SPAR: Self-supervised Placement-Aware Representation Learning for Distributed Sensing
This addresses a central challenge in distributed sensing applications like vehicle monitoring and earthquake localization, offering a novel approach to improve representation learning.
The paper tackles the problem of sensor placement affecting signals in distributed sensing by proposing SPAR, a self-supervised framework that uses positional embeddings and dual reconstruction to learn placement-aware representations, achieving superior robustness and generalization in experiments on three real-world datasets.
We present SPAR, a framework for self-supervised placement-aware representation learning in distributed sensing. Distributed sensing spans applications where multiple spatially distributed and multimodal sensors jointly observe an environment, from vehicle monitoring to human activity recognition and earthquake localization. A central challenge shared by this wide spectrum of applications, is that observed signals are inseparably shaped by sensor placements, including their spatial locations and structural roles. However, existing pretraining methods remain largely placement-agnostic. SPAR addresses this gap through a unifying principle: the duality between signals and positions. Guided by this principle, SPAR introduces spatial and structural positional embeddings together with dual reconstruction objectives, explicitly modeling how observing positions and observed signals shape each other. Placement is thus treated not as auxiliary metadata but as intrinsic to representation learning. SPAR is theoretically supported by analyses from information theory and occlusion-invariant learning. Extensive experiments on three real-world datasets show that SPAR achieves superior robustness and generalization across various modalities, placements, and downstream tasks.