Ship network real-time abnormity detection method based on programmable data plane

Abstract

The invention discloses a ship network real-time abnormity detection method based on a programmable data plane. The method comprises the following steps: 1) collecting network state data of each nodei in a network; 2) at each time point t, inputting the sequence data of the network state data into a hierarchical memory network HTM to obtain a prediction error; 3) taking the distribution of the original prediction error as a measure, calculating the likelihood value of the abnormal degree by using the distribution, and judging whether to trigger an abnormal alarm; 4) in multivariable joint anomaly detection, decomposing the input into a plurality of sub-modules according to the number of parameters, and obtaining a global metric through joint likelihood anomaly score calculation to represent an abnormity likelihood score of the whole system; and 5) carrying out joint estimation on the distribution of the original abnormity scores of the plurality of detection systems, and judging whether to trigger an anomaly alarm. The invention provides the abnormity detection method capable of quantitatively evaluating the abnormity degree of a network.

Description

technical field [0001] The invention relates to network technology, in particular to a real-time anomaly detection method for a ship network based on a programmable data plane. Background technique [0002] my country's military ship network technology has experienced three generations of technological development, which originated from the development and construction of the shipboard combat command system. In the early 1980s, my country's shipboard combat command system adopted a military 1553B bus with a rate of 1Mbps, baseband transmission and master-slave communication mode. By the early 1990s, in order to increase the communication rate, the modified commercial 10Mbps Ethernet adapted to the shipboard combat system network was adopted, and the network topology in the form of shared media and bus was generally used, and a small amount of shared media centered on hubs was used. , Star network topology. In the mid-to-late 1990s, with the further development of the shipb...

AI Technical Summary

Problems solved by technology

The detection process of this method is too complicated, the amount of calculation is large, and it has a long time delay.

Images