Sensor System Data Enhancement Mapping Function
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low-quality sensors produce data that lacks the desired accuracy, sensitivity, range, and resolution, making them inadequate for various applications, while high-quality sensors are more expensive.
Innovation Solution
A processor-based system that determines and adjusts tunable parameters of a mapping function to enhance low-quality data samples, aligning them with high-quality data samples, thereby producing high-quality data outputs without the need for expensive high-quality sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If low-quality sensors are used to reduce cost, then device cost is reduced, but measurement precision deteriorates
Solution Approach 1:
A mapping function serves as an intermediary between low-quality sensor data and high-quality data representation. The mapping function transforms LQ data into a form that approximates HQ data characteristics, allowing the system to achieve high measurement precision while using low-cost sensors. The mapping function learns the transformation relationship during training and applies it during inference to bridge the quality gap.
Solution Approach 2:
The system changes the parameters of the mapping function through training to optimize the transformation from LQ to HQ data. During training, the mapping function's parameters are adjusted to minimize the difference between transformed LQ data and actual HQ data. This parameter optimization enables the system to achieve high measurement precision without using expensive HQ sensors.
2Measurement precision
If high-quality sensors are used to improve measurement precision, then data quality is improved, but device cost increases
Solution Approach 1:
Instead of using expensive high-quality sensors directly, the system creates a computational copy of HQ data by transforming LQ sensor data through a trained mapping function. This copying approach replicates the data quality characteristics of HQ sensors without requiring the physical HQ sensor hardware, thereby reducing device cost while maintaining measurement precision.
Solution Approach 2:
The system replaces expensive, durable high-quality sensors with inexpensive low-quality sensors combined with computational processing. The LQ sensors are used as disposable or low-cost components, while the value is added through software-based mapping function transformation, achieving HQ data quality without the high hardware cost.
3Ease of manufacture
If low-quality sensors are used to reduce cost, then device cost is reduced, but accuracy deteriorates
Solution Approach 1:
The mapping function acts as an intermediary that corrects accuracy deficiencies in LQ sensor data. By learning the relationship between LQ and HQ data during training, the mapping function compensates for accuracy losses and transforms LQ measurements into accurate representations of the physical quantity being measured.
Solution Approach 2:
The system optimizes the mapping function parameters during training to maximize accuracy of the transformed data. Through iterative adjustment of parameters, the system learns to accurately represent physical quantities from LQ sensor inputs, achieving measurement accuracy comparable to HQ sensors while using low-cost hardware.
4Ease of manufacture
If low-quality sensors are used to reduce cost, then device cost is reduced, but resolution deteriorates
Solution Approach 1:
The mapping function serves as an intermediary that enhances the resolution of LQ sensor data. By learning the transformation patterns during training, the mapping function recovers fine details and improves the effective resolution of measurements, allowing the system to achieve high resolution without using expensive high-resolution sensors.
Data Source
AI summary
A sensor system may comprise a sensor; a processor in electronic communication with the sensor; and/or a tangible, non-transitory memory configured to communicate with the processor, the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations. The operations may comprise recording, by the sensor, a low quality data sample; and/or applying, by the processor, a mapping function having a plurality of tuned parameters to the low quality data sample, producing a high quality data output.