기계학습 기반 비트코인 채굴 난이도 예측 연구
Machine Learning Based Prediction of Bitcoin Mining Difficulty
0
조회
0
다운로드
2
인용
0
참고문헌
키워드
초록
비트코인은 탈중앙화와 분산원장을 특징으로 하는 암호화폐로서 “작업증명”이라는 채굴시스템을 통해 유지된다. 채굴 시스템에서는 블록 생성시간을 일정하게 유지하기 위해 채굴 난이도를 조정하게 되는데, 기존의 채굴 난이도변경 방식은 미래의 해시파워를 반영할 수 없다는 문제가 있다. 따라서 실제시간과 예정시간 사이에 발생하는 오차로 인해 블록생성과 실세계 시간의 불일치를 가중시키게 되고, 결국 거래 기한을 맞추지 못하거나 코인 호핑 공격에취약점을 노출시키게 된다. 블록 생성시간을 일정하게 유지시키기 위한 기존 연구도 여전히 오차 문제를 갖는다. 본연구에서는 이러한 오차를 줄이기 위한 기계학습 기반 채굴 난이도 예측 방안을 제시한다. 이전 해시파워를 학습하여 미래의 해시파워를 예측하고 예측한 값을 이용하여 채굴 난이도를 조정한다. 우리의 실험 결과는 이와 같은 경우기존 채굴 난이도 조정방식보다 오차율을 약 36% 더 줄일 수 있음을 보여준다.
Abstract
Bitcoin is a cryptocurrency with characteristics such as de-centralization and distributed ledger, and these features aremaintained through a mining system called "proof of work". In the mining system, mining difficulty is adjusted to keep theblock generation time constant. However, Bitcoin’s current method to update mining difficulty does not reflect the futurehash power, so the block generation time can not be kept constant and the error occurs between designed time and realtime. This increases the inconsistency between block generation and real world and causes problems such as not meetingdeadlines of transaction and exposing the vulnerability to coin-hopping attack. Previous studies to keep the block generationtime constant still have the error. In this paper, we propose a machine-learning based method to reduce the error. Bytraining with the previous hash power, we predict the future hash power and adjust the mining difficulty. Our experimentalresult shows that the error rate can be reduced by about 36% compared with the current method.
참고문헌 (18)
- S. Nakamoto, "Bitcoin : A Peer-to-Peer Electronic Cash System", ~ (2008)
- "A partial hash collision bas ed postage scheme"[URL]
- Profitability of the coin-hopping strategy, "Király", EGRES, ~, pp.EGRES quick proof, no. 2018-03 (2018)
- D. Kraft, "Difficulty control for blockchain-based consensus systems", Peer-to-Peer Networking and Applications, 9(2), pp.397~413 (2016)
- "BIP 34 : Consensus (soft fork)"[URL]
- Meshkov, Dmitry, "Short paper: revisiting difficulty control for blockchain systems", Data Privacy Management, Cryptocurrencies and Blockchain Technology, 10436, pp.429~436 (2017), Springer
- Kalodner, Harry, "BlockSci:Design and applications of a blockchain analysis platform", arXiv preprint arXiv:1709.02489, ~ (2017)
- "Bitcoin Wiki : Difficulty"[URL]
- "Bitcoin Wiki : Block timestamp"[URL]
- Akita, Ryo, "Deep learning for stock prediction using numerical and textual information", Computer and Information Science (ICIS), 2016IEEE, 2016, 1~6[DOI]
- McNally, Sean, "Predicting the price of Bitcoin using Machine Learning", Parallel, Distributed and Networkbased Processing (PDP), 2018 26th Euromicro International Conference on, pp.339~343 (2018), IEEE
- "Bitcoin cash : Difficulty Adjustment A lgorithm Update"[URL]
- "Automated bitcoin trading via machine learning algorithms"[URL]
- Jang, Huisu, "An empirical study on modeling and prediction of bitcoin prices with bayesian neural networks based on blockchain information", IEEE Access, 6, pp.5427~5437 (2017)
- Rosenfeld, Meni, "Analysis of hashrate-based double spending", arXiv preprint arXiv:1402.2009, ~ (2014)
- Ittay Eyal, "Majority is not enough : Bitcoin mining is vulnerable", Financial Cryptography and Data Security, 8437, pp.436~454 (2014)
- Bahack, Lear, "Theoretical Bitcoin Attacks with less than Half of the Computational Power (draft)", arXiv preprint arXiv:1312.7013, ~ (2013)