Security and Privacy in Communication Networks. 13th International Conference, SecureComm 2017, Niagara Falls, ON, Canada, October 22–25, 2017, Proceedings

Research Article

Gray-Box Software Integrity Checking via Side-Channels

Download
207 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-78813-5_1,
        author={Hong Liu and Eugene Vasserman},
        title={Gray-Box Software Integrity Checking via Side-Channels},
        proceedings={Security and Privacy in Communication Networks. 13th International Conference, SecureComm 2017, Niagara Falls, ON, Canada, October 22--25, 2017, Proceedings},
        proceedings_a={SECURECOMM},
        year={2018},
        month={4},
        keywords={Security Embedded systems Side-channel analysis Software attestation Soft-core processors FPGA},
        doi={10.1007/978-3-319-78813-5_1}
    }
    
  • Hong Liu
    Eugene Vasserman
    Year: 2018
    Gray-Box Software Integrity Checking via Side-Channels
    SECURECOMM
    Springer
    DOI: 10.1007/978-3-319-78813-5_1
Hong Liu1,*, Eugene Vasserman1,*
  • 1: Kansas State University
*Contact email: hongl@ksu.edu, eyv@ksu.edu

Abstract

Enforcing software integrity is a challenge in embedded systems which cannot employ modern protection mechanisms. In this paper, we explore feasibility of software integrity checking from measuring passive electromagnetic emissions of FPGA-implemented SoCs. We show that clock-cycle-accurate side-channel models can be built by utilizing gray-box analysis and regression techniques. The generality and effectiveness of our methods are shown by three different SoCs, profiled and tested on different chips of the same model. Our technique is non-invasive, and does not interrupt normal execution or change hardware/software configuration of the target device, making it particularly attractive for already-deployed systems.