Multi-threshold network attack detection method integrating ellipsoid and multi-cell body state estimation domain

By combining the ellipsoid and multicellular state estimation domain methods, the intersection area and clustering threshold of network attacks are calculated, which solves the problems of insufficient detection accuracy and large computational load in the existing technology, and realizes flexible attack degree judgment and reduced computational load.

CN118972133AActive Publication Date: 2024-11-15SHANGHAI MARITIME UNIVERSITY
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411127474.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-15
Estimated Expiration
2044-08-16

AI Technical Summary

Technical Problem

Existing network attack detection methods are difficult to meet the unknown but bounded conditions of noise and disturbance in actual industrial production scenarios, resulting in insufficient detection accuracy and large computational load.

Method used

A multi-threshold network attack detection method integrating ellipsoids and multi-cell state estimation domains is adopted. By constructing a discrete time-varying networked control system, a prediction filter and an observer are designed. The state prediction set and observation set are described by centrosymmetric multi-cells and ellipsoids. The area of ​​the intersection is calculated, and the attack degree is determined by Monte Carlo method and K-means clustering algorithm.

Benefits of technology

While ensuring detection accuracy, it reduces computational load, provides flexible attack level assessment, and avoids the conservatism and computational overload of existing methods.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118972133A_ABST
    Figure CN118972133A_ABST
Patent Text Reader

Abstract

The invention relates to a multi-threshold network attack detection method integrating ellipsoid and multi-cell body state estimation domains. The method comprises the following steps: constructing a networked control system in a discrete time-varying form; aiming at the condition that the networked control system is attacked by false information injection, a prediction filter and an observer are designed respectively, a system state prediction set and a system state observation set are obtained, the system state prediction set is described by a centrosymmetric polytope, and the system state observation set is described by an ellipsoid; and calculating the area of the intersection part of the centrosymmetric multi-cell body and the ellipsoid, and judging the relationship between the area of the intersection part and different set thresholds to obtain a network attack detection result containing the attack degree. Compared with the prior art, the method has the advantages that the accuracy and the real-time performance are ensured, and meanwhile, the excessive calculation amount is avoided.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Joint interval estimation method for state and fault of networked control system under FDI attack

    CN114019944A

  • Active attack detection method for improving detection rate

    CN114063602A

  • Power distribution network safety state estimation method and system based on event triggering mechanism

    CN117578444A

  • Multi-cell-body double-filtering state estimation algorithm for radar data

    CN117910201A

  • Method and apparatus for assessing risk of vehicle, and system for monitoring attack

    WO2024065283A1