Electromagnetic interference filter for spacecraft power bus

Zhechev, Yevgeniy S.; Kosteletskii, V P; Zabolotsky, A. M.; Gazizov, Talgat R.

doi:10.1088/1757-899x/560/1/012133

Cited by 5 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering the high electromagnetic noise interference environment on a spacecraft 46 , and the physical activity context in which the ECG data might be recorded, a foundation of this work was to develop an effective denoising procedure tailored to the deep-space edge computing context. To enable proper inference, this denoising algorithm detected and removed noise without compromising important biometric data, as shown in Fig.…”

Section: Denoising and Feature Extractionmentioning

confidence: 99%

Machine learning workflow for edge computed arrhythmia detection in exploration class missions

Mani,

Paul,

Archambault

et al. 2024

npj Microgravity

View full text Add to dashboard Cite

Deep-space missions require preventative care methods based on predictive models for identifying in-space pathologies. Deploying such models requires flexible edge computing, which Open Neural Network Exchange (ONNX) formats enable by optimizing inference directly on wearable edge devices. This work demonstrates an innovative approach to point-of-care machine learning model pipelines by combining this capacity with an advanced self-optimizing training scheme to classify periods of Normal Sinus Rhythm (NSR), Atrial Fibrillation (AFIB), and Atrial Flutter (AFL). 742 h of electrocardiogram (ECG) recordings were pre-processed into 30-second normalized samples where variable mode decomposition purged muscle artifacts and instrumentation noise. Seventeen heart rate variability and morphological ECG features were extracted by convoluting peak detection with Gaussian distributions and delineating QRS complexes using discrete wavelet transforms. The decision tree classifier’s features, parameters, and hyperparameters were self-optimized through stratified triple nested cross-validation ranked on F1-scoring against cardiologist labeling. The selected model achieved a macro F1-score of 0.899 with 0.993 for NSR, 0.938 for AFIB, and 0.767 for AFL. The most important features included median P-wave amplitudes, PRR20, and mean heart rates. The ONNX-translated pipeline took 9.2 s/sample. This combination of our self-optimizing scheme and deployment use case of ONNX demonstrated overall accurate operational tachycardia detection.

show abstract

Section: Denoising and Feature Extractionmentioning

confidence: 99%

Machine learning workflow for edge computed arrhythmia detection in exploration class missions

Mani,

Paul,

Archambault

et al. 2024

npj Microgravity

View full text Add to dashboard Cite

show abstract

“…Considering the high electromagnetic noise interference environment on a spacecraft 38 , and the physical activity context in which the ECG data might be recorded, a foundation of this work was to define an effective denoising procedure. To enable proper inference, this developed denoising algorithm detects and removes noise without compromising important biometric data, as shown in Fig.…”

Section: Denoising and Feature Extractionmentioning

confidence: 99%

Machine-Learning Workflow for Edge Computed Arrhythmia Detection in Exploration-Class Missions

Mani,

Paul,

Archambault

et al. 2023

Preprint

View full text Add to dashboard Cite

Deep-space missions require preventative care methods based on predictive models for identifying in-space pathologies. Deploying such models requires flexible edge computing, which Open Neural Network Exchange (ONNX) formats enable by optimizing inference directly on wearable edge devices. This work demonstrates an innovative approach to point-of-care machine learning model pipelines by combining this capacity with an advanced self-optimizing training scheme to classify periods of Normal Sinus Rhythm (NSR), Atrial Fibrillation (AFIB) and Atrial Flutter (AFL). 742 hours of electrocardiogram (ECG) recordings were pre-processed into 30-second normalized samples where variable mode decomposition purged muscle artifacts and instrumentation noise. Seventeen heart rate variability and morphological ECG features were extracted by convoluting peak detection with Gaussian distributions and delineating QRS complexes using discrete wavelet transforms. The decision tree classifier’s features, parameters and hyperparameters were self-optimized through stratified triple nested cross-validation ranked on F1-scoring against cardiologist labelling. The selected model achieved a macro F1-score of 0.899 with 0.993 for NSR, 0.938 for AFIB and 0.767 for AFL. The most important features included median P-wave amplitudes, PNN20, and mean heart rates. The ONNX-translated pipeline took 9.2 s/sample. The novel combination of our self-optimizing scheme and deployment use case of ONNX demonstrated overall accurate operational tachycardia detection.

show abstract

Quasi-static analysis of a two-conductor modal filter with a thin passive conductor

Zhechev

Zabolotsky

Gazizov

2021

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The paper presents the results of the quasi-static analysis of a two-conductor modal filter with a thin passive conductor. The geometrical parameters of the structure under investigation were optimized using a genetic algorithm. The authors performed the quasi-static simulation in the time domain with and without losses. It revealed that with a significant reduction in the thickness of the passive conductor it is possible to preserve or improve the electromechanical properties of the modal filter. The authors showed that in the proposed configuration the difference of per-unit-length delays increased from 0.43 to 0.93 ns, and the maximum output voltage decreased from 501 to 498 mV.

show abstract

Electromagnetic interference filter for spacecraft power bus

Cited by 5 publications

References 4 publications

Machine learning workflow for edge computed arrhythmia detection in exploration class missions

Machine learning workflow for edge computed arrhythmia detection in exploration class missions

Machine-Learning Workflow for Edge Computed Arrhythmia Detection in Exploration-Class Missions

Quasi-static analysis of a two-conductor modal filter with a thin passive conductor

Contact Info

Product

Resources

About