ML Measurement Translator for Test Instruments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-quality test and measurement instruments are expensive and provide accurate measurements, while lower-cost alternatives often result in measurement errors, leading to device failures and unnecessary scrapping due to their lower performance.
Innovation Solution
A machine learning-based measurement translator system that uses neural networks to translate measurements from lower-cost manufacturing test instruments to those of higher-performing reference instruments, considering both hardware and software differences, thereby enhancing measurement accuracy without the need for expensive equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If lower-cost test instruments are used, then cost is reduced, but measurement accuracy deteriorates
Solution Approach 1:
A machine learning-based measurement translator is introduced as an intermediary system between the lower-cost test instrument and the measurement results. The translator receives measurements from the affordable instrument and transforms them into accurate measurement values equivalent to those from expensive reference instruments, thereby resolving the contradiction between cost and accuracy
Solution Approach 2:
The system changes the parameter representation of measurements through machine learning transformation. By training neural networks on the relationship between measurements from different instrument types, the system recodes the measurement parameters from the lower-cost instrument to match the accuracy characteristics of high-cost instruments
2Ease of manufacture
If lower-cost instruments are used, then cost is reduced, but measurement errors increase
Solution Approach 1:
The measurement translator acts as a mediator that processes measurements from lower-cost instruments and transforms them into reliable results. The machine learning model compensates for systematic errors and discrepancies between instrument types, ensuring that measurements from affordable instruments achieve the same reliability as expensive ones
Solution Approach 2:
The system uses training data that includes feedback loops where measurements from reference instruments are compared with measurements from the lower-cost instruments. This feedback mechanism allows the machine learning model to learn and correct errors, improving measurement reliability
3Measurement precision
If expensive reference instruments are used for all measurements, then measurement accuracy is maintained, but cost increases
Solution Approach 1:
Instead of using expensive reference instruments for all measurements (excessive action), the system uses affordable instruments for routine measurements and only deploys expensive reference instruments when needed for calibration or validation. The machine learning translator enables this partial use of high-cost instruments while maintaining overall measurement accuracy
Data Source
AI summary
A test and measurement system includes a first test and measurement instrument having an input to allow the test and measurement instrument to receive signals from one or more devices under test (DUT), and one or more digitizers to convert the signals from the one or more DUTs to digital waveforms, a machine learning network, and one or more processors to: perform one or more measurements of the digital waveforms, send the one or more measurements of the digital waveforms to the machine learning network as an input, use the machine learning network to translate the one or more measurements to measurements made by a reference instrument to produce one or more translated measurements, the reference instrument being more accurate than the first test and measurement instrument, and determine whether the DUT meets a performance requirement based upon the one or more translated measurements.