Abstract—In recent years, many application domains, such as environmental monitoring and management, smart homes and buildings, and intelligent traffic, including parking and traffic lights, require edge computing platforms for processing data at the stations before sending preliminaries to cloud servers. Although processor-based computing platforms like Raspberry Pi or Jetson Nano/Xavier offer many advantages when applied to this system, the platforms exhibit high energy consumption, security vulnerabilities, and high non-recurring costs. Therefore, this paper introduces a secure and energy efficient edge computing platform built with Field Programmable Gate Array (FPGA) technology. This platform’s customized RISC-V processor, developed based on the Open-Source RISC-V architecture, is augmented with the security algorithm SHA-256 to allow data to be encrypted before sending to cloud servers. The security approach is built as a custom instruction processor, allowing users to implement their systems without hardware design skills. The customized processor with SHA-256 extension instruction is developed with Verilog-HDL and built with different low-end/-price FPGA boards for testing and comparing with Micro:Bit and Raspberry Pi embedded boards. Compared with the Micro:Bit system, at the same price as our low-end experimental platform (GW1NR9), our proposed system achieves speedups by up to 27.01×. Compared with Raspberry Pi 4, five times more expensive than our platforms, we need a slightly longer execution time but use 4.24× less energy consumption. For a deep dive evaluating our customized processor, we will use the Dhrystone benchmark to compare it with the ARM-Cortex high-end embedded processor. Experimental results show that we achieve the value DMIPS/Hz of 1.96, better than ARM-Cortex 3 processor and other RISC-V softcore implementations.
 
Keywords—Field Programmable Gate Array (FPGA), secured and lightweight edge devices, RISC-V, extension instructions


Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.