The Application of an Information Model to C4ISR Systems Modeling

1
The Application of an Information Model to C4ISR
Systems Modeling
Gary W. Stutts Gregory M. Roszyk Booz Allen
Hamilton Simulation Interoperability Workshop
Fall 2005 September 20-21, 2005
2
Agenda

• Background
• Survey of the C4ISR cycle
• The reality of simulating the C4ISR cycle
• The Goal of Information Modeling
• Capabilities
• Qualities and Characteristics
• Nature of Understanding and Challenges of
  representing it
• Information Model Description
• High, Low, High approach
• Application to Simulation of C4ISR

3
The C4ISR Cycle
4
TODAY Our Stove-piped World.
SIGINT
MASINT
IMINT
OSINT
HUMINT
5
Reflected in Stove-Piped Simulations

• INT Domain
• Sponsoring organizations
• Cycle Segmentation
• Sponsored by acquisition activities
• Tasking Collection
• PEDs
• Focus on Readily Quantifiable Data
• System Process Centric
• Concrete Metrics
• Data Alone Rarely Provides the Basis for
  Higher-order Analytic Utility

6
The Goal Information Modeling

• Conceptually
• Provide means to demonstrate effectiveness via
  situational awareness impact, not mere data
  satisfaction
• Provide end-to-end analytic capability by
  representing or facilitating
• Multi-intelligence synergies
• Dynamic reaction
• Evolving Information Needs
• Structurally
• Versatile Maximize degrees of freedom in
  construction
• Open ended Address full spectrum of problem
  types

7
Cognition A Two-Fold Challenge
8
Terminology

• Data
• Observation
• Neither true nor false..it just is
• Information
• Data with meaning
• Contextual
• Derives from fusion of Data

• Knowledge
• Similar to information
• Both are about meaning
• Associated with or driving action
• Used to some end
• Cyclic relationship, not linear

(KID)
9
Information Model (IM)

• Purpose
• Treats particular Information / Intelligence
  problem
• Abstract problem decomposition
• Actual or postulated analysis centers
• Scope Enterprise, Mission, Engagement
• Role in C4ISR simulation
• Provide decomposed information requirements
  leading to Requests for Information (RFI)
• Absorb / aggregate data responses to RFIs back
  into information / knowledge
• Quantify value added with respect to knowledge
  gained
• Provide adaptive decision making capability to
  C4ISR cycle

Conceptually Hierarchical Structurally A
Graph, NOT a Tree!
10
IM Building Block Agent (BBA)

• IM quantum structure
• Agent of particular sub topic
• Two Components
• Subject
• Cognition Element
• Input
• Lower order BBA
• Data
• Outputs (pulled)
• Inferred state variables

11
Subject

• Two varieties
• Attribute based
• Characteristics of an object
• Position, Composition, Temperature....
• Behavior based
• Intangible profile/conditions
• Posture, Intent (e.g. exercise vs.
  mobilization)

Entity ID Attribute/Hypothesis State
Variable Belief Distribution Perishability
12
Cognition Element

• Functionalities
• Input Receipt
• Makes inferences
• Determines belief distributions
• Changes state variables
• Bias Computation
• Dynamically Reactive
• Determines preferences among lower- order
  pursuits
• Varying independence
• Needs Definition
• Dynamically Reactive
• Issues INT-independent RFI
• Optional

Bias Computation RFI Generation
Inference Computation
13
Macroscopic View Revisited

• Continual Propagation
• Inference
• Bias
• RFI Issuance

• IM Mechanism
• Realizes Cyclic relation
• Knowledge, Information, Data
• Remember Graph not a tree
• Emergent Behaviors

14
IM Modularity

• Compatibly-defined BBAs or assemblies may be
  added and/or interchanged
• Libraries of BBAs solving particular problem
  types

15
Application to C4ISR Simulation

• True C4 integration into simulation
• Highly configurable
• Dynamically reactive
• Reflects cyclic KID relationships while
  appreciating conceptual hierarchies
• Information Paradigm
• Domain Independent
• Inference Bias a feedback loop for the full
  scope of simulation

16
Application of an IM Paradigm Makes New Analyses
Possible

• Information Strategies and CONOPS
• Value of Multi-INT strategies
• Multi-INT alternatives (A vs. B)
• System Hardware / Architectures / Processes
• Tasking, Collection, Processing, Exploitation,
  Dissemination
• Knowledge Management / Development
• Alternative treatments of a problem
• Are we asking the right question(s)?
• Algorithmic Exploration Merits of various
  Inference and Bias algorithms against different
  problem sub types
• Such Analyses Become Possible for any Software
  Able to Supply Data

17
Summary

• The Information Model
• Independent of INT domain
• Structurally simple modular
• Conceptually highly configurable extensible
• Captures cyclic Knowledge, Information, and Data
  relationship
• The Information Model in an analytic simulation
  environment
• Metrics
• Information centric, not data or product centric
• Normalizes across disparate subject domains and
  their simulators
• IM is the one medium of measure
• New versatility fields of analysis made
  possible

18
QUESTIONS ?
19
Contact Information

• Gary W. Stutts
• 703-902-5000
• 703-456-0732
• stutts_gary_at_bah.com
• Gregory M. Roszyk
• 703-902-7162
• roszyk_greg_at_bah.com
• Booz Allen Hamilton
• 8283 Greensboro Dr.
• McLean, VA 22102

20
BACKUP 1
21
BACKUP 2