Detecting Code Alteration by Creating a Temporary Memory Bottleneck

Abstract

Abstract-We develop a new technique whereby a poll worker can determine whether the software executing on electronic voting machines on election day has been altered from its factory version. Our generalized approach allows a human, using a known challenge-response pair, to detect attacks that involve modification or replacement of software on a computer based on the time it takes the computer to provide a correct response to a challenge. We exploit the large difference between main memory access times and cache memory access or CPU clock cycle times to significantly increase the time required to compute the right response when the software has been changed.

Security and privacy challenges in industrial internet of things

“…Common to all of them is that the device to be attested, called prover, sends a status report of its current software configuration to another device, called verifier, to demonstrate that it is in a known and, thus trustworthy, state. Since malicious software on the prover's platform could forge this report, its authenticity is typically assured by secure hardware [49,67,48,14,29,56,26] and/or trusted software [3,24,60,59,58,57,16,32,29,68]. Attestation based on secure hardware is most suitable for advanced computing platforms, such as smartphones, tablets, laptops, personal computers, and servers.…”

“…However, the underlying security hardware is often too complex and/or expensive for low-end embedded systems. In contrast, software-based attestation [24,60,59,58,57,16,32], does not require secure hardware or cryptographic secrets. However, security guarantees of software-based attestation are often unclear since it usually relies on strong assumptions, such as (1) the adversary being passive while the attestation protocol is executed, and (2) optimality of the attestation algorithm and its implementation.…”

A security framework for the analysis and design of software attestation

“…Several works consider the strengthening of checksum algorithms and their implementations against unintended modifications by either limiting the memory available to the prover during attestation [10,29] or by using self-modifying and/or obfuscated attestation algorithms [25,11]. Many works investigate the suitability and extension of software attestation to a variety of computing platforms, including sensors, peripherals and voting machines [21,17,9,23,16].…”

SobTrA

“…Either by generating the checksum function itself as the challenge [11] or by introducing a memory bottleneck during the checksum calculation to slow down the attacker's code [5,7]. Downside of the first approach is that it reintroduces a higher variance for the checksum function execution time.…”