ProtocolLearner
Class PassiveLearner

java.lang.Object
  ProtocolLearner.PassiveLearner

public class PassiveLearner
extends java.lang.Object

A passive synthesis protocol learner that builds Finite State Machine from a set of protocol messages traces

Author:

Serge Gorbunov

Nested Class Summary

class

PassiveLearner.Loop

Constructor Summary

PassiveLearner()

Method Summary

boolean	checkRootsChildrenColors(FSAutomaton inputAutomaton) Checks whether the the children of the root of the tree have associated colors.
FSAutomaton	constructAutomaton(java.util.ArrayList<MessageTrace> traces, AbstractionFunct input, AbstractionFunct output) Constructs a FSM from a sample of message traces and input/output abstraction function
int	findStateHeights(FSAutomaton inputAutomaton, int statePointer) A recursive function that sets a height for each state in the FSA.
java.util.HashMap<java.lang.Integer,java.lang.Integer>	findStatesParent(FSAutomaton inputAutomaton, int statePointer) Function traverses the FSA and stores the parent ID for each child.
FSAutomaton	minimizeAutomaton(FSAutomaton inputAutomaton, java.util.ArrayList<MessageTrace> traces, AbstractionFunct input, AbstractionFunct output) Minimizes a finite state automaton by reducing similar states And restores the loops that were initially removed from the automaton
void	setTraceLoops(MessageTrace trace, AbstractionFunct inab) Identifies loops in the trace

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

PassiveLearner

public PassiveLearner()

Method Detail

constructAutomaton

public FSAutomaton constructAutomaton(java.util.ArrayList<MessageTrace> traces,
                                      AbstractionFunct input,
                                      AbstractionFunct output)

Constructs a FSM from a sample of message traces and input/output abstraction function

Parameters:

traces - ArrayList of message traces

input - Input abstraction function

output - Output abstraction function

Returns:

Generated FSM

minimizeAutomaton

public FSAutomaton minimizeAutomaton(FSAutomaton inputAutomaton,
                                     java.util.ArrayList<MessageTrace> traces,
                                     AbstractionFunct input,
                                     AbstractionFunct output)

Minimizes a finite state automaton by reducing similar states And restores the loops that were initially removed from the automaton

Parameters:

inputAutomaton - Finite state automaton

traces - A collection of protocol traces that were used to construct the automaton

input - An abstraction function for the input messages

output - An abstraction function for the output messages

Returns:

Minimized finite state automaton

checkRootsChildrenColors

public boolean checkRootsChildrenColors(FSAutomaton inputAutomaton)

Checks whether the the children of the root of the tree have associated colors. This method is used for FSM minimization

Parameters:

inputAutomaton - FS automaton

Returns:

True if all children of the root have colors, false otherwise

findStateHeights

public int findStateHeights(FSAutomaton inputAutomaton,
                            int statePointer)

A recursive function that sets a height for each state in the FSA.

Parameters:

inputAutomaton - A finite state automaton

statePointer - Index of a state for which height is calculated.

Returns:

The height of the state

findStatesParent

public java.util.HashMap<java.lang.Integer,java.lang.Integer> findStatesParent(FSAutomaton inputAutomaton,
                                                                               int statePointer)

Function traverses the FSA and stores the parent ID for each child.

Parameters:

inputAtomaton - input FAS

Returns:

A hash map where state is the key, and its parent is the value

setTraceLoops

public void setTraceLoops(MessageTrace trace,
                          AbstractionFunct inab)

Identifies loops in the trace