Anomaly detection network intelligence fuzzy neural network using a type II

摘要

A detector (102) abnormality comprising a fuzzy neural network type II having a structure (112) control on three levels tracking attack symptoms and suggests scale actions (110) corrective until attack symptoms start disappear, wherein the control structure includes: a first level (202) including a plurality of membership functions, each membership function is adapted to: compile a statistic (108) network; and processing the collected statistical metric that is collected statistical divided by a theoretical maximum of the collected statistics; a second level (204) including a plurality of adders, each adder is adapted to: receive a unique set of metrics associated with the membership functions; and calculating an average based on the unique set of metrics and an exchange rate of each of the metrics in the single set; and a third level (206) including at least one aggregator and at least one table (2081, 208k), each aggregator is adapted to: receive a unique set of calculated averages; and add the unique set of averages calculated; and wherein each table is used to analyze the mean calculated summed to determine whether a course of action is necessary to treat the symptoms of the attack.