Towards interpretability in fingerprint based indoor positioning: May attention be with us
Abstract: Highlights•Interpretability in indoor positioning is defined and achieved through attention.•The model uses ranked-based fingerprinting, deep learning (RNN) and attention.•Attention has a strong spatial characterization in the indoor positioning context.•Different kinds of attention mechanisms exhibit rather different behaviours.•Attention contributes to multiple positioning-related tasks, e.g., improving. accuracy.
Loading