الفهرس 
Dissertation Organization
INTRUSION DETECTION SYSTEM BASED ON ARTIFICIAL IMMUNE SYSTEMSOnly 14 pages are availabe for public view
 |  | from | 206 |  |  |
| | | from | 206 | | |
Abstract
Wireless sensor networks (WSNs) are promlsmg technology for
surveillance and reconnaissance in many applications, such as next
generation C4ISR, digital battlefield and perimeter defence.
A sensor network is composed of a large number of sensor nodes that are
densely deployed either inside the phenomenon or very close to it and
communicate wirelessly. The main purpose of a sensor network is to
gather information by monitoring the environment. This is possible by
collecting information from all the sensors distributed in the environment.
A sensor node combines the abilities to compute, communicate and sense.
The structure of a sensor node IS influenced by
device complexity, high performance, wireless networking technologies,
and advances in the development of micro- electromechanical
system (MEMS).
The dearth of effective security mechanisms is the main obstacle for the
sensor networks. Traditional security techniques used in the other
networks cannot be applied directly in wireless sensor networks.
In wireless sensor networks several important security challenges,
including standard security requirements, such as availability,
confidentiality, integrity, authentication, robustness to den ial-of-service,
and other special requirements for wireless sensor networks, such as
secure routing, node capture, secure group management, and
intrusion detection.
Due to inherent limitations in wireless sensor network (limited
processing, storages power and energy), security for these sensor
llct\\orks is not easy. Since sensor networks pose unique challenges.
traditional security techniques cannot be applied directly. lnuusion
I)Ll..:ctiol1S:stem (IDS) is a must ror this critical application. Providing
adaptive new intrusion detection systems remains a challenging research
problem.
In our PhD thesis for securing the wireless sensor networks we develop
an adaptive intrusion detection system for WSNs based on Artificial
Immune System (ArS) that is inspired and simulated by natural and
Biological Immune System (BIS).
We present the integration of some useful immunological functions to
develop a robust and intelligent detection system.
By implementing Artificial Immune system in our security system, which
has the learnability as one of the main features, we found that: the system
response was better and faster while detecting different types of intruder
patterns, that providing a system with AI’s functions and capabilities
enables it to differentiate between the different actions done through the
system to detect and discover the adversary’s pattern even if it is the first
time to be exposed to any of these patterns.
As a result of using artificial immune system to build intrusion detection
system, the wireless sensor networks become more secured and immune.
Also we added enhancement for previous work for key pre-distribution
scheme (Eschenauer and Gligor (EG)) by using Artificial Immune
System technique.
In short, this thesis presents two contributions IJ1 the field of WSN
security:
Contribution I: Intrusion Detection System
We develop an adaptive intrusion detection system (rDS) for WSNs
based on Artificial Immune System (ArS) that is inspired and simulated
by natural and Biological Immune System (BrS). Taking into
consideration the security properties, storage- and energy-efficiency of’ a
wireless sensor network node. The result should serve as a practical
guideline on the suitable IDS to use according to the constraint of
available memory and the required level of detect the intruders for WSNs.
Contribution 2: Key pre-distribution
We present, cluster-based, key pre-distribution for Wireless Sensor
Networks. Our contribution uses only symmetric-key cryptography, and
is based on Artificial Immune System. The balance between security and
energy consumption is achieved in our contribution.
The second main contribution is about modifying an already existed key
pre-distribution for wireless sensor networks. Our proposal covers the
aspects of key setup and re-keying phase to overcome the drawbacks of
the existed scheme.