GRAFICS: Graph Embedding-based Floor Identification Using Crowdsourced RF Signals

TL;DR

GRAFICS is a novel graph embedding-based system for floor identification using crowdsourced RF signals, effectively handling heterogeneous data and limited labels to achieve high accuracy.

Contribution

It introduces a new bipartite graph model and a graph embedding algorithm called E-LINE for accurate floor identification with minimal labeled data.

Findings

Achieves 96% accuracy with few labeled samples

Outperforms state-of-the-art algorithms by over 45% in micro-F score

Validated on large-scale datasets from multiple cities

Abstract

We study the problem of floor identification for radiofrequency (RF) signal samples obtained in a crowdsourced manner, where the signal samples are highly heterogeneous and most samples lack their floor labels. We propose GRAFICS, a graph embedding-based floor identification system. GRAFICS first builds a highly versatile bipartite graph model, having APs on one side and signal samples on the other. GRAFICS then learns the low-dimensional embeddings of signal samples via a novel graph embedding algorithm named E-LINE. GRAFICS finally clusters the node embeddings along with the embeddings of a few labeled samples through a proximity-based hierarchical clustering, which eases the floor identification of every new sample. We validate the effectiveness of GRAFICS based on two large-scale datasets that contain RF signal records from 204 buildings in Hangzhou, China, and five buildings in Hong Kong. Our experiment results show that GRAFICS achieves highly accurate prediction performance with only a few labeled samples (96% in both micro- and macro-F scores) and significantly outperforms several state-of-the-art algorithms (by about 45% improvement in micro-F score and 53% in macro-F score).