Active learning traceability attack method based on multi-layer sampling

Abstract

The invention discloses an active learning traceability attack method based on multilayer sampling, and the method comprises the steps: proposing an active learning traceability attack model based on multilayer sampling to simulate a problem that a source node of an industrial information physical system is attacked, building an intelligent traceability attack model which is an iterative learning model of random walk and active learning, and solving a problem that the source node of the industrial information physical system is attacked; an attacker adds active learning on the basis of random walk of network nodes to improve the traceability walk performance. According to the method, a sampling strategy algorithm based on sample spatiality is constructed, and the distribution condition of the whole sample space can be effectively expressed; a sampling strategy algorithm based on sample diversity is constructed, and the problem of information redundancy in the sampling process is solved; the traceability attack of the active learning proposed by the invention has better traceability attack capability than the traceability attack of random walk, and the proposed algorithm has better sampling performance than other active learning algorithms, thereby improving the attack effect of the traceability attack.

Description

technical field [0001] The invention relates to the field of deep learning security technology, in particular to an active learning traceability attack method based on multi-layer sampling. Background technique [0002] Industrial Cyber-Physical Systems (ICPS) is a general term for a class of control systems used in industrial production, which includes supervisory control and data acquisition systems (SCADA), distributed control systems (DCS) and other common in Small control systems (such as programmable logic controllers) for industrial sectors and critical infrastructure, etc. It includes chemical chemistry, pharmaceuticals, hydropower energy, oil, natural gas, discrete manufacturing, automated production, transportation, aerospace and other fields, and plays a vital role in national infrastructure. [0003] With the development of machine learning, more and more attackers are beginning to monitor system network traffic and start modeling and analysis, starting from the...

AI Technical Summary

Problems solved by technology

It mainly solves the problem of classifier learning iteration in the case of a small amount of labeled training sample data. By increasing the information, space and diversity of the training sample data, high-value samples are screened out to solve the learning of the classifier under a small number of samples. question

Images