Pluggable Hardware Barrier for verifying execution and preventing unauthorized outputs

Background

Computer systems today are incredibly complex and opaque, making it nearly impossible for users to fully verify the integrity and trustworthiness of all underlying hardware components. This opacity, coupled with globalized supply chains involving multiple parties, creates significant vulnerabilities. There is a critical need to ensure system integrity, data privacy, and reliable operation, as malicious alterations, subtle design flaws, or sophisticated hardware Trojans can compromise systems at a fundamental level, leading to unauthorized data leakage or incorrect computations.

Current approaches to ensuring hardware trustworthiness are largely inadequate against these pervasive threats. Existing verification methods often fail to guarantee comprehensive hardware integrity, particularly against sophisticated or latent attacks that can evade detection. They struggle to provide assurances against malicious modifications introduced during manufacturing or within the complex supply chain. Consequently, users lack reliable mechanisms to independently confirm that their systems are executing as intended, leaving them vulnerable to compromised hardware that can undermine privacy and system reliability without detection.

Technology

This technology introduces a pluggable trusted hardware element positioned between an untrusted processor and external peripherals to ensure system trustworthiness. This element independently re-executes instructions based on the processor’s trace information, comparing its computed results against the processor’s reported outputs. For memory operations, it employs cryptographic Message Authentication Codes (MACs) to verify data integrity during both stores and loads, and also validates program code integrity. Only upon successful verification of execution and data integrity does this hardware element permit outputs to be communicated to external devices, thereby preventing unauthorized or incorrect information from leaving the system. It is designed to be simple and separately manufactured for independent verification.

This technology is differentiated by establishing trust in systems where underlying hardware components, including processors and memory, cannot be inherently trusted due to potential malicious alterations or supply chain compromises. Unlike traditional methods, it provides independent verification by acting as a simple, verifiable gatekeeper that can be deployed post-manufacturing. Its pluggable nature allows retrofitting existing systems, and its design minimizes performance impact by decoupling execution from verification. This approach offers comprehensive protection against diverse hardware and software attacks, ensuring system integrity and privacy by strictly controlling external outputs based on verified execution, a critical capability in complex, opaque computing environments.

Benefits

• Establishes system trustworthiness despite untrusted hardware components.
• Enhances supply chain security by verifying hardware integrity.
• Prevents unauthorized or incorrect data from leaving the system.
• Ensures data and program code integrity through cryptographic verification.
• Minimizes performance impact on the main processor.
• Offers flexibility through a pluggable and upgradable design.

Applications

• Secure critical infrastructure
• Secure hardware supply chain
• Secure IoT devices
• Secure cloud infrastructure
• Protect financial transactions

Additional info

A pluggable hardware element verifies untrusted processor execution by independently re-executing instructions from trace data and comparing results. It employs cryptographic Message Authentication Codes for memory and program code integrity. This barrier circuit permits outputs to external devices only upon successful verification, preventing unauthorized or incorrect information from leaving the system and ensuring overall trustworthiness.