OUTLIER SENSING BASED DDOS ATTACKER DISTINCTION METHOD AND APPARATUS USING STATISTICAL INFORMATION OF FLOW

摘要

The present invention relates to an algorithm to detect a DDoS attack and a method thereof. According to an embodiment of the present invention, the algorithm to detect a DDoS attack includes: a step of collecting statistical data by flow; a step of extracting data to determine if the flow is at least one DDoS attack from the collected statistical data using data extraction criteria; a step of applying at least one outlier detection technique to the data to determine if the flow is at least one DDoS attack to determine if the data is an outlier and outputting the resu and a step of determining if the flow is an DDoS attack according to the result. According to an embodiment of the present invention, the present invention can minimize wrong detection of a DDoS attack.