Cyber-attacks can seriously affect the security of computers and network systems. Thus, developing an efficient anomaly detection mechanism is crucial for information protection and cyber security. To accurately detect TCP SYN flood attacks, two statistical schemes based on the continuous ranked probability score (CRPS) metric have been designed in this paper. Specifically, by integrating the CRPS measure with two conventional charts, Shewhart and the exponentially weighted moving average (EWMA) charts, novel anomaly detection strategies were developed: CRPS-Shewhart and CRPS-EWMA. The efficiency of the proposed methods has been verified using the 1999 DARPA intrusion detection evaluation datasets.
|Original language||English (US)|
|Title of host publication||2018 IEEE Symposium Series on Computational Intelligence (SSCI)|
|Publisher||Institute of Electrical and Electronics Engineers (IEEE)|
|Number of pages||5|
|State||Published - Feb 28 2019|