Skip to main navigation Skip to search Skip to main content

Hardware-Based Protection for Data Security at Run-Time on Embedded Systems

  • Xiang Wang*
  • , Xiaobing Zhang
  • , Weike Wang
  • , Pei Du
  • , Zhun Zhang
  • , Yuntong Tian
  • , Qiang Hao
  • , Bin Xu
  • *Corresponding author for this work
  • Beihang University

Research output: Contribution to journal › Conference article › peer-review

Abstract

The security of embedded systems has attracted much attention as they are being used in more and more fields. The rapid growth and pervasive use of embedded systems make it easier for a sophisticated attacker to gain physical access to launch physical attacks on insecure off-chip memory and bus. This paper presents a novel hardware-based security mechanism to protect confidentiality and integrity of data, preventing the system data from being stolen or tampered by a malicious attacker. The proposed mechanism protects the confidentiality of data using advanced encryption standard (AES) stream encryption algorithm in parallel with the memory access process. This mechanism provides integrity protection for data by attaching integrity signatures generated using hash algorithm to data stored in external memory. The signature is verified when data is fetched into the chip. The security architecture has been tested and validated on the system on a programmable chip (SoPC) with OR1200(processor based on OpenRISC1000 architecture) processor. The experimental result shows that the proposed security mechanism ensures the integrity and confidentiality of system data, introducing low performance penalties.

Original languageEnglish
Article number012070
JournalIOP Conference Series: Materials Science and Engineering
Volume466
Issue number1
DOIs
StatePublished - 28 Dec 2018
Event2018 2nd Annual International Conference on Cloud Technology and Communication Engineering, CTCE 2018 - Nanjing, China
Duration: 17 Aug 2018 → 19 Aug 2018

Fingerprint

Dive into the research topics of 'Hardware-Based Protection for Data Security at Run-Time on Embedded Systems'. Together they form a unique fingerprint.

Cite this