Post-quantum TEE with GPU-accelerated Lattice Encryption

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current trusted execution environments (TEEs) are limited by their reliance on classical cryptography, which is vulnerable to quantum attacks, and they do not effectively utilize off-CPU accelerators like GPUs for enhanced security and performance.

Innovation Solution

The implementation of a post-quantum encrypted trusted execution environment on edge devices, which incorporates GPUs for accelerating lattice-based encryption and decryption operations within the TEE, ensuring secure processing and communication of sensitive data like AI model inferences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If classical cryptography is used in TEEs, then current security standards are maintained, but vulnerability to quantum attacks increases

Engineering Contradiction:

ImprovesecurityVSAvoidquantum attack vulnerability

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent changes the cryptographic parameter from classical cryptography to post-quantum cryptography, specifically using lattice-based encryption schemes. This parameter change maintains security reliability while eliminating vulnerability to quantum attacks, as lattice-based cryptography is designed to be resistant to both classical and quantum computational approaches.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If TEEs are implemented without GPU acceleration, then system complexity is lower, but encryption and decryption performance is insufficient

Engineering Contradiction:

Improveencryption performanceVSAvoidsystem architecture

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent merges the TEE functionality with GPU acceleration capabilities by implementing a hybrid architecture where the TEE encompasses both CPU and GPU units. This combination allows the system to leverage the parallel processing power of GPUs for encryption operations while maintaining the security guarantees of TEEs, thereby improving encryption performance without proportionally increasing complexity.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The TEE is designed to perform multiple functions: it provides secure execution environments, enables post-quantum encryption operations, and utilizes both CPU and GPU processing units. This multi-functionality allows a single system to handle diverse workloads from AI inferencing to cryptographic operations, improving overall productivity while managing complexity through unified architecture.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If GPUs are excluded from TEE, then TEE security isolation is simpler, but AI inferencing capability is limited

Engineering Contradiction:

ImproveAI processing capabilityVSAvoidTEE structure

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent combines AI inferencing capabilities with TEE security by implementing GPUs within the TEE boundary. This allows AI models to be executed securely within the trusted execution environment while maintaining security isolation. The GPU units inside the TEE can perform AI workloads encrypted with post-quantum cryptography, enabling versatile AI processing without compromising security isolation.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS12309263B2Post-quantum cryptography secured execution environments for edge devices

Publication Date: 2025.05.20 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Disclosed are techniques for post-quantum encrypted trusted execution environments on edge devices. An edge computing device includes a trusted execution environment that encompasses at least some SIMD processing units such as Graphics Processing Units (GPUs). A data record, such as machine learning inferences from a machine learning or artificial intelligence model, is generated on the edge computing device within the trusted execution environment and encrypted with post-quantum encryption (such as lattice based encryption) using SIMD processing units in the trusted execution environment. Workloads received for the trusted execution environment, also encrypted with post-quantum encryption, are decrypted using the SIMD processing units in the trusted execution environment.