Complex Networks
and Systems
Dr Robert J. Bonneau
Program Manager
AFOSR/RSL
Air Force Research Laboratory
1
Complex Networks
Roadmap
Complex networks uses the results of the mathematical quantification of critical
information delivery to assure, manage, predict, and design Air Force networks
Local Network Research: Coding that assures information delivery and security
Network Management Research: Network protocol to maximize information flow
Global Network Research: Predict network performance and design robustness
Unified Mission Assured
Design
Dynamic, Heterogeneous,
Air Force Network
Raw Network
Data
Network
Management Research
Manage
Information Flow
Local Network
Research
Assure Critical
Information Delivery
Critical
Information
Guaranteed Delivery
Of Time Critical
Information
Communications
Networks
Diverse Types of
Networks
(Fundamental
Principles)
Mathematical
Characterization of
Network
Global Network
Research
Predict Network
Performance
2
Unified Mission Assured
Architecture
• Current networks are managed with multiple protocols depending on their taxonomy
• Air Force networks, particularly Airborne Networks are heterogeneous
• A unified network approach
pp
should adapt
p to the conditions and provide
p
design
g
principles
Design Principles
According
To Constraints
Less: Disruption Tolerant,
Less: Information Loss
Latency
More: Information Loss
Under Interference,
Observable/Controllable
Under Interference,
Observable/Controllable
More: Disruption Tolerant,
Latency
Adapt According
To Measurements
3
Fundamental Network Principles
Units of information transfer do not have to be packets – generalizing this approach to other
scientific areas allows generalized network design and analysis within constraints
- Taking this approach can lead to an integrated strategy of stable design formulation
Basic Information Unit Scales
Data
Network
Packet
Packet
Groups
Packet
Blocks
Wireless Hardware/
Social
Network Software
Modulation
Unit
Register/
g
Variable
Waveform
Ram/
Subroutine
Signal
Array
Virtual
Mem./
Program
Words
Network Design Principles
Frequency
Biological
DNA
(1/information
timescale)
Content
Co
te t
(local)
Phrases Protein
Synth.
News
Cell
Reports/
Function
Blogs
Network Policy/
Protocol
(management)
Network
Structure
(global)
Deterministic
Content
Heterogeneous
Content
Deterministic
Protocol
Heterogeneous
Protocol
Deterministic
Network
Heterogeneous
Network
Random
Content
Random
Protocol
Random
Network
Distribution
Deterministic Heterogeneous Random
Communications
C
i ti
Networks
G
General
l
Networks
Design
Included Properties
Design
Excluded Properties
Time Evolution
(Global
Properties)
Not Resourced,
Not Stable,
Not Secure
Resourced,
Resourced
Stable,
Secure
4
Comprehensive Systems Modeling
- Model heterogeneous distributed systems using unified, modular,
composable and scalable mathematical framework from previous
measurement and system specification
- Use new statistical, algebraic, and geometric representations and theory
for modularized representations and composable into a modeling
framework
System of Interest
Mission
Applications
Physical Environ.
Resource Const.
Processing
Hardware
Network
Mathematical Models
Resource Policy
Security
Framework
Database Arch.
Operating System
Prog. Languages
Design Tools
Statistical, Algebraic, Geometric, …
Unified Representation
Modular,
Composable,
Scalable Model of
Unified System
5
Measurement-Based System Verification
- Verify the properties of a given unified system through measurement of a limited
set of parameters and calculate system risk of not meeting mission requirements
- Assess
A
risk
i kb
by di
distance
t
between
b t
properties
ti off desired
d i d representation
t ti
(model) and measured properties
- Incorporate risk of sparse measurement
M
Measured
d Properties
P
ti
Mission
Requirements
Desired Properties
p
Measurement
Risk Assessment
Low
Mission
Risk
Desirable Properties:
(Example)
Robustness to Disruption
Medium
Mission
Risk
Performance Verification
High
Mission
Risk
Undesirable Properties:
(Examples)
Latency, Interference,
p
Overhead
Computational
6
System Design Trade-space
- Define general application architectural and policy design principles
g unified assessment of system
y
operating
p
g risk
through
- Apply to existing architectures through policy implementation
System Operating Trade-space
Architecturally Excluded Modalities
(high mission risk)
Architecturally Included Modalities
(low mission risk)
7
Systems Engineering Framework
- Measure and Model, for Design using a comprehensive, modular,
compositional, and scalable framework
- Models inform measurement based verification of system properties
- Strategy enables designs to quantifiably meet mission performance
objectives in heterogeneous dynamic systems
Comprehensive
System
Modeling
Mission
Performance
Objectives
System
Measurement
System
Design
System
Performance
Trade-space
Verification
Of System
Performance
8