Decimated Sweep Measurements Using Neural Networks for DUT Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current test and measurement systems require extensive full sweep measurements for devices under test, which are time-consuming and inefficient, necessitating a need for faster data collection methods while maintaining accuracy for pass/fail determinations.
Innovation Solution
The system employs a machine learning approach by training a neural network on full sweep data to associate pass/fail labels with a decimated set of measurements, allowing for reduced data collection without compromising accuracy, using tensor building and normalization processes to prepare inputs for the neural network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full sweep measurements are performed on all data points, then measurement accuracy is maintained, but measurement time increases significantly
Solution Approach 1:
The system performs a preliminary full sweep measurement to collect training data, then uses this data to train a neural network model. The trained model can subsequently make accurate pass/fail predictions on new devices using only decimated measurements, eliminating the need to perform full sweep measurements for each new device while maintaining high accuracy.
Solution Approach 2:
The neural network model creates a virtual copy of the full sweep measurement process. Instead of physically performing measurements on all data points, the model replicates the decision-making capability of full sweep measurements by learning from training data, allowing accurate predictions with fewer actual measurements.
2Productivity
If the number of measurements is reduced to speed up data collection, then measurement time decreases, but measurement accuracy deteriorates
Solution Approach 1:
The system performs preliminary full sweep measurements during a training phase to build a comprehensive data set. This preliminary action creates a knowledge base that enables accurate predictions with decimated measurements, allowing the system to achieve both high productivity and high accuracy in the operational phase.
Solution Approach 2:
The system changes the measurement parameters by transitioning from measuring all data points to measuring only a decimated subset. The neural network compensates for this parameter change by learning the relationships between decimated measurements and full sweep outcomes, maintaining accuracy while improving productivity.
Data Source
AI summary
A test and measurement instrument includes one or more ports to allow the test and measurement instrument to receive a signal from a device under test (DUT), a user interface to allow the user to send inputs to the test and measurement instrument and receive results, and one or more processors configured to acquire the signal from the DUT, make measurements on the signal to create a decimated measurement set, convert the decimated measurement set into a tensor, send the tensor to a machine learning network, and receive a pass/fail value from the machine learning network. A method includes acquiring a signal from a device under test (DUT), making measurements on the signal to create a decimated measurement set, convert the decimated measurement set into a tensor, sending the tensor to a machine learning network, and receiving a pass/fail value from the machine learning network.