Trusted Job Execution Across Distributed Processing Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced driver assistance systems (ADAS) and autonomous vehicle (AV) systems face challenges in protecting data generated by object classifiers, which are crucial for detecting objects in a vehicle's environment, especially as these systems progress towards fully autonomous operation.
Innovation Solution
The implementation of a system that includes cameras and a processor to monitor the vehicle's environment, using neural networks or deep learning algorithms to estimate a future path based on images, while also incorporating additional sensors and a distributed job allocation scheme to manage processing units securely and efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is protected through secure distributed execution, then security is improved, but processing speed may deteriorate
Solution Approach 1:
The patent divides the processing system into multiple isolated processing units, each executing jobs in separate trusted execution environments. This segmentation allows parallel processing across multiple units while maintaining security isolation, thus improving both security and maintaining processing speed through concurrent execution.
Solution Approach 2:
The patent introduces a job manager as an intermediary component that coordinates job allocation between the controller and processing units. This mediator optimizes job distribution and manages data flow efficiently, ensuring secure execution while minimizing overhead and maintaining high processing throughput.
2Productivity
If distributed job allocation is implemented, then processing efficiency is improved, but system complexity worsens
Solution Approach 1:
The patent designs processing units with universal trusted execution environments that can handle multiple types of jobs and data processing tasks. This multi-functionality allows a single distributed architecture to manage diverse workloads efficiently, reducing the need for specialized components and simplifying the overall system while maintaining high productivity.
3Measurement precision
If neural networks are used for path estimation, then navigation accuracy is improved, but computational resource consumption worsens
Solution Approach 1:
The patent segments neural network processing into discrete jobs that can be distributed across multiple processing units. Each unit executes specific inference tasks in isolated trusted environments, enabling parallel processing of image data and path estimation. This distribution reduces the computational burden on single units and optimizes energy consumption while maintaining high accuracy through collective processing power.
Data Source
AI summary
A processing unit, where the processing unit one of a group of processing units of a system, includes a processor; and memory including instructions, which when executed by the processor while avoiding interrupting a controller that does not belong to the group of processing units, cause the processor to: perform at least one iteration of the steps of: (a) entering a trusted mode, (b) selecting a selected job to be executed by the processing unit, (c) retrieving access control metadata related to the selected job, (d) entering, by the processing unit, an untrusted mode, (e) executing the selected job by the processing unit while adhering to the access control metadata related to the job, and (f) resetting the processing unit.