High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Scrutinizing Well Known Counter Measures of Anomaly Detection in Intrusion Detection System

Author(s):

C. Renuga , HOD cum Assistant Professor,Bharathiyar Arts and science college for women,deviyakurichi,salem; D. Gomathi, Research Scholar,Bharathiyar Arts and science college for women,deviyakurichi,salem

Keywords:

Intrusion Detection System, Anomaly Detection, Sequential Probability Ratio Test, Shiryaev-Roberts Procedure

Abstract

This project contemplates the difficult of efficient on-line anomaly recognition in computer network traffic. The problem is move toward statistically, as that of consecutive (quickest) change point detection. A multi-cyclic setting of rapidest change detection is a usual fit for this problem. It recommends a novel score-based multi-cyclic discovery algorithm. The procedure is grounded on the so-called Shiryaev–Roberts procedure. This technique is as easy to employ in preparation and as computationally economical as the widespread Cumulative Sum chart and the Exponentially Weighted Moving Average arrangement. The likelihood ratio based Shiryaev–Roberts procedure has interesting optimality properties; predominantly it is precisely optimum in a multi-cyclic setting geared to perceive a change occurring at a far time horizon. It is therefore predictable that an intrusion detection algorithm based on the Shiryaev–Roberts technique will perform better than other recognition schemes. This is confirmed experimentally for real suggestions. We also discuss the opportunity of accompanying our anomaly detection algorithm with a spectral-signature intrusion detection system with false apprehension filtering and true occurrence confirmation capability, so as to obtain a synergistic organization.

Other Details

Paper ID: IJSRDV4I90110
Published in: Volume : 4, Issue : 9
Publication Date: 01/12/2016
Page(s): 123-127

Article Preview

Download Article