Secure execution environment using post-quantum cryptography for edge devices
A lattice-encrypted TEE accelerated by GPUs addresses the limitations of current TEEs by securing AI inferences and workloads with post-quantum cryptography, ensuring confidentiality and resistance to quantum attacks, suitable for edge devices.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-02-11
- Publication Date
- 2026-04-07
AI Technical Summary
Current trusted execution environments (TEEs) are limited to central processing units (CPUs) and lack resistance to quantum computing attacks, and existing edge devices do not effectively utilize graphics processing units (GPUs) for enhanced security and parallel processing of AI and machine learning tasks.
Implementing a lattice-encrypted trusted execution environment accelerated by GPUs for edge devices, utilizing post-quantum cryptography to secure and process AI inferences and workloads, with encryption and decryption operations performed by GPUs within the TEE.
Provides enhanced security and confidentiality for AI inferences and workloads by ensuring data remains encrypted outside the TEE, with efficient parallel processing and resistance to quantum attacks, suitable for sensitive applications like battlefield communications and IoT devices.
Smart Images

Figure 0007841824000001 
Figure 0007841824000002 
Figure 0007841824000003
Abstract
Citation Information
Patent Citations
Trusted execution environment construction method and system for graphics processor
CN111949369A
Device, secure element, program, information processing method and information processing system
JP2019109680A
Edge system, control method of edge system, computer program used to control edge system, and recording medium thereof
JP2021010129A
Protection of privacy and data on smart edge devices
US20200134230A1
Elastic power scaling
US20210004265A1