Deep Attention-based Sequential Ensemble Learning for BLE-Based Indoor Localization in Care Facilities

TL;DR

This paper presents DASEL, a deep attention-based sequential ensemble learning framework that significantly improves BLE-based indoor localization accuracy in care facilities by modeling human movement as a sequential process.

Contribution

Introduces DASEL, a novel deep learning framework that captures sequential movement data for improved BLE-based indoor localization in care settings.

Findings

DASEL achieves a macro F1 score of 0.4438, a 53.1% improvement over traditional methods.

The framework effectively models human movement trajectories using attention mechanisms.

Real-world data evaluation demonstrates significant performance gains.

Abstract

Indoor localization systems in care facilities enable optimization of staff allocation, workload management, and quality of care delivery. Traditional machine learning approaches to Bluetooth Low Energy (BLE)-based localization treat each temporal measurement as an independent observation, fundamentally limiting their performance. To address this limitation, this paper introduces Deep Attention-based Sequential Ensemble Learning (DASEL), a novel framework that reconceptualizes indoor localization as a sequential learning problem. The framework integrates frequency-based feature engineering, bidirectional GRU networks with attention mechanisms, multi-directional sliding windows, and confidence-weighted temporal smoothing to capture human movement trajectories. Evaluated on real-world data from a care facility using 4-fold temporal cross-validation, DASEL achieves a macro F1 score of 0.4438, representing a 53.1% improvement over the best traditional baseline (0.2898).