Double-Spending Detection for Fast Bitcoin Payment Based on Artificial Immune

Abstract

With the rapid development of Bitcoin, it is frequently used in the scene of fast payment. But the strategy which Bitcoin takes to prevent the double-spending attack is waiting for six confirmations (about one hour), this is not suitable for the fast payment scenarios where service time is about tens of seconds. The default strategy in fast payment is that do not offer the service until a payment transaction is added to the wallet of vendor, which is useless for the detection of double-spending attack. In this paper, an immune-based model is proposed to detect the double-spending attack in the fast Bitcoin payment. This model is composed of many immune-based Bitcoin nodes which include a detection modula and a traditional node. Antigen character is first extracted from a transaction by the detection modula, and initial detectors (mature detectors) are generated based on these antigens. Then, memory detectors and mature detectors are used to detect the double-spending attack, and a mature detector which matches an attack will evolve into a memory detector and be delivered to other immune-based nodes in the Bitcoin network, in order to rapidly detect the double-spending attack. Experimental result shows that this model can efficiently detect double-spending attacks in fast Bitcoin payment.