D1.1 GUM2ALA – Uncertainty Propagation Algorithm for the Adaptive Linear Approximation According to the GUM

Dorst, Tanja; Schneider, T.; Schütze, Andreas; Eichstädt, Sascha

doi:10.5162/smsi2021/d1.1

Cited by 6 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another reason for the application of data preprocessing is the reduction of data dimensionality. This may be necessary simply due to storage or data transfer bandwidth limitations [3]. In the project EMPIR Met4FoF, the previously developed Python library PyDynamic [16] was extended to include the data pre-processing steps typically required in IoT.…”

Section: Measurement Uncertainty In Sensor Network Data Processingmentioning

confidence: 99%

“…An important aspect that distinguishes sensor network applications from single sensor measurements is that rather than the individual sensors, the combined information from all sensors is the main object of interest. For instance, the combination of microphone data and vibration measurement in predictive maintenance provides more insights into the actual status of the monitored machine than the individual measurements alone [3]. A consequence of the focus on combined sensor data rather than individual sensors is that the definition of the quantity of interest, the measurand, is not straightforward.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fundamental aspects in sensor network metrology

et al. 2023

View full text Add to dashboard Cite

Sensor networks have appeared on the 'radar' of metrology only recently and a rigorous treatment and metrological assessment have not yet been established. However, sensor networks as measuring systems underpin many developments in digital transformation, with applications ranging from regulated utility networks to low-cost Internet of Things (IoT). The metrological assessment of sensor networks necessitates a fundamental revision of calibration, uncertainty propagation and performance assessment and new approaches for information and data handling regarding the individual sensors and their interactions in the network to allow a systems metrology approach to be established. This contribution introduces some initial findings from recent research and gives an outlook into future developments.

show abstract

Section: Measurement Uncertainty In Sensor Network Data Processingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fundamental aspects in sensor network metrology

et al. 2023

View full text Add to dashboard Cite

show abstract

“…This study uses the adaptive linear approximation as a feature extraction method [43]. Although the algorithm can identify the optimal number of splits, this time, the algorithm is forced to make exactly 49 splits for each sub-sensor independently, which ensures that every temperature step can be accurately reconstructed.…”

Section: Feature Extraction Selection and Regressionmentioning

confidence: 99%

Comparison of Transfer Learning and Established Calibration Transfer Methods for Metal Oxide Semiconductor Gas Sensors

et al. 2023

View full text Add to dashboard Cite

Although metal oxide semiconductors are a promising candidate for accurate indoor air quality assessments, multiple drawbacks of the gas sensors prevent their widespread use. Examples include poor selectivity, instability over time, and sensor poisoning. Complex calibration methods and advanced operation modes can solve some of those drawbacks. However, this leads to long calibration times, which are unsuitable for mass production. In recent years, multiple attempts to solve calibration transfer have been made with the help of direct standardization, orthogonal signal correction, and many more methods. Besides those, a new promising approach is transfer learning from deep learning. This article will compare different calibration transfer methods, including direct standardization, piecewise direct standardization, transfer learning for deep learning models, and global model building. The machine learning methods to calibrate the initial models for calibration transfer are feature extraction, selection, and regression (established methods) and a custom convolutional neural network TCOCNN. It is shown that transfer learning can outperform the other calibration transfer methods regarding the root mean squared error, especially if the initial model is built with multiple sensors. It was possible to reduce the number of calibration samples by up to 99.3% (from 10 days to approximately 2 h) and still achieve an RMSE for acetone of around 18 ppb (15 ppb with extended individual calibration) if six different sensors were used for building the initial model. Furthermore, it was shown that the other calibration transfer methods (direct standardization and piecewise direct standardization) also work reasonably well for both machine learning approaches, primarily when multiple sensors are used for the initial model.

show abstract

“…Another reason for the application of data pre-processing is the reduction of data dimensionality. This may be necessary simply due to storage or data transfer bandwidth limitations [3]. In the project EMPIR Met4FoF, the previously developed Python library PyDynamic [15] was extended to include the data pre-processing steps typically required in IoT.…”

Section: Measurement Uncertainty In Sensor Network Data Processingmentioning

confidence: 99%

“…An important aspect that distinguishes sensor network applications from single sensor measurements is that rather than individual sensors, the combined information from all sensors is the main object of interest. For instance, the combination of microphone data and vibration measurement in predictive maintenance provides more insights into the actual status of the monitored machine than the individual measurements alone [3]. A consequence of the focus on combined sensor data rather than individual sensors is that the definition of the quantity of interest, the measurand, is not straightforward.…”

Section: Introductionmentioning

confidence: 99%

Fundamental Aspects in Data Analysis for Sensor Network Metrology

Eichstädt¹,

Vedurmudi²,

Gruber³

et al. 2022

Proceedings of the First International IMEKO TC6 Conference on Metrology and Digital Transformation - M4Dconf2022

View full text Add to dashboard Cite

Sensor networks underpin many developments in digital transformation, with applications ranging from regulated utility networks to low-cost Internet of Things (IoT). The metrological assessment of sensor networks necessitates a fundamental revision of calibration, uncertainty propagation and performance assessment and new approaches for information and data handling regarding the individual sensors and their interactions in the network to allow a systems metrology approach to be established. This contribution summarizes initial findings from three research projects and gives an outlook into future developments.

show abstract

D1.1 GUM2ALA – Uncertainty Propagation Algorithm for the Adaptive Linear Approximation According to the GUM

Cited by 6 publications

References 0 publications

Fundamental aspects in sensor network metrology

Fundamental aspects in sensor network metrology

Comparison of Transfer Learning and Established Calibration Transfer Methods for Metal Oxide Semiconductor Gas Sensors

Fundamental Aspects in Data Analysis for Sensor Network Metrology

Contact Info

Product

Resources

About