LOCUS: LOcalization with Channel Uncertainty and Sporadic Energy

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Biswas, Subrata, Khan, Mohammad Nur Hossain, Colwell, Violet, Adiletta, Jack, Islam, Bashima
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912489572139008
author Biswas, Subrata
Khan, Mohammad Nur Hossain
Colwell, Violet
Adiletta, Jack
Islam, Bashima
author_facet Biswas, Subrata
Khan, Mohammad Nur Hossain
Colwell, Violet
Adiletta, Jack
Islam, Bashima
contents Accurate sound source localization (SSL), such as direction-of-arrival (DoA) estimation, relies on consistent multichannel data. However, batteryless systems often suffer from missing data due to the stochastic nature of energy harvesting, degrading localization performance. We propose LOCUS, a deep learning framework that recovers corrupted features in such settings. LOCUS integrates three modules: (1) Information-Weighted Focus (InFo) to identify corrupted regions, (2) Latent Feature Synthesizer (LaFS) to reconstruct missing features, and (3) Guided Replacement (GRep) to restore data without altering valid inputs. LOCUS significantly improves DoA accuracy under missing-channel conditions, achieving up to 36.91% error reduction on DCASE and LargeSet, and 25.87-59.46% gains in real-world deployments. We release a 50-hour multichannel dataset to support future research on localization under energy constraints. Our code and data are available at: https://bashlab.github.io/locus_project/
format Preprint
id arxiv_https___arxiv_org_abs_2302_09409
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle LOCUS: LOcalization with Channel Uncertainty and Sporadic Energy
Biswas, Subrata
Khan, Mohammad Nur Hossain
Colwell, Violet
Adiletta, Jack
Islam, Bashima
Machine Learning
Accurate sound source localization (SSL), such as direction-of-arrival (DoA) estimation, relies on consistent multichannel data. However, batteryless systems often suffer from missing data due to the stochastic nature of energy harvesting, degrading localization performance. We propose LOCUS, a deep learning framework that recovers corrupted features in such settings. LOCUS integrates three modules: (1) Information-Weighted Focus (InFo) to identify corrupted regions, (2) Latent Feature Synthesizer (LaFS) to reconstruct missing features, and (3) Guided Replacement (GRep) to restore data without altering valid inputs. LOCUS significantly improves DoA accuracy under missing-channel conditions, achieving up to 36.91% error reduction on DCASE and LargeSet, and 25.87-59.46% gains in real-world deployments. We release a 50-hour multichannel dataset to support future research on localization under energy constraints. Our code and data are available at: https://bashlab.github.io/locus_project/
title LOCUS: LOcalization with Channel Uncertainty and Sporadic Energy
topic Machine Learning
url https://arxiv.org/abs/2302.09409