Abstract

۱٫ Introduction

۲٫ Related work

۳٫ Problem statement and summary of contributions

۴٫ Background knowledge

۵٫ Master Node Based Clustering (MNBC)

۶٫ System design

۷٫ Performance evaluation

۸٫ Security evaluation

۹٫ Conclusion

References

Abstract

     Bitcoin is a digital currency based on a peer-to-peer network to propagate and verify transactions. Bitcoin is gaining wider adoption than any previous crypto-currency. However, the mechanism of peers randomly choosing logical neighbours without any knowledge about the underlying physical topology can cause a delay overhead in information propagation which makes the system vulnerable to double spend attacks. Aiming at alleviating the propagation delay problem, this paper introduces a proximity-aware extension to the current Bitcoin protocol, named Master Node Based Clustering (MNBC). The ultimate purpose of the proposed protocol, which is based on how clusters are formulated and how nodes can define their membership, is to improve the information propagation delay in the Bitcoin network. In the MNBC protocol, physical internet connectivity increases as well as the number of hops between nodes decreases through assigning nodes to be responsible for maintaining clusters based on physical Internet proximity. Furthermore, a reputation-based blockchain protocol is integrated with MNBC protocol in order to securely assign a master node for every cluster. We validate our proposed methods through a set of simulation experiments and the findings show how the proposed methods run and their impact in optimising the transaction propagation delay.

Introduction

     Bitcoin is the first digital currency to attract mainstream, businesses, and people’s attention. Bitcoin is a virtual, decentralised cryptography that no one is in charge of it and it is not tracked by any hard asset or government. Instead, Bitcoin relies on a peer-to-peer (P2P) network that protects the Bitcoin’s value by means of cryptography that is performed by peers mining Bitcoins through brute-forcing double SHA-256 hash function. The concept of Bitcoin was proposed by an anonymous programmer using the pseudonymous Satoshi Nakamoto [1]. Due to the Bitcoin advantages such as the absence of intermediates and its decentralised architecture, Bitcoin is now deployed as a currency by many businesses and companies. Bitcoin performs global transactions which allow non-refundable, reasonably fast money transfers to any part of the world [2].

     Based on the publicly distributed ledger that is shared by the entire Bitcoin network nodes, a distributed trust mechanism is achieved [3]. This mechanism is considered as a monitoring technique by which the amount of available bitcoins1 in Bitcoin will be tracked. To achieve this mechanism, two main requirements need to be fulfilled: (i) transactions verification process has to be achieved in a distributed manner to ensure the validity of transactions; and (ii) successfully processed transactions have to be quickly announced to everyone in order to guarantee the consistent state of the blockchain [4]. As transactions are validated against the blockchain, reaching a consistent state over the blockchain is considered as a fundamental requirement towards achieving the distributed transactions verification process.

Conclusion

     In this paper, a brief background of the Bitcoin system as well as analysing the information propagation in the real Bitcoin network were presented. In addition, how propagation delay in the Bitcoin network could affect the security by offering an opportunity to double spend the same coins; thereby abusing the consistency of the public ledger was discussed in this paper. The MNBC, a novel clustering protocol that incorporates master node based physical proximity and reputation scheme based blockchain into the existing Bitcoin protocol, was presented in this paper. By conducting extensive simulations, MNBC evaluation results indicate an improvement in the transaction propagation delay over the Bitcoin network protocol. However, MNBC maintains a lower variance of delays than the BCBSN protocol. Furthermore, experiments with different latency thresholds have been conducted to identify the distance threshold that would give a better improvement in the transaction propagation delay. We discovered that providing a less latency distance threshold would improve the transaction propagation delay with a high proportion. Furthermore, evaluation of partitioning attacks in the Bitcoin network as well as the MNBC and BCBSN protocols were presented in this paper. The results revealed that attackers still need more resources to split the network in the proposed protocol, especially with a higher number of nodes. Furthermore, the effect of malicious nodes on the MNBC protocol was validated in this paper. The results proved that MNBC throughput had not been significantly affected by the malicious behaviour.