Adaptive Learning Framework for AoA-Based Outdoor Localization in 5G/6G
A new adaptive framework for angle-of-arrival (AoA) based outdoor localization is proposed, designed to address varying training dataset sizes in 5G and 6G networks. The framework offers two alternative learning strategies: one optimized for large datasets and another for small datasets. It is evaluated on a real massive MIMO OFDM outdoor channel state information dataset. Localization is critical for intelligent transportation, smart factories, and smart cities. Deep learning has improved accuracy, but training processes vary by deployment and data collection effort. Recent work has shown AoA-based localization to be robust. The framework adapts to these conditions, providing a comprehensive solution for outdoor localization.
Key facts
- Proposed adaptive framework for AoA-based outdoor localization in 5G/6G networks.
- Two alternative learning strategies: one for large datasets, one for small datasets.
- Evaluated on real massive MIMO OFDM outdoor CSI dataset.
- Localization essential for intelligent transportation, smart factories, smart cities.
- Deep learning improves accuracy but training varies by deployment.
- AoA-based localization shown to be robust in recent works.
- Framework adapts to training dataset size.
- Addressed in arXiv:2605.05055.
Entities
—