Human-centric visual monitoring of multi-clients system behavior and BiT for trust computing

Eun Ha Song, Su Hyun Yang, Young Sik Jeong

Research output: Contribution to journalArticlepeer-review

Abstract

This paper proposes the trusted platform board monitoring (TPBM) system that enabled effectively detecting and managing abnormal phenomenon based on the trusted platform board not only for security but also for multiple clients connected on the web. In other words, TPBM strengthens weak points of the existing security and provides the function to monitor with human-centric method not only system resources and process status for remote multi-client systems that are operated based on hardware security but also behavior of multiple clients in the cloud computing environment. In addition, measuring instruction level behavior is potentially a more effective protection than a pure software approach. From program's run-time behavior, system state safety can also be analyzed. However, dynamic characteristic of program trace also introduces challenges on what to measure and how to measure. To solve such problems, branch instruction trace (BiT) profiling tool is introduced to measure processor's instruction level run-time behavior. However, the BiT has provided only the text results of logical error information but not the visual monitoring function by tracing abnormal branch at the level of instruction. In order to resolve this problem, this paper includes the human-centric visual monitoring function for information on abnormal branch trace to TPBM.

Original languageEnglish
Pages (from-to)1441-1453
Number of pages13
JournalSecurity and Communication Networks
Volume7
Issue number10
DOIs
StatePublished - 1 Oct 2014

Keywords

  • Branch prediction
  • Execution path
  • Human-centric trusted visual monitoring
  • Indirect branch
  • Non-control data attack
  • Software attacks
  • Trusted platform module

Fingerprint

Dive into the research topics of 'Human-centric visual monitoring of multi-clients system behavior and BiT for trust computing'. Together they form a unique fingerprint.

Cite this