Semantic Consistency in Information Exchange

1
Semantic Consistency in Information Exchange
Pleiades Project

• Dwork, Kannan, Lee, Lincoln, Mitchell, Rubinfeld,
  Scedrov
• Cervesato, Sokolsky, Stern

2
Consistency

• Information and executable code are understood
  and behave in same way when shared.
• Goals
• Basic agreement on form and behavior
• Resistance to faults or attacks
• Correctness when possible
• In general
• shared information may be incorrect
• shared code might give incorrect results

3
Analogy

• Consistency plug ends match
• Correctness devices configured properly
• Data in same units, but not necessarily
  correct.

4
Goal Securely transmit, monitor and check
software on the fly
5
Main Project Threads

• Network communication protocols
• Security protocol analysis
• Interactive, prob primitives based on complexity
  theory
• System design and analysis
• Design and requirements checking tools
• Specifications with real-time, probability
• Schedulability analysis
• Program monitoring and checking
• Java run-time monitoring framework
• Functional checking
• Security requirements and checking

6
Integrated Diversity

• Systems
• Protocols
• Java Applets
• Modular Systems

• Properties
• Security
• Real-time
• Safety
• Schedul-ability
• Functional correctness
• Statistics

• Techniques
• Process calculus
• Model checking
• Linear logic,tools
• Code analysis, instrumentation
• Probabilistic algorithms and complexity theory

7
MURI Multidisciplinary Project
Cornell
Stanford
Penn
SRI
IBM
Pleiades Constellation of 7
members Logic and languages
Algorithms and complexity Model
checking and real-time systems
8
Gravitational Attraction
Project evolution
Year 1
Year 2
Year 3
9
Multiple Disciplines

• Computational Complexity
• Cryptography
• Probabilistic algo-rithms and classes
• Approximation
• Functional checking
• Learning

• Logical Methods
• Games, Proof theory
• Linear logic
• Type systems and program analysis

• Real-time Systems
• Timed process algebra
• Temporal logic

10
Accomplishments

• Analysis of security protocols
• Finite-state analysis, new general framework
• Probabilistic poly-time process language
• Complexity and cryptography
• Probabilistic methods for transmitting proofs
• Concurrent zero knowledge
• Program monitoring and checking
• Framework with intermediate languages
• Broader scope for functional checking

11
Accomplishments (contd)

• System Design and Analysis
• Probabilistic extension of real-time formalism
• Abstraction methods for verification
• Incremental re-verification of adaptive systems
• Mobile code
• Java bytecode verification study
• Java bytecode modification for security

12
Further directions

• Bootstrap secure communication in hostile
  environment
• Use mobile code to securely install new comm.
  protocol from previously untrusted agent
• Increase trust in adaptive systems
• Evolving systems and monitors
• Pro-active security, reliability
• Refresh system to repair result of intrusion or
  malfunction

• Adaptive
• network

• Adaptive
• system

• Adaptive
• response

13
Outline of presentations

• Program Monitoring and Checking
• Security Protocols
• Probabilistic methods, functional checking

14
Web site

• Organizational information
• Publications to date

http//theory.stanford.edu/muri/
15
(No Transcript)