Using immunological principles in anomaly detection
Abstract
In this paper, a method for anomaly detection is described which is based on principles of the natural immune system. Compared to existing methods, this method has the advantage of not requiring prior knowledge about all possible failure modes of the monitored system. Instead it uses the normal behavior pattern to generate a set of detectors which can detect any deviation that exceed allowable variation from the defined normal. An example using Mackey-Glass equation is presented and the performance of the proposed anomaly detection algorithm is reported.
Publication Title
Intelligent Engineering Systems Through Artificial Neural Networks
Recommended Citation
Dasgupta, D. (1996). Using immunological principles in anomaly detection. Intelligent Engineering Systems Through Artificial Neural Networks, 6, 443-448. Retrieved from https://digitalcommons.memphis.edu/facpubs/3319