Uncertainty-Based Fingerprinting Model Monitoring for Radio Localization

Abstract

Indoor radio environments often consist of areas with mixed propagation conditions. In line-of-sight (LoS)-dominated areas, classic time-of-flight (ToF) methods reliably return accurate positions, while in nonline-of-sight (NLoS) dominated areas (AI-based) fingerprinting methods are required. However, fingerprinting methods are only cost-efficient if they are used exclusively in NLoS-dominated areas due to their expensive life cycle management. Systems that are both accurate and cost-efficient in LoS- and NLoS-dominated areas require identification of those areas to select the optimal localization method. To enable a reliable and robust life cycle management of fingerprinting, we must identify altered fingerprints to trigger update processes. In this article, we propose methods for uncertainty estimation of AI-based fingerprinting to determine its spatial boundaries and validity. Our experiments show that we can successfully identify spatial boundaries of the fingerprinting models and detect corrupted areas. In contrast to the state-of-the-art, our approach employs an intrinsic identification through out-of-distribution (OOD) detection, rendering external detection approaches unnecessary.