Method and device for detecting network intrusion by anomaly in network communications

FR3142058B1Active Publication Date: 2024-11-08COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES
-1 Cites 0 Cited by

Patent Information

Application Number
FR2022011862
Authority / Receiving Office
FR · FR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-11-15
Publication Date
2024-11-08
Estimated Expiration
2042-11-15
Patent Text Reader

Abstract

The invention relates to a device and method for anomaly-based network intrusion detection using a normal traffic model learned in an unsupervised learning phase. The method, implemented by a processor of an anomaly-based network intrusion detection system, comprises, during at least one inference phase, steps consisting of determining data samples containing anomalous data and data samples containing normal data.The method further includes steps of: receiving (302) a notification that a data sample En designated as containing normal data is a false negative, and should have been designated as containing abnormal data; constructing (306) a dummy sample Fn such that the dummy sample Fn is the opposite of said data sample En; and updating (310) said normal traffic model by running (308) at least once a false negative learning algorithm routine with said dummy sample Fn. Figure for the abstract: Fig. 3.
Need to check novelty before this filing date? Find Prior Art