Academic Journals Database
Disseminating quality controlled scientific knowledge

ATTACK DETECTION AND CLASSIFICATION OF HETEROGENEOUS WIRELESS SENSORS USING CO-CLUSTERING

ADD TO MY LIST
 
Author(s): K.V.RAMANA | N.S.V.SRINIVAS

Journal: International Journal of Engineering and Technology
ISSN 0975-4024

Volume: 4;
Issue: 1;
Start page: 11;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Attack Forecasting | Heterogeneous Sensor | Co-Clustering | Attack Graphs | Transitive Closures.

ABSTRACT
In a Wireless Sensor Network a large number of sensors are deployed for the purpose of sensing data and then to bring the data back securely to nearby base stations. The base stations then perform thecostly computation on behalf of the sensors to analyze the data sensed by the sensors. Due to resource limitations of the nodes and also due to the vulnerability of physical captures of the nodes, the traditional cryptographic techniques are very complex and should not fit for energy constrained environments. Data mining techniques can be applied to find the malicious behavior of the nodes of the SensorNetwork. By analyzing the traffic patterns one can differentiate the normal behavior from malicious behaviour.Those techniques are used to identify various attacks. This paper addresses the issue of Attacks using data mining techniques. There exist two types of attacks: (i) external and (ii) internal. External attacks are those in which an attacker manipulates the communication between pairs of trusted nodes and causes the nodes to de-synchronize. Internal attacks are those in which internal attackers report false clock references to their neighboring nodes proposed an approach to develop a protocol. The protocol not only finds malicious node(s) but also counts them withinthe group using data mining clustering techniques.
Why do you need a reservation system?      Save time & money - Smart Internet Solutions