Detecting Ponzi Schemes on Ethereum: Towards Healthier Blockchain Technology
Eclipse Attacks on Bitcoin’s Peer-to-Peer Network
The author of this paper has studied the source code of Bitcoin's network layer in depth and found a major security hole in Bitcoin design at the network layer. The eclipse attack [Tampering with the delivery of blocks and transactions in bitcoin] is a possible attack to Bitcoin network. In such an attack, attackers control parts of network communication, create an adverse network to isolate victims from the rest of normal peers on the network, and block views of these victims. Then, attackers can take advantage of the victims’ computing power for their own malicious purposes, such as selfish mining, double spending and etc. To avoid the happen of eclipse attacks on Bitcoin network, several countermeasures, like deterministic random eviction, more ongoing connections and etc, have been explored and purposed in their work》
Peer to Peer
OmniLedger: A Secure, Scale-Out, Decentralized Ledger via Sharding
The paper presents a novel distributed ledger OmniLedger. Previous distributed ledgers (DLs) are unable to avoid trade-offs between decentralization, scaling and security, i.e. those DLs only hold two out of the three features. So A great innovation of OmniLedger is its assemble all the three features. The structure of the paper is clear. Detailed descriptions of the system design are given. Ideal models and practical experiments show its good performance. Weaknesses and future works are appropriately discussed.
A survey on the security of blockchain systems
S-gram: Towards Semantic-Aware Security Auditing for Ethereum Smart Contracts
In this paper, the authors present the S-gram semantic-aware security auditing technique for Ethereum smart contracts. The S-gram highlighted the insight that statistical abnormality is very much likely to indicate the existence of vulnerabilities. It's novel that the paper introduces the natural language processing technic to analysis the smart contract code. The n-gram language model achieved an over 90% accuracy in identifying different types of potential vulnerabilities.
Who Spent My EOS? On the (In)Security of Resource Management of EOS.IO
A well understanding of potential threats and their underlying causes is necessary when building a secure system. This paper is the first study that analyzes the security of the EOS.IO system and points out that all of the security vulnerabilities they find stem from the EOS.IO design, which is beneficial to the whole EOS ecosystem.
Page 1 of 2
Smart Contract (10)
data mining (5)
Transaction Graph (4)
Ponzi scheme (3)
Bitcoin Clustering (2)
Peer to Peer (2)
Generals Byzantine Problem (2)