作者: Liwei Zhou , Yiorgos Makris
DOI: 10.23919/DATE.2017.7927236
关键词:
摘要: We introduce a hardware-based methodology for performing on-line intrusion detection in microprocessors. The proposed method extracts fingerprints from the basic blocks of routine executed response to system call and examines their validity using Bloom filter. Implementation hardware renders spoofing attacks, which operating or hypervisor-level methods are vulnerable, ineffective. is evaluated kernel rootkits covertly modify service routines Linux running on 32-bit x86 architecture, implemented Simics simulation environment, while overhead predictive 45nm PDK.