The Logical Meeting Point of Multiset Rewriting and Process Algebra - PowerPoint PPT Presentation

1 / 27
About This Presentation
Title:

The Logical Meeting Point of Multiset Rewriting and Process Algebra

Description:

To major premise. Emphasis on G, D and S. C is output, at best. Does not change. Possibly infinite ... For S = x1, ..., xn. De Bruijn's telescopes. Observation ... – PowerPoint PPT presentation

Number of Views:58
Avg rating:3.0/5.0
Slides: 28
Provided by: ilianoce
Category:

less

Transcript and Presenter's Notes

Title: The Logical Meeting Point of Multiset Rewriting and Process Algebra


1
The Logical Meeting Pointof Multiset
Rewritingand Process Algebra
  • Iliano Cervesato iliano_at_itd.nrl.navy.mil
  • ITT Industries, inc _at_ NRL Washington, DC
  • http//theory.stanford.edu/iliano

MFPS 20 _at_ CMU
May 25, 2004
2
Motivations
  • Security protocol specifications
  • Transition-based
  • Process-based
  • Different languages and techniques
  • Ad-hoc translations
  • Attempt at a unified approach
  • Rewriting re-interpretation of logic
  • Open derivations
  • Left rule semantics
  • Foundation of multiset rewriting
  • Bridge to process algebra
  • Effective protocol specification language

3
Linear Logic
  • Formulas
  • A, B a 1 A ? B A ?? B ! A
  • T A B ?x. A ?x. A
  • LV sequents
  • G D --gtS C
  • Constructor ,
  • Empty ?

- logic- system w- rewriting - processes-
security
Goalformula
Unrestrictedcontext
Linearcontext
Signature
4
Some LV Rules
Structural rules
G A --gtS A
Left rules
G D, A, B --gtS C G D, A?B --gtS C
G, A D , A --gtS C G, A D --gtS C
G D --gtS A G D, B --gtS C G D, D , A??B
--gtS C
Cut rules
G D --gtS A G D, A --gtS C G D,D --gtS C
S - t G D, t/xA --gtS C G D, ?x.A --gtS C
- logic- system w- rewriting - processes-
security
G ? --gtS A G, A D --gtS C G D --gtS C
G D, A --gtS,x C G D, x.A --gtS C
G, A D--gtS C G D , !A --gtS C
Right rules


5
Logical Derivations
G C --gtS C
  • Proof of C from D and G
  • Emphasis on C
  • C is input
  • Finite
  • Closed
  • Rules shown
  • Major premise
  • Preserves C
  • Minor premise
  • Starts subderivation

G D --gtS CG D --gtS C
- logic- system w- rewriting - processes-
security
G D --gtS C
6
A Rewriting Re-Interpretation
  • Transition
  • From conclusion
  • To major premise
  • Emphasis on G, D and S
  • C is output, at best
  • Does not change
  • Possibly infinite
  • Open
  • Minor premise
  • Auxiliary rewrite chain
  • Finite
  • Topped with axiom

G C --gtS C
G D --gtS CG D --gtS C
- logic- system w- rewriting - processes-
security
G D --gtS C
7
State and Transitions
  • States
  • S G D
  • S is a list
  • G and D are commutative monoids
  • No C
  • Does not change
  • Transitions
  • S G D ? S G D
  • ? for reflexive and transitive closure
  • Constructor ,
  • Empty ?

- logic- system w- rewriting - processes-
security
8
Interpreting Unary Rules
G D, A, B --gtS C G D, A?B --gtS C
S G (D, A?B ) ? S G (D, A, B)
G D, A --gtS,x C G D, x.A --gtS C
S G (D, ?x. A) ? (S, x) G (D, A)
- logic- system w- rewriting - processes-
security
G, A D --gtS C G D , !A --gtS C
  • S G (D, !A) ? S (G, A) D



9
Binary Rules and Axiom
G A --gtS A
  • Minor premise
  • Auxiliary rewrite chain
  • Top of tree
  • Focus shifts to RHS
  • Axiom rule
  • Observation

G D --gtS A G D, B --gtS C G D, D , A??B
--gtS C
- logic- system w- rewriting - processes-
security
10
Observations
G,G A --gtS,S A
G D --gtS ?S. A
  • Observation states
  • S D
  • In D, we identify
  • , with ?
  • ? with 1
  • Categorical semantics
  • Identified with ?x1. ?xn. D
  • For S x1, , xn
  • De Bruijns telescopes
  • Observation transitions
  • S G D ? S D

A
D ?D
- logic- system w- rewriting - processes-
security
S D ?S. ?D
11
Interpreting Binary Rules
G A --gtS A
S G D ? S D S G D ? S Dif S G
D ? S G Dand S G D ? S D
  • S G (D, D, A ?? B) ? S G (D, B)if S G D
    ? S A

- logic- system w- rewriting - processes-
security
S G D, D ? S G (A, D) if S G D ? S A


12
Formal Correspondence
  • Soundness
  • If S G D ? S,S Dthen G D --gtS ?S.
    ? D
  • Completeness?
  • No! We have only crippled right rules
  • ? ? a ?? b, b ?? c ? ? a ?? c

- logic- system w- rewriting - processes-
security
13
System w
  • With cut, rule for ?? can be simplified to
  • S G (D, A, A ?? B) ? S G (D, B)
  • Cut elimination holds
  • in-lining of auxiliary rewrite chains
  • But
  • Careful with extra signature symbols
  • Careful with extra persistent objects
  • No rule for ? needs a premise
  • ? does not depend on ?

- logic- system w- rewriting - processes-
security
14
Multiset Rewriting
  • Multiset set with repetitions allowed
  • a ? a, a
  • Commutative monoid
  • Multiset rewriting (a.k.a. Petri nets)
  • Rewriting within the monoid
  • Fundamental model of distributed computing
  • Alternative Process Algebras
  • Basis for security protocol spec. languages
  • MSR family
  • several others
  • Many extensions, more or less ad hoc

- logic- system w- rewriting - processes-
security
15
First-Order Multiset Rewriting
  • Multiset elements are F0 atomic formulas
  • Rules have the form
  • ?x1xn. a(x) ? ?y1yk. b(x,y)
  • Semantics
  • Several encodings into linear logic
  • Martí-Oliet, Meseguer, 91

- logic- system w- rewriting - processes-
security
S a(t), s ?R, (a(x) ? ?y. b(x,y)) S,y
b(t,y), s
if S - t
16
w-Multisets vs. Multiset Rewriting
  • MSR 1 is an instance of w-multisets
  • Uses only ?, 1, ?, ?, and ??
  • ?? never nested, always persistent
  • S s ?R S siff S R s ? S
    s
  • Interpretation of MSR as linear logic
  • Logical explanation of multiset rewriting
  • MSR is logic
  • Guideline to design rewrite systems

- logic- system w- rewriting - processes-
security
17
The Asynchronous p-Calculus
  • Another fundamental model of distributed
    computing
  • Language
  • P 0 PQ n x. P !P x(y).P
    xltygt
  • Semantics
  • Structural equivalence
  • Comm. monoidal congruence of and 0
  • Binder mobility congruence of n
  • n x. n y. P ? n y. n x. P
  • 0 ? n x. 0
  • P n x. Q ? n x. (P Q) if x ? FN(P)
  • !P ? !P P
  • Reaction law
  • xltygt x(z). P Q ?? y/zP Q

- logic- system w- rewriting - processes-
security
18
Properties
  • If P ?? Q
  • then ? ? P ? S G D
  • where Q ?S. !G ? D mod !A !A?A
  • Note with !P ? !P P as a transition
  • If P ?? Q
  • then ? ? P ? S G D
  • where Q ?S. !G ? D

- logic- system w- rewriting - processes-
security
19
w-Multisets vs. Process Algebra
  • Simple encoding of asynchronous p-calculus into
    w-multisets
  • Doesnt show that p-calculus is logic
  • Uses only a fraction of w-multiset syntax
  • Inverse encoding?
  • As hard as going from multiset rewriting to
    p-calculus
  • Other languages
  • Join calculus
  • Strand spaces
  • To do Synchronous p-calculus

- logic- system w- rewriting - processes-
security
20
MSR 3
  • Instance of w-multisets for cryptographic
    protocol specification
  • Security-relevant signature
  • Typing infrastructure
  • Modules, equations,
  • 3rd generation
  • MSR 1 First-order multiset rewriting with ?
  • Undecidability of protocol analysis
  • MSR 2 MSR 1 typing
  • Actual specification language
  • More theoretical results
  • Implementation underway

- logic- system w- rewriting - processes-
security
21
Example
  • Needham-Schroeder public-key protocol
  • A ? B nA, AkB
  • B ? A nA, nBkA
  • A ? B nBkB
  • Can be expressed in several ways
  • State-based
  • Explicit local state
  • As in MSR 2
  • Process-based embedded ?
  • Continuation-passing style
  • As in process algebra
  • (Intermediate approaches)

- logic- system w- rewriting - processes-
security
22
State-Based
A ? B nA, AkB B ? A nA, nBkA A ? B nBkB
MSR 2 spec.
  • ?A princ.
  • ?L princ ? ?Bprinc.pubK B ? nonce ? mset.
  • ?B princ. ?kB pubK B.
  • ?
  • ? ?nA nonce.
  • net (nA, AkB), L (A, B, kB, nA)
  • ?B princ. ?kB pubK B.
  • ?kA pubK A. ?kA' prvK kA.
  • ?nA nonce. ?nB nonce.
  • net (nA, nBkA), L (A, B, kB, nA)
  • ? net (nBkB)
  • Interpretation of L
  • Rule invocation
  • Implementation detail
  • Control flow
  • Local state of role
  • Explicit view
  • Important for DOS

- logic- system w- rewriting - processes-
security
23
Process-Based
A ? B nA, AkB B ? A nA, nBkA A ? B nBkB
  • ?Aprinc.
  • ?B princ. ?kB pubK B.
  • ? ? ?nA nonce.
  • net (nA, AkB),
  • (?kA pubK A. ?kA' prvK kA. ?nB
    nonce.
  • net (nA, nBkA) ? net (nBkB))

- logic- system w- rewriting - processes-
security
  • Succinct
  • Continuation-passing style
  • Rule asserts what to do next
  • Lexical control flow
  • State is implicit
  • Abstract

24
NSPK in Process Algebra
A ? B nA, AkB B ? A nA, nBkA A ? B nBkB
  • ?Aprinc.
  • ?B princ. ?kB pubK B.
  • ?kA pubK A. ?kA' prvK kA. ?nB nonce.
  • nnA nonce.
  • net (nA, AkB) .
  • net ltnA, nBkAgt .
  • net (nBkB) . 0
  • Same structure !
  • Not a coincidence
  • MSR 3 very close to Process Algebra
  • w-multiset encodings of p-calculus
  • and Join Calculus

- logic- system w- rewriting - processes-
security
  • MSR 3 is promising middle-ground for relating
  • State-based
  • Process-based
  • representations of a problem

25
State-Based vs. Process-Based
  • State-based languages
  • Multiset Rewriting
  • NRL Prot. Analyzer, CAPSL/CIL, Paulsons
    approach,
  • Statetransitionsemantics
  • Process-based languages
  • Process Algebra
  • Strand spaces, spi-calculus,
  • Independentcommunicatingthreads

- logic- system w- rewriting - processes-
security
26
MSR 3 Bridges the Gap
  • Difficult to go from one to the other
  • Different paradigms

- logic- system w- rewriting - processes-
security
State ? Process translation done once and for all
in MSR 3
27
Conclusions
  • w-multisets
  • Logical foundation of multiset rewriting
  • Relationship with process algebras
  • Unified logical view
  • Better understanding of where we are
  • Hint about where to go next
  • MSR 3.0
  • Language for security protocol specification
  • Succinct representations
  • Simpler specifications
  • Economy of reasoning
  • Bridge between
  • State-based representation
  • Process-based representation
Write a Comment
User Comments (0)
About PowerShow.com