10 Clock-Periods Pipelined Implementation of AES-128 Encryption-Decryption Algorithm up to 28 Gbit/s Real Throughput by Xilinx Zynq UltraScale+ MPSoC ZCU102 Platform

The security of communication and computer systems is an increasingly important issue, nowadays pervading all areas of human activity (e.g., credit cards, website encryption, medical data, etc.). Furthermore, the development of high-speed and light-weight implementations of the encryption algorithms...

Full description

Bibliographic Details
Main Authors: Paolo Visconti, Stefano Capoccia, Eugenio Venere, Ramiro Velázquez, Roberto de Fazio
Format: Article
Language:English
Published: MDPI AG 2020-10-01
Series:Electronics
Subjects:
Online Access:https://www.mdpi.com/2079-9292/9/10/1665
Description
Summary:The security of communication and computer systems is an increasingly important issue, nowadays pervading all areas of human activity (e.g., credit cards, website encryption, medical data, etc.). Furthermore, the development of high-speed and light-weight implementations of the encryption algorithms is fundamental to improve and widespread their application in low-cost, low-power and portable systems. In this scientific article, a high-speed implementation of the AES-128 algorithm is reported, developed for a short-range and high-frequency communication system, called Wireless Connector; a Xilinx ZCU102 Field Programmable Gate Array (FPGA) platform represents the core of this communication system since manages all the base-band operations, including the encryption/decryption of the data packets. Specifically, a pipelined implementation of the Advanced Encryption Standard (AES) algorithm has been developed, allowing simultaneous processing of distinct rounds on multiple successive plaintext packets for each clock period and thus obtaining higher data throughput. The proposed encryption system supports 220 MHz maximum operating frequency, ensuring encryption and decryption times both equal to only 10 clock periods. Thanks to the pipelined approach and optimized solutions for the Substitute Bytes operation, the proposed implementation can process and provide the encrypted packets each clock period, thus obtaining a maximum data throughput higher than 28 Gbit/s. Also, the simulation results demonstrate that the proposed architecture is very efficient in using hardware resources, requiring only 1631 Configurable Logic Blocks (CLBs) for the encryption block and 3464 CLBs for the decryption one.
ISSN:2079-9292