Data Eye CNN Scoring for Unstable Link Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data communication links in computing devices, such as those used in autonomous vehicles and IoT systems, face reliability issues due to noise and environmental stressors, which can distort the data eye and lead to inaccurate data sampling, and current training methods do not effectively determine when links have become severely impaired.
Innovation Solution
A method using a convolutional neural network to collect and analyze a two-dimensional array of data points representing the data eye, determining a score that compares to a threshold, and initiating actions based on this comparison to maintain link reliability, including potential fail-over to alternate subsystems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data link training is performed periodically to re-align clock edge with data eye center, then timing alignment is improved, but the system cannot detect when the link has become severely distorted and training becomes ineffective
Solution Approach 1:
The patent replaces traditional mechanical/electronic data eye analysis methods with a vision-based system using cameras to capture and analyze data eye diagrams. This substitution enables more precise measurement of data eye characteristics and allows detection of severe distortion conditions where conventional training methods fail.
Solution Approach 2:
The patent introduces an intermediary detection system that monitors data eye quality independently of the training mechanism. This intermediary system provides early warning of link degradation before training becomes ineffective, allowing proactive maintenance actions to be taken.
2Ease of operation
If conventional data link training methods are used, then timing alignment can be maintained under normal conditions, but the system lacks the capability to detect severe impairment requiring maintenance
Solution Approach 1:
The patent substitutes conventional electronic monitoring with vision-based data eye capture and analysis. This enables automated timing alignment while simultaneously providing detailed visual measurement capabilities to detect severe link impairment that would be difficult to identify with traditional methods.
3Productivity
If high-speed data communication is implemented to meet increasing computing demands, then processing capability is improved, but noise and environmental stressors cause signal distortion and reliability degradation
Solution Approach 1:
The patent uses vision-based measurement to characterize signal distortion caused by noise and environmental stressors in high-speed data communication systems. This substitution enables precise quantification of harmful effects without interfering with the high-speed operation.
Solution Approach 2:
The patent implements a feedback mechanism where captured data eye images are analyzed to assess the impact of noise and environmental stressors. This information feeds back into the system to trigger maintenance actions before reliability is compromised, allowing continuous operation at high speeds despite adverse conditions.
Data Source
AI summary
The reliability of a data communication link may be analyzed and otherwise maintained by collecting a two-dimensional array representing a functional data eye, and using a convolutional neural network to determine a score of the functional data eye. The determined score may be compared with a threshold, and an action may be initiated based on the result of the comparison.