Too many researches have been done using artificial immune systems AIS to solve intrusion detection problems due to several reasons. The self and non-self model based on the Negative Selection Algorithm NSA is the dominant model since it is adopted by the vast majority of these researches. However, this model has some problems especially in terms of scalability and coverage. This paper tries to exploit some interesting concepts proposed by the new danger theory to overcome the problems associated with the self and non-self model. That by improving NSA in order to achieve better detection rates by integrating the basic danger concepts. In this approach, the intrusion detection is related to the damage that can occur in the system and that can be caused by both external elements such as internal elements. The proposed algorithm integrates and combines the basic concepts of intrusion detection systems IDS based on the role of T cells described by the negative selection algorithm, with those inspired by the role of dendritic cells to process the alarm signals and to judge thereafter whether there is presence of a dangerous element or not.
[1]
Mark Burgess.
Recent Developments in Cfengine
,
2001
.
[2]
P. Matzinger,et al.
An innate sense of danger.
,
1998,
Seminars in immunology.
[3]
P. Matzinger.
The Danger Model: A Renewed Sense of Self
,
2002,
Science.
[4]
Leandro Nunes de Castro,et al.
Artificial Immune Systems: Part I-Basic Theory and Applications
,
1999
.
[5]
Mark Burgess.
Evaluating cfengine’s immunity model of site maintenance
,
2007
.
[6]
Uwe Aickelin,et al.
Danger Theory: The Link between AIS and IDS?
,
2003,
ICARIS.
[7]
P. Matzinger.
Tolerance, danger, and the extended family.
,
1994,
Annual review of immunology.
[8]
Stephanie Forrest,et al.
Infect Recognize Destroy
,
1996
.
[9]
Mark Burgess.
Two Dimensional Time-Series for Anomaly Detection and Regulation in Adaptive Systems
,
2002,
DSOM.