Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A Cross-Domain Benchmark of Intrinsic and Post Hoc Explainability for 3D Deep Learning Models.

Journal of imaging·2026
Same author

Deep learning based phase retrieval with complex beam shapes for beam shape correction.

Optics express·2025
Same author

Use of Domain Labels during Pre-Training for Domain-Independent WiFi-CSI Gesture Recognition.

Sensors (Basel, Switzerland)·2023
Same author

A Codeword-Independent Localization Technique for Reconfigurable Intelligent Surface Enhanced Environments Using Adversarial Learning.

Sensors (Basel, Switzerland)·2023
Same author

Selfish Behavior in IEEE 802.11ah Networks: A Detection Algorithm and Mitigation Strategies.

Sensors (Basel, Switzerland)·2022
Same author

A Domain-Independent Generative Adversarial Network for Activity Recognition Using WiFi CSI Data.

Sensors (Basel, Switzerland)·2021
Same journal

RETRACTED: Zhang et al. A Novel Framework for Reconstruction and Imaging of Target Scattering Centers via Wide-Angle Incidence in Radar Networks. <i>Sensors</i> 2025, <i>25</i>, 6802.

Sensors (Basel, Switzerland)·2026
Same journal

Enhancing Unsupervised Multi-Source Domain Adaptation for Person Re-Identification via Mixture of Experts and Graph-Based Relation.

Sensors (Basel, Switzerland)·2026
Same journal

Development of an Instrumented Glove for Palmar Pressure Assessment in Kayakers.

Sensors (Basel, Switzerland)·2026
Same journal

Development and Experimental Validation of an Autonomous IoT-Based Monitoring System for Real-Time Water Quality Assessment in the Amazon River.

Sensors (Basel, Switzerland)·2026
Same journal

Semi-Supervised Adversarial Learning Framework for Controller Area Network Bus Intrusion Detection.

Sensors (Basel, Switzerland)·2026
Same journal

Smart Optimization Method for Safety Signs in Innovative Manufacturing Environments Integrating Industrial Field IoT Sensors and Knowledge Graphs.

Sensors (Basel, Switzerland)·2026
See all related articles

Related Experiment Video

Updated: Jul 9, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

558

Mini-Batch Alignment: A Deep-Learning Model for Domain Factor-Independent Feature Extraction for Wi-Fi-CSI Data.

Bram van Berlo1, Camiel Oerlemans1, Francesca Luigia Marogna1

  • 1Department of Mathematics and Computer Science, Eindhoven University of Technology, P.O. Box 513, 5600 MB Eindhoven, The Netherlands.

Sensors (Basel, Switzerland)
|December 9, 2023
PubMed
Summary
This summary is machine-generated.

Mini-batch alignment is a new technique for Wi-Fi sensing that helps models generalize better by reducing sensitivity to domain factors. This method shows promise in improving gesture recognition without needing domain labels or extensive retraining.

Keywords:
Wi-Fi–CSIdevice-free sensingdomain adaptationdomain shiftdomain-independent learningunobtrusive sensing

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.2K
Successful In vivo Calcium Imaging with a Head-Mount Miniaturized Microscope in the Amygdala of Freely Behaving Mouse
09:39

Successful In vivo Calcium Imaging with a Head-Mount Miniaturized Microscope in the Amygdala of Freely Behaving Mouse

Published on: August 26, 2020

12.3K

Related Experiment Videos

Last Updated: Jul 9, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

558
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.2K
Successful In vivo Calcium Imaging with a Head-Mount Miniaturized Microscope in the Amygdala of Freely Behaving Mouse
09:39

Successful In vivo Calcium Imaging with a Head-Mount Miniaturized Microscope in the Amygdala of Freely Behaving Mouse

Published on: August 26, 2020

12.3K

Area of Science:

  • Computer Science
  • Machine Learning
  • Signal Processing

Background:

  • Unobtrusive sensing aims to integrate sensing capabilities into daily life by repurposing existing communication technologies.
  • Wireless Fidelity (Wi-Fi) sensing, utilizing Channel State Information (CSI), is a promising approach due to ubiquitous Wi-Fi networks.
  • A significant challenge is the sensitivity of CSI data to domain factors (e.g., subject position/orientation), leading to domain shifts and poor inference generalization.

Purpose of the Study:

  • To introduce and evaluate 'mini-batch alignment', a novel domain factor-independent feature-extraction pipeline for Wi-Fi sensing.
  • To test the hypothesis that mini-batch alignment can eliminate the need for domain labels, reduce retraining, and save computational resources.
  • To assess the effectiveness of mini-batch alignment in mitigating domain shifts in gesture recognition tasks.

Main Methods:

  • Developed a feature-extraction pipeline called 'mini-batch alignment' that trains models to be invariant to intermediate feature-probability density variations across data batches.
  • Conducted extensive experiments using the SignFi and Widar3 gesture recognition datasets.
  • Evaluated performance using one- and two-domain-factor leave-out cross-validation with Doppler Frequency Spectrum (DFS) and Gramian Angular Difference Field (GADF) as input types, comparing against existing domain-shift mitigation techniques.

Main Results:

  • Mini-batch alignment performed comparably to other domain-shift mitigation techniques on the Widar3 dataset with DFS input for one-domain cross-validation (position/orientation).
  • With a memory-optimized GADF input, mini-batch alignment showed potential in recovering baseline model performance without performance loss from weight steering in one- and two-domain cross-validation scenarios.
  • Despite promising results, the experiments did not provide sufficient evidence to fully validate the mini-batch alignment hypothesis due to dataset limitations, probability distribution assumptions, and scaling issues.

Conclusions:

  • Mini-batch alignment demonstrates potential as a domain-shift mitigation technique in Wi-Fi sensing for gesture recognition.
  • Further research is needed with improved benchmark datasets and refined methods to address identified pitfalls and fully validate the hypothesis.
  • The study highlights the challenges in achieving robust and generalizable device-free sensing systems.