Device Parameter Search Using Exponential Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing binary search algorithms for finding device parameters in test and measurement instruments are inefficient due to long measurement times, inaccuracies caused by assuming exact measured values, and inability to account for errors, leading to non-recoverable results.
Innovation Solution
A method involving exponential modeling based on multiple measurements to determine a device's behavior, using an exponential function to refine the model iteratively and adjust the test and measurement instrument's behavior for improved accuracy and precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If binary search algorithm is used to find device parameters, then measurement robustness is improved, but measurement time increases significantly
Solution Approach 1:
The patent performs preliminary measurements to gather data about the device's actual behavior before conducting the search. This preliminary characterization allows the algorithm to use a more informed search strategy that adapts to the specific device being tested, reducing the number of iterations needed compared to a generic binary search.
Solution Approach 2:
The patent implements feedback mechanisms where measurement results are used to update the search algorithm's understanding of the device characteristics. This feedback loop allows the system to adjust its search strategy dynamically, improving efficiency while maintaining robustness across different device types.
2Device complexity
If binary search algorithm assumes exact measured values, then computational simplicity is improved, but measurement accuracy deteriorates due to unaccounted errors
Solution Approach 1:
The patent compensates for systematic measurement errors by characterizing the device's actual behavior in advance and using this information to correct subsequent measurements. This cushioning approach prevents error accumulation that would otherwise occur with the binary search's assumption of exact values.
Solution Approach 2:
The patent changes the search algorithm's parameters based on actual device measurements rather than assuming ideal conditions. By adapting the search parameters to match the real device characteristics, the system maintains computational simplicity while improving measurement accuracy.
3Measurement precision
If more measurements are taken to narrow down resolution, then parameter accuracy is improved, but measurement time increases
Solution Approach 1:
The patent implements a dynamic measurement strategy that adapts the number and spacing of measurements based on the device's characteristics and the current search state. This dynamic approach concentrates measurements where they provide the most information, achieving high accuracy with fewer total measurements than a static approach.
Solution Approach 2:
The patent moves beyond simple voltage sweeps by incorporating multiple measurement dimensions including different current levels, device states, and operational conditions. This multi-dimensional characterization achieves higher accuracy by capturing the device's full behavior rather than relying on numerous one-dimensional measurements.
Data Source
AI summary
In some implementations, a method may include acquiring a first plurality of measurements for the DUT, where the test and measurement instrument is not configured with the characteristic of the DUT and the first plurality of measurements may include measurements for the characteristic. In addition, the method may include determining an exponential model of the DUT based on the first plurality of measurements, where the exponential model is representative of behavior of the DUT. The method may include acquiring a second plurality of measurements different from the first plurality of measurements, where the second plurality of measurements may include measurements for the characteristic of the DUT. Moreover, the method may include verifying the exponential model based on the second plurality of measurements. Also, the method may include adjusting the behavior of the test and measurement instrument based on the verified exponential model of the DUT.