Parameter Efficiency Testing for an Intrusion Detection System Inspired by the Human Immune System

Abstract

In this paper, a study of parameter efficiency for an intrusion detection system (IDS) inspired by the human immune system has been conducted. The IDS monitors a folder in the operating system to check for modifications or infections. The IDS is based on the negative selection algorithm to generate binary strings called receptors. The receptors are used to check for anomalies in the monitored folder. The receptors are gen-erated based on fundamental parameters of receptor size and activation threshold. This work investigates the efficiency of said parameters in a test environment. A wide range of receptor sizes and activation thresholds are tested, their efficiency factors are calculated, and the results are presented, analyzed and concluded.