A Multi-sensor Data Fusion Approach for Detecting Direct Debit Frauds

Abstract

Electronic payment systems have always represented an attractive target for cyber criminals. In this context the Single Euro Payments Area Direct Debit (SDD) service is gaining more and more importance since it has been promoted by the European banking industry as an innovative payment infrastructure. This service allows to perform electronic payments across the Euro zone as simple as domestic payments currently are. This schema facilitates the access to new markets by enterprises and reduces the overall cost to move capitals in Europe, but the other side of the coin is that it is the only financial system that has recorded an increase in the number of frauds, as highlighted by European Central Bank in a report dated 2015. In this paper SDD service attack patterns are analysed and a SDD fraud detection system based on multi-sensor data fusion is presented. Specifically, the Dempster-Shafer Theory is used in the proposed system to correlate security-relevant data from multiple information sources in order to detect anomalous behaviours that could be the evidence of an ongoing SDD fraud.