Unlock instant, AI-driven research and patent intelligence for your innovation.

Intelligent contract endless loop detection method based on graph convolutional neural network

A convolutional neural network and dead-loop detection technology, applied to biological neural network models, neural architectures, instruments, etc., can solve problems such as smart contract dead-loop detection, achieve good versatility and practical value, and have versatility and applicability performance, high accuracy

Active Publication Date: 2020-01-10
HANGZHOU YUNXIANG NETWORK TECH
View PDF8 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] A smart contract dead-loop detection method based on graph convolutional neural network, through GCN model training and learning to realize the automatic judgment of smart contract dead-loop, thereby solving the problem of smart contract dead-loop detection, the method specifically includes the following steps:

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Intelligent contract endless loop detection method based on graph convolutional neural network
  • Intelligent contract endless loop detection method based on graph convolutional neural network
  • Intelligent contract endless loop detection method based on graph convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment

[0052] 3. The smart contract dead loop detection method in this embodiment takes the Fallback cycle of the VNT smart contract as an example, such as image 3 As shown, the specific implementation process is as follows:

[0053] (1) Use the automatic drawing tool to extract the core nodes, VAR variable nodes and directed edges from the smart contract source code, that is, convert the target contract source code into a graph structure. The specific implementation steps are as follows:

[0054] (1-1) Extract all function modules from the target contract, and the function modules are the core nodes. The characteristic attributes of core nodes are: core node ID (ie A, B, C,...), function return value (ie void, uint8, bool, ...), function call node ID (ie A, B, C, ...), edge-out timing, edge-in timing, and function call methods (including CALL, INNCALL, SELFCALL, and FALLCALL; respectively representing function calls outside the contract, function calls inside the contract, functio...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an intelligent contract endless loop detection method based on a graph convolutional neural network, which realizes automatic judgment of an intelligent contract endless loop through training and learning of a GCN model, and specifically comprises the following steps: collecting and designing an intelligent contract endless loop case, and making an intelligent contract source code data set; performing case analysis on the endless loop intelligent contract data set; converting the endless loop smart contract source code data set into a core node graph data set with a unified structure; ablating the graph data characteristics to the core node; mapping the core node features into vectors by using a vector conversion tool; and constructing a GCN model for learning and training, and automatically outputting an intelligent contract endless loop judgment result. Compared with an existing intelligent contract detection tool based on a traditional method, the method hasthe advantages that the accuracy is higher, the current blank of intelligent contract security hole detection based on deep learning is filled up, and the method has good universality and practical value and also has good reference significance.

Description

technical field [0001] The invention belongs to the technical field of block chain smart contract security loophole detection, and in particular relates to a smart contract dead-loop detection method based on a graph convolutional neural network. Background technique [0002] In recent years, computer technology has developed rapidly, but there are also more and more computer security loopholes, especially if some security loopholes are exploited by criminals, it will not only increase the probability of computer being attacked, but may also lead to the loss of important information. lost, resulting in major losses. How to discover and repair vulnerabilities in time and continuously improve the security of computers has always been a hot issue of public concern. [0003] The blockchain smart contract is a computer protocol running on the blockchain. It has the characteristics of high efficiency, real-time update, non-tampering, and decentralization. However, the security lo...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F21/56G06F21/57G06Q40/04G06N3/04
CPCG06F21/563G06F21/577G06Q40/04G06N3/045
Inventor 黄步添刘振广钱鹏周伟华陈建海周峰
Owner HANGZHOU YUNXIANG NETWORK TECH