Trusted Execution Environment Policy Ingestion via Segmented Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Untrusted computing platforms can introduce malicious code into compositions, compromising the security of trusted execution environments (TEEs) during the authoring process, as they lack verification mechanisms to detect and prevent such threats.
Innovation Solution
Segmenting the authoring environment into an untrusted computing platform for composition creation and a trusted execution environment (TEE) for verification, using secure I/O and limited peripherals for user authentication and integrity checks, ensuring compositions are verified and signed before ingestion into the runtime TEE.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If compositions are authored on untrusted computing platforms, then comprehensive functionality and ease of authoring are provided, but malicious code can be introduced into compositions without detection
Solution Approach 1:
The system segments the authoring process into two distinct environments: an untrusted rich execution environment (REE) for composition creation with full functionality, and a trusted execution environment (TEE) for verification. The composition is divided into a tainted version created in REE and a clean version verified in TEE, allowing comprehensive authoring capabilities while ensuring integrity through separate verification
Solution Approach 2:
The patent introduces an intermediary verification process where the composition created in the untrusted REE is transferred to and verified in the trusted TEE. This intermediary step acts as a mediator that validates the composition's integrity, detecting any malicious code introduced during authoring before the composition is deployed to production
2Reliability
If the entire authoring environment is moved into TEE, then composition integrity is ensured, but device complexity and cost increase significantly
Solution Approach 1:
Instead of moving the entire authoring environment into TEE, the system segments functionality by keeping composition authoring in REE and only moving verification into TEE. This segmentation allows the complex authoring tools to remain in the untrusted environment while the critical verification function resides in the trusted environment, reducing overall system complexity
Solution Approach 2:
The patent extracts only the essential verification functionality into the TEE, leaving the complex authoring tools, editors, and development environments in the REE. This extraction approach minimizes the TEE's complexity while maintaining integrity, as only the critical verification step requires trusted execution
3Productivity
If comprehensive authoring tools are provided in untrusted environment, then productivity is enhanced, but vulnerability to malware increases
Solution Approach 1:
The system segments the authoring workflow into creation and verification phases, allowing comprehensive tools to operate in REE for high productivity while the verification phase in TEE provides security protection against malware infiltration
Solution Approach 2:
The patent applies preliminary anti-action by performing verification in the trusted TEE before the composition is deployed to production. This preemptive verification detects and prevents malicious code introduced by malware during authoring, countering the security threat before it can affect the production system
Data Source
AI summary
Before a composition is ingested into a runtime environment at a runtime device, the composition may be verified at an authoring trusted execution environment (TEE) operating on an authoring device. A user can operate an untrusted computing platform (e.g., a personal computer, laptop computer, tablet computer, etc.) to write code, generate data, or create some other composition. Since this composition is created on an untrusted device, the authoring TEE may output the composition on a trusted peripheral device to a user for review and approval. Responsive to receiving approval at the trusted peripheral device, the authoring TEE can sign the composition with a local key and forward the composition for execution by the runtime device. The signature can be utilized by the runtime device to prove that it was reviewed and verified by an authorized user operating the authoring device.


