin to solving a system of
unknown functions of
unknown variables
Most of this learning is
supposed to occur during
“sleep” modes

53
© Cognitive Radio Technologies, 2007
Won’t be observed during
operation
General Model (Focus on OODA Loop
Interactions)

Cognitive Radios


Set N
Particular radios, i, j
Outside
World
54
© Cognitive Radio Technologies, 2007
General Model (Focus on OODA Loop
Interactions)
Actions




Different radios may
have different
capabilities
May be constrained
by policy
Should specify each
radio’s available
actions to account
for variations
Actions for radio i
–
Ai
55
© Cognitive Radio Technologies, 2007
Act
General Model (Focus on OODA Loop
Interactions)

Decision Rules
Maps observations
to actions
–

ui:OAi
Intelligence implies
that these actions
further the radio’s
goal
–

Implies very simple,
deterministic function,
e.g., standard
interference function
ui:O
The many different
ways of doing this
merit further
discussion
Decide
56
© Cognitive Radio Technologies, 2007
Modeling Interactions (1/3)
Radio 1
Radio 2
Actions
Decision
Rules
u1
57
Actions
Decision
Rules
Action Space
Informed by
Communications
Theory
u1 ˆ1 
f :AO
Outcome Space
ˆ1, ˆ2 
© Cognitive Radio Technologies, 2007
u2 ˆ2 
u2
Modeling Interactions (2/3)



Radios implement actions, but observe outcomes.
Sometimes the mapping between outcomes and
actions is one-to-one implying f is invertible.
In this case, we can express goals and decision
rules as functions of action space.
–

Simplifies analysis
One-to-one assumption invalid in presence of noise.
58
© Cognitive Radio Technologies, 2007
Modeling Interactions (3/3)


When decisions are made
also matters and different
radios will likely make
decisions at different time
Tj – when radio j makes its
adaptations
–
–
Generally assumed to be
an infinite set
Assumed to occur at
discrete time



Consistent with DSP
implementation
T=T1T2Tn
tT

Decision timing classes
Synchronous
–

Round-robin
–
–

One at a time in order
Used in a lot of analysis
Random
–

All at once
One at a time in no order
Asynchronous
–
–
Random subset at a time
Least overhead for a
network
59
© Cognitive Radio Technologies, 2007
Cognitive Radio Network Modeling
Summary






Radios
Actions for each radio
Observed Outcome
Space
Goals
Decision Rules
Timing






i,j N, |N| = n
A=A1A2An
O
uj:O (uj:A)
dj:OAi (dj:A Ai)
T=T1T2Tn
60
© Cognitive Radio Technologies, 2007
DFS Example


Two radios
Two common channels
–


Implies 4 element action space
Both try to maximize Signal-to-Interference
Ratio
Alternate adaptations
61
© Cognitive Radio Technologies, 2007
Questions?
62
© Cognitive Radio Technologies, 2007