[Paper Review] OAT: Attesting Operation Integrity of Embedded Devices

Zhichuang Sun, Bo Feng+2 more|arXiv (Cornell University)|Feb 9, 2018

Security and Verification in Computing40 references10 citations

TL;DR

This paper proposes OAT, a system for remote attestation of operation execution integrity (OEI) in ARM-based embedded devices, enabling detection of both control-flow hijacks and data-only attacks. OAT uses lightweight control-flow measurement and abstract execution for efficient, verifiable integrity checks with only 2.7% runtime overhead, allowing IoT backends to verify trustworthiness of device operations in real time.

ABSTRACT

Due to the wide adoption of IoT/CPS systems, embedded devices(IoT frontends) become increasingly connected and mission-critical, which in turn has attracted advanced attacks (e.g., control-flow hijacks and data-only attacks). Unfortunately, IoT backends are unable to detect if such attacks have happened while receiving data, service requests, or operation status from IoT devices. As a result, currently, IoT backends are forced to blindly trust the IoT devices that they interact with. To fill this void, we first formulate a new security property for embedded devices, called "Operation Execution Integrity" or OEI. We then design and build a system, OAT, that enables remote OEI attestation for ARM-based bare-metal embedded devices. Our formulation of OEI captures the integrity of both control flow and critical data involved in an operation execution. Therefore, satisfying OEI entails that an operation execution is free of unexpected control and data manipulations, which existing attestation methods cannot check. Our design of OAT strikes a balance between prover's constraints (embedded devices' limited computing power and storage) and verifier's requirements(complete verifiability and forensic assistance). OAT uses a new control-flow measurement scheme, which enables light-weight and space-efficient collection of measurements (97% space reduction from the trace-based approach). OAT performs the remote control-flow verification through abstract execution, which is fast and deterministic. OAT also features lightweight integrity checking for critical data (74% fewer instrumentation needed than previous work). Our security analysis shows that OAT allows remote verifiers or IoT backends to detect both control-flow hijacks and data-only attacks that affect the execution of operations on IoT devices. In our evaluation using real embedded programs, OAT incurs a runtime overhead of 2.7%.

Motivation & Objective

• To address the critical gap in remote trust verification for IoT devices, where backends currently blindly trust remote devices due to lack of runtime integrity checks.
• To formalize a new security property, Operation Execution Integrity (OEI), that captures both control-flow and critical data integrity during operation execution.
• To design a system that enables lightweight, efficient, and verifiable remote attestation on resource-constrained embedded devices without sacrificing security.
• To support forensic analysis by enabling reconstruction of attack execution traces post-incident, unlike traditional binary attestation outcomes.

Proposed method

• Formalize Operation Execution Integrity (OEI) as a security property requiring compliance with the operation's control flow graph and preservation of critical variable values across def-use chains.
• Design a new control-flow measurement scheme based on abstract execution that reduces measurement storage by 97% compared to trace-based methods.
• Implement a lightweight instrumentation technique for critical data integrity checks, reducing instrumentation overhead by 74% compared to prior work.
• Build OAT as a full-stack system with a customized compiler, runtime measurement engine on the device (prover), and a verification engine on the backend (verifier).
• Use unforgeable measurements and deterministic abstract execution to ensure complete verifiability and support postmortem forensic analysis.
• Integrate the system into real embedded programs and evaluate performance on ARM-based platforms.

Experimental results

Research questions

• RQ1 Can a remote verifier detect control-flow hijacks and data-only attacks on embedded devices that traditional attestation mechanisms fail to identify?
• RQ2 Is it feasible to achieve complete verifiability of operation integrity on resource-constrained embedded devices without excessive runtime or storage overhead?
• RQ3 Can a system provide both control-flow and data integrity verification with minimal instrumentation and measurement overhead?
• RQ4 Can the system support forensic reconstruction of attack traces to enable post-incident analysis?

Key findings

• OAT achieves 97% reduction in measurement storage space compared to trace-based control-flow measurement by using abstract execution.
• The system incurs only 2.7% runtime overhead on real embedded programs, demonstrating high efficiency on ARM-based devices.
• OAT reduces instrumentation requirements for data integrity checks by 74% compared to prior approaches, improving deployability.
• The system enables remote verification of both control-flow and data integrity, detecting attacks that existing methods miss, such as data-only attacks and ROP exploits.
• OAT supports forensic trace reconstruction, allowing verifiers to analyze attack behavior after detection, which is not supported by traditional binary attestation.
• Security analysis confirms that OAT can detect both control-flow hijacks and data-only attacks, ensuring that only genuine operation executions are accepted by IoT backends.