Abstract
Blockchain technology is currently evolving rapidly, and smart contracts are the hallmark of the second generation of blockchains. Currently, smart contracts are gradually being used in power system networks to build a decentralized energy system. Security is very important to power systems and attacks launched against smart contract vulnerabilities occur frequently, seriously affecting the development of the smart contract ecosystem. Current smart contract vulnerability detection tools suffer from low correct rates and high false positive rates, which cannot meet current needs. Therefore, we propose a smart contract vulnerability detection system based on the Siamese network in this paper. We improved the original Siamese network model to perform smart contract vulnerability detection by comparing the similarity of two sub networks with the same structure and shared parameters. We also demonstrate, through extensive experiments, that the model has better vulnerability detection performance and lower false alarm rate compared with previous research results.
Funding Information
  • National Natural Science Foundation of China (62172353, U20B2046, FNSRFP-2021-YB-48, 1221045)
  • Future Network Scientific Research Fund Project (62172353, U20B2046, FNSRFP-2021-YB-48, 1221045)
  • Innovation Fund Program of the Engineering Research Center for Integration and Application of Digital Learning Technology of Ministry of Education (62172353, U20B2046, FNSRFP-2021-YB-48, 1221045)

This publication has 23 references indexed in Scilit: