Абстрактный

Cluster Based Intrusion Detection System for Wireless Sensor Networks

Gopi K, Sivaprakash S

The intrusion detection system identifies the legitimate and attackers in the network area. The intrusion detection system is designed in two categories. In the first model a system component is used for monitoring the security of a WSN and diagnosing compromised or vulnerable. Second model is the monitoring or surveillance system for detecting a malicious intruder that invades the network domain. Intrusion detection is applied to detect malicious or unexpected attackers in Wireless Sensor Network (WSN). The intruder can be an enemy in a battlefield, or a malicious moving object in the area of interest. Same detection probability is used in the uniform distribution modeled WSN. Gaussian-distributed WSNs can provide differentiated detection capabilities at different locations. Different degrees of probability is used in Gaussian distribution modeled WSN. There are two sensing scenarios are used in the intrusion detection system. They are single-sensing detection and multiple-sensing detection scenarios. Relaxed intrusion detection and immediate intrusion detection models are used in the single sensing scenarios. In the immediate intrusion detection model the intruders are detected before any movement in the WSN. In the relaxed intrusion detection model the intruders are detected after some movements in the WSN. The intrusion detection system is enhanced to support different deployment schemes. Automatic parameter learning model is integrated with the intrusion detection system. Sensor node clustering scheme is integrated with the IDS to handle single and multiple detection mechanisms. Sensing and transmission coverage factors are included in the intrusion detection process.