A blockchain eclipse attack detection method based on CNN

By using a CNN-based blockchain eclipse attack detection method, a two-dimensional grayscale image is generated in real time and features are extracted using a pre-trained model. This solves the problem of identifying and defending against eclipse attacks in blockchain networks, achieving high-precision, real-time attack detection and defense linkage, and building a flexible defense ecosystem.

CN122120010BActive Publication Date: 2026-07-10SHANGHAI CRIMINAL SCI TECH RES INST +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHANGHAI CRIMINAL SCI TECH RES INST
Filing Date
2026-04-10
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively identify and defend against eclipse attacks in blockchain networks. They suffer from challenges such as difficulty in extracting high-dimensional nonlinear features, lack of a closed-loop system for detection and defense, feature coupling and ambiguity in dynamic network environments, and an imbalance between the need for high-precision detection and real-time response performance.

Method used

A CNN-based blockchain eclipse attack detection method is adopted. It generates two-dimensional grayscale images by capturing underlying communication messages in real time, extracts abnormal texture and spatial correlation features using a pre-trained CNN model, and combines response level-triggered defense measures to achieve iterative optimization.

Benefits of technology

It significantly improves the accuracy of identifying covert attacks, enhances the anti-interference capability in dynamic environments, achieves millisecond-level real-time detection and protection response, and builds a dynamic and flexible closed-loop defense ecosystem.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122120010B_ABST
    Figure CN122120010B_ABST
Patent Text Reader

Abstract

This invention provides a CNN-based method for detecting blockchain eclipse attacks, belonging to the field of blockchain and network security detection technology. The method includes: using a data acquisition engine deployed on key nodes of a blockchain P2P network to capture underlying communication packets in real time, and preprocessing these packets to obtain a two-dimensional grayscale image; inputting the two-dimensional grayscale image into a pre-trained CNN detection model to capture abnormal textures, edges, and spatial correlation features, and outputting binary classification results and probability values; matching the binary classification probability values ​​with a preset threshold, and combining this with the attributes of affected nodes and the proportion of malicious connections to determine the response level; triggering corresponding defense measures based on the response level, while continuously monitoring the defense effect and feeding it back to the pre-trained CNN detection model for iterative optimization. This method significantly improves the accuracy of identifying covert attacks, significantly enhances anti-interference capabilities in dynamic environments, and achieves millisecond-level real-time detection and protection response.
Need to check novelty before this filing date? Find Prior Art