Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/986A9DC4-CE6F-4301-B810-EF5A0A8E9AE7 | ||
| Year | 2022 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Asterios Oikonomou, "Implementation of lightweight cryptography algorithms in FPGAs", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2022 https://doi.org/10.26233/heallink.tuc.93352 | ||
| Appears in Collections |
IoT applications typically involve a tree-like structure of devices, with the top levels occupied by a small number of highly complex powerful machines and a plethora of more restricted in computational and energy resources devices as one moves to the bottom levels. While cost and physical size restrictions mandate that those limited devices employ simplistic computing resources with small capacity batteries, the penetration of IoT applications in all forms of activities has multiplied the computational tasks (both in number and complexity) that have to be executed, increasing the pressure to use highly computationally and energy efficient devices and implementations of algorithms to be executed on their hardware. One of those tasks is data encryption, employed as one of the fundamental means to secure the exchange and storage of sensitive information. Lightweight encryption algorithms have been proposed as a good balance between data protection and computational complexity, enabling simple devices to handle those tasks. In this diploma thesis, three such algorithms are examined (Clefia, Simon and Present) focusing on devices that are placed at the edge of the IoT infrastructure. These devices communicate with a significant number of leaf-nodes and concentrate the traffic between the leaves and the higher levels of infrastructure. While they retain much of the limitations of the lower-levels of the IoT architecture (low cost, small size and limited energy resources), they have to provide significant computational performance and this work aims to explore the implementation of the encryption algorithm in custom reconfigurable hardware in order to achieve higher performance than common CPU-based systems and even higher energy efficiency. A comparison is therefore made between implementations of the three algorithms in software and hardware Specifically, the three algorithms were tested in software on four different processors(ARM Cortex-A9, ARM Cortex-A53, Intel Core I5-4200U and Intel Xeon E5-2630) and for the hardware at Xilinx KRIA KV260 FPGA. The results of the measurements show that Present block cipher is 1,473 times faster when applied on hardware compared to the Intel Xeon E5- 2630. The same comparison is made with the others algorithms where Simon cipher appears to 29 times faster when applied to the hardware while Clefia block cipher is 252 times faster when applied to the hardware compared to the Intel Xeon processor E5-2630.
Copyright © DIAS 2013
