Architecture
Schematic overview of LOCUS. The highlighted components are the contributions of this paper.
Accurate sound source localization (SSL) requires consistent multichannel data for reliable degree of arrival (DoA) estimation. However, intermittently powered batteryless systems often suffer from incomplete sensor data due to the stochastic nature of energy harvesting. Existing methods struggle with missing channels, leading to significant performance degradation. In this paper, we propose LOCUS, a novel deep learning-based system designed to recover corrupted features for SSL in batteryless systems. LOCUS addresses missing data by leveraging information entropy estimation and conditional interpolation, combining three modules: (1) Information-Weighted Focus (InFo), which identifies and quantifies corrupted data elements, (2) Latent Feature Synthesizer (LaFS), which synthesizes missing features, and (3) Guided Replacement (GRep), which intelligently replaces missing elements while preserving valid data. We demonstrate significant performance improvements using two datasets: DCASE and LargeSet, where LOCUS achieves up to 36.91% lower DoA error compared to existing methods. Real-world evaluations across three environments with intermittent power sources show a 25.87-59.46% improvement in performance when channels are stochastically missing. Additionally, we release a 50-hour multichannel dataset to support further research in SSL.
@article{biswas2023locus,
title={LOCUS: LOcalization with Channel Uncertainty and Sporadic Energy},
author={Biswas, Subrata and Khan, Mohammad Nur Hossain and Colwell, Alex and Adiletta, Jack and Islam, Bashima},
journal={arXiv preprint arXiv:2302.09409},
year={2023}
}