Utilizing a Variable Activation Threshold Approach to Calculate and Increase Probabilities of Anomaly Detection
Krzysztof W. Wawryn, Patryk Widuliński · 2024
In this paper, a variable activation threshold method for increasing the probabilities of anomaly detection is described. The method has been implemented in an intrusion detection system (IDS) based on artificial immune systems (AIS). The proposed method involves decreasing the size of the activation threshold bit by bit to increase the size of the receptor set and in turn, extend receptor coverage for more successful anomaly detection. Equations for calculating anomaly detection probabilities using the proposed method are presented and tested against experimental results to verify their usability, and the work is then concluded.