Abstract—This research paper is an elaboration of
Incremental Radial Based Function Neural Network model
with Particles Swarm Optimization (IRBF-PSO) in Intrusion
Detection System. This system is helpful to find the most
featured misuse and anomaly detection. RBF network is most
popular real-time classifier method. RBF method comprises of
mostly analysis and the thorny part is finding the right weights
and bias values for dynamic systems. The intrusion detection
system has become highly dynamic. Many large or small
enterprise systems are still facing with different problems in
this area with dynamic form. So the main objective of my work
is to employ Particles Swarm Optimization to detect the right
weight and bias values for RBF method.
In this method, apart from training with existing data and
information for design, there is a need to extend or redesign the
existing system to identify different pattern types and modulate
the system using PSO with new patterns. After experimentation,
this method has improved to identify the difficulty in anomaly
detections and reduce the rate of false alarm and fail cases.
Index Terms—Incremental method, intrusion detection
system, particles swarm optimization and radial based.
M. V. Siva Prasadis is with Anurag Engineering College, Kodad 508206,
India (e-mail: magantisivaprasad@gmail.com).
Ravi Gottipati is with Tripod Technologies, Hyderabad 500082, India
(e-mail: softtotime@gmail.com).
[PDF]
Cite: M. V. Siva Prasad and Ravi Gottipati, "A Novel Incremental Instruct Dynamic Intrusion Detection System Using PSO-RBF," International Journal of Future Computer and Communication vol. 4, no. 4, pp. 280-285, 2015.
