Triple Modular Redundancy Optimization for Threshold Determination in Intrusion Detection Systems

摘要: This paper presents a novel approach for an Intrusion Detection System (IDS) based on one kind of asymmetric optimization which use any three already well-known IDS algorithms and Triple Modular Redundancy (TMR) algorithm together. Namely, variable threshold indicates attack observed protected network is determined by using all values obtained with known i.e., previously recorded data making decision majority. For these authors used k-nearest neighbors, cumulative sum algorithm, exponentially weighted moving average. Using proposed method we can get that more precisely than in the case individual. Practically, TMR obtain dynamically adjustment software, reduces existence false alarms undetected attacks, so efficiency such software notably higher better results. Today, Denial Service attacks (DoS) are most present type reason special attention paid to them this paper. In addition, CIC-DDoS2019 dataset, contains various recordings attacks. Obtained results solution showed characteristics each individual solution. worked timely, means were triggered properly efficiently.