MEM-DnP-A Novel Energy Efficient Approach for Memory Integrity Detection and Protection in Embedded Systems

摘要

The pervasiveness of modern day embedded systems has led to the storing of huge amounts of sensitive information in them. These embedded devices often have to operate under insecure environments and are hence susceptible to software and physical attacks. Thus, security has been and will remain one of the prime concerns in the embedded systems. Although a lot of hardware and software techniques have been proposed to provide high levels of security, they are hampered by the trade-offs created by the design constraints in embedded systems. This paper presents a novel energy efficient approach for MEMory integrity Detection and Protection (MEM-DnP). The key feature of the proposed MEM-DnP is that it can be adaptively tuned to a memory integrity verification module by using a sensor module. This significantly reduces the energy overheads imposed on an embedded system as compared to the conventional memory integrity verification mechanisms. The simulation results show that the average energy saved in the combined detection and protection mechanism ranges from 85.5 % to 99.998 %. This is substantially higher compared to the results achieved in basecase simulations with traditional memory integrity verification techniques.