Elliptic Curve Cryptography hardware accelerator for high-performance secure servers
JOURNAL OF SUPERCOMPUTING
Authors: Parrilla, Luis; Alvarez-Bermejo, Jose A.; Castillo, Encarnacion; Lopez-Ramos, Juan A.; Morales-Santos, Diego P.; Garcia, Antonio
Security threats affecting electronics communications in the current world make necessary the encryption and authentication of every transaction. The increasing levels of security required are leading to an overload of transaction servers due to cryptographic tasks. In this paper, a hardware-implemented coprocessor for Elliptic Curve Cryptography operations is presented. This coprocessor enables the acceleration of secure services and can be implemented in the last generations FPGA, thus allowing to host in the same chip a software secure web/database server and the cryptographic coprocessor. Obtained results show advantages of the proposed solution when compared to software implementations and classical acceleration using graphics processing units. Moreover, the proposed cryptographic coprocessor presents improvements over other hardware implementations when area, performance and scalability are considered. The developed crypto-processor has been implemented in a xc7z020clg484-1 device from Xilinx, taking advantage of the synergy provided by the ARM microprocessors and the programmable logic for hardware implementations included in the device. This design requires only 9852 LUTs, while providing 8930 scalar-point operations per second when operating at 50 MHz, with a power consumption of 0.42 W.
Serial and large investors in initial coin offerings
SMALL BUSINESS ECONOMICS
Authors: Boreiko, Dmitri; Risteski, Dimche
This study is the first to provide systematic evidence regarding investor behaviour in initial coin offerings (ICOs), their investment patterns and their role in the success of campaigns. Using hand-collected data on 472 public token sales over the period of 2013-2017, we advance the ICO literature by demonstrating that some contributors often invest in more than one campaign, and such serial investors contribute earlier. However, they are not more informed and fail to pick better-quality ICOs. Only large serial investors invest more in campaigns that raise more funds, attract more contributors, are more likely to reach their hard caps, and distribute tokens that are listed on crypto exchange. Our findings raise the question of whether regulatory or industry self-regulation agreements on information provision measures are needed to protect smaller retail ICO investors that exhibit naive reinforcement learning behaviour.