Integrating Geometry-Driven and Data-Driven Positioning via Combinatorial Data Augmentation
share
Precise positioning has become one core topic in wireless communications by facilitating candidate techniques of B5G. Nevertheless, most existing positioning algorithms, categorized into geometric-driven and data-driven approaches, fail to simultaneously fulfill diversified requirements for practical use, e.g., accuracy, real-time operation, scalability, maintenance, etc. This article aims at introducing a new principle, called combinatorial data augmentation (CDA), a catalyst for the two approaches' tight integration. We first explain the concept of CDA and its critical advantages over the two standalone approaches. Then, we confirm the CDA's effectiveness from field experiments based on WiFi round-trip time and inertial measurement units. Lastly, we present its potential beyond positioning, expected to play a critical role in B5G.
READ FULL TEXT