Abstract
Trusted computing allows attesting remote system’s trustworthiness based on the software stack whose integrity has been measured. However, attacker can corrupt system as well as measurement operation. As a result, nearly all integrity measurement mechanism suffers from the fact that what is measured may not be same as what is executed. To solve this problem, a novel integrity measurement called dynamic instruction trace measurement (DiT) is proposed. For DiT, processor’s instruction cache is modified to stores back instructions to memory. Consequently, it is designed as a assistance to existing integrity measurement by including dynamic instructions trace. We have simulated DiT in a full-fledged system emulator with level-1 cache modified. It can successfully update records at the moment the attestation is required. Overhead in terms of circuit area, power consumption, and access time, is less than 3% for most criterions. And system only introduces less than 2% performance overhead in average.
Highlights
Nowadays, computer under different platforms interacts with each other through internet environment
In order to narrow down the range of such measures, Trusted Computer Base (TCB) is defined as hardware components and/or software modules whose integrity decides the status of a whole platform
In order to provide updated integrity measurement as system evolves, we propose an original dynamic instruction trace measurement (DiT) to include in the metric dynamic instructions-level behaviour in the processor with the help of simple micro-architecture modification
Summary
Computer under different platforms interacts with each other through internet environment. Integrity measurement is strongly related to security policy applied to specific computer system and results in different attestation mechanism. A challenger (the remote entity which is interested in attesting the system) should be able to tell the system’s updated security-related capability such as whether the memory has been ever corrupted by attacker, or whether programs can be properly executed in isolation, or whether cryptography keys are securely stored, and so on. As different attacks occur from internet, this is becoming less sufficient for a remote challenger For those programs running for a long time, such as server programs, a static measurement prior to execution may have little relation to the system status at the current moment. Validation of system against a certain security policy introduces little performance overhead This policy-driven attestation or validation schemes are largely based on limited information specific per intended attack scenarios.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
