Learning Displacement-Aware WiFi Representations for Weakly Supervised Relative Localization

TL;DR

This paper introduces a novel weakly supervised learning framework for WiFi-based relative indoor localization, focusing on directly estimating displacement between traces without dense position annotations.

Contribution

It proposes the Intersection Pathway (IP), a cross-modal learning method that aligns WiFi fingerprint and displacement traces in a shared latent space with additive structure for direct relative localization.

Findings

Achieves accurate relative localization across various displacement ranges.

Learns displacement-aware WiFi representations effectively.

Can be extended to few-shot absolute localization with sparse anchors.

Abstract

WiFi fingerprint-based indoor localization has been widely studied, but most existing approaches focus on absolute positioning and rely on dense coordinate annotations, which are costly to obtain at scale. In this paper, we study a fundamentally different problem: relative localization, where the goal is to directly estimate the displacement between two WiFi fingerprint traces without predicting their absolute positions. To reduce annotation overhead, we adopt weak supervision in the form of stepwise motion vectors obtained from inertial sensing. We propose Intersection Pathway (IP), a cross-modal learning framework that aligns fingerprint traces (f-traces) and displacement traces (d-traces) in a shared latent space. The key idea is to enforce an additive structure in the latent space, such that latent addition and subtraction correspond to physical motion composition, enabling direct relative-displacement inference. Experiments on a synthesized dataset derived from real measurements demonstrate that the proposed method learns displacement-aware WiFi representations and achieves accurate relative localization across varying displacement ranges. Furthermore, the learned model can be extended to few-shot absolute localization with sparse anchors.