DNN-Based Indoor Localization Under Limited Dataset Using GANs and Semi-Supervised Learning

This work addresses indoor localization when labeled training data is limited. By combining generative adversarial networks (GANs) for data augmentation with semi-supervised learning, the deep neural network localizer attains competitive accuracy using a fraction of the labeled samples required by fully-supervised baselines.

Recommended citation: W. Njima, A. Bazzi and M. Chafii, "DNN-Based Indoor Localization Under Limited Dataset Using GANs and Semi-Supervised Learning," in IEEE Access, vol. 10, pp. 69896-69909, 2022. https://ieeexplore.ieee.org/abstract/document/9812625

@article{njima2022dnn-based,
  title   = {DNN-Based Indoor Localization Under Limited Dataset Using GANs and Semi-Supervised Learning},
  author  = {Wafa Njima and Ahmad Bazzi and Marwa Chafii},
  journal = {IEEE Access},
  volume  = {10},
  pages   = {69896--69909},
  year    = {2022},
  month   = {jul},
  publisher = {IEEE},
  url     = {https://therealbazzi.github.io/publication/dnn-indoor-localization-gan-access22},
}