Information Technology Reference
In-Depth Information
is the number of predicates in the pre condition of B .In
other terms ω S ( B ) gives the rate of preconditions in B that can be achieved
using a scenario S .
Similarly, ω S ( O ) gives the ratio of state conditions in intrusion objective O
that can be satisfied in a scenario S , namely:
Where
|
Pre ( B )
|
0
if there exists at least one element in S O
which has a negative influence on O
ω S ( O )=
U ( S O ,O )
|StateCondition ( O ) |
otherwise
5 Weighting Scenarios
This section aims at showing how correlation weights can be used to rank-order
a set of possible scenarios leading to the same intrusion objective.
In the following, to each scenario S =( A 1 ,A 2 , ..., A n ,O ) we associate its
vector of correlation weights ( ω S ( A 1 ) S ( A 2 ) , ..., ω S ( A n ) S ( O )). The question
is how to aggregate these weights in order to evaluate the plausibility of a given
scenario and how to compare two weighted scenarios. By g we designate the
aggregation operator.
A first natural aggregation mode is to consider the mean operator, namely:
Mean-based agregation mode:
ω S ( A i )+ ω S ( O )
g ( A 1 , ..., A n ,O )=
i =1
n +1
However this aggregation mode is not desirable since scenarios containing
actions with a null weight are not excluded.
Conjunctive-based agregation mode:
A natural condition that g should satisfy is:
If
ω S ( A i )=0or ω S ( O ) = 0 then g ( A 1 , ..., A n ,O )=0.
Aggregation functions satisfying this condition are called conjunctive oper-
ators. A weaker form of such operators can be if ω S ( A i )=0or ω S ( O )=0
then scenario S should be among the least plausible ones. The weakest form of
a conjunctive operator would be to say that a scenario S should not be among
the most plausible ones if ω S ( A i )=0or ω S ( O )=0.
An example of aggregation operator which is conjunctive is the minimum
operator, namely:
i
∈{
1 , .., n
}
Definition 11: A scenario S =( A 1 ,A 2 , ..., A n ,O ) is said to be more plausible
than S =( B 1 ,B 2 , ..., B n ,O )if
min (( ω S ( A 1 ) S ( A 2 ) , ..., ω S ( A n ) S ( O ))) >
min(( ω S ( B 1 ) S ( B 2 ) , ..., ω S ( B n ) S ( O )))
 
Search WWH ::




Custom Search