A vulnerability detection method and system for Ponzi schemes in smart contracts

Abstract

The invention discloses a method and system for detecting loopholes of Ponzi schemes in smart contracts. The invention firstly extracts its control flow diagram and data flow diagram according to the bytecode of the target smart contract. Then combined with the characteristics of the Ponzi scheme, its code performance in the smart contract includes operations such as capital injection, contract transfer, and dividends. The instructions in the bytecode mainly involve CALL and CALLVALUE instructions, etc., to summarize the characteristics of the Ponzi scheme contract. , and then use this to construct data flow and control flow constraints, and finally use the constraint solver to solve it to determine whether the target contract has the characteristics of a Ponzi scheme, so as to alert subsequent users to call the contract and avoid financial losses caused by the contract to subsequent callers.

Description

technical field [0001] The invention belongs to the technical field of blockchain information security, and designs a method and system for detecting vulnerabilities in smart contracts, in particular to a method and system for detecting vulnerabilities in Ponzi schemes in smart contracts. Background technique [0002] A blockchain is an ever-growing list of peer-to-peer network value transfer transaction records maintained through a distributed consensus mechanism. As an important part of Bitcoin, blockchain technology has anonymity and non-repudiation, which makes it possible to transfer value between anonymous participants without relying on an authoritative third party. As a hot spot in academia and industry, blockchain technology is often referred to as the next-generation Internet. A smart contract is an application of blockchain technology, a computer agreement between mutually distrusting participants that is automatically enforced on the blockchain when preset condi...

AI Technical Summary

Problems solved by technology

This makes it difficult for regulators to trace the source of the Ponzi scheme

[0007] (2) Smart contracts are non-tamperable

[0012] (2) The contract income only comes from investors, that is, the return of existing investors can only come from new investors;

[0014] (4) The later the investor who participates, the greater the risk, that is, when it is difficult to deceive new investors, the scam will inevitably be exposed

[0018] (3) Waterfall type: The dividend distribution method is different from the chain type. After the contract creator retains a part of the investment, the remaining funds are sent to investors in accordance with the investment ratio in sequence. Subsequent investors are required to invest enough funds, otherwise it will be difficult to support the contract run;

[0021] Both of these schemes are based on deep learning to detect vulnerabilities in smart contracts, but the models in their schemes need to continuously learn the "training set" to predict new content, which has a hysteresis. However, the blockchain and smart contracts have a fast update frequency. , a large amount of data and other characteristics, which means that the training set is difficult to keep pace with the times, and the detection false positive rate is high

Images