Date of Award
Master of Science (MS)
Electrical and Computer Engineering (Holcomb Dept. of)
The growing interest in indoor localization has been driven by its wide range of applications in areas such as smart homes, industrial automation, and healthcare. With the increasing reliance on wireless devices for location-based services, accurate estimation of device positions within indoor environments has become crucial. Deep learning approaches have shown promise in leveraging wireless parameters like Channel State Information (CSI) and Received Signal Strength Indicator (RSSI) to achieve precise localization. However, despite their success in achieving high accuracy, these deep learning models suffer from limited generalizability, making them unsuitable for deployment in new or dynamic environments without retraining. To address the generalizability challenge faced by conventionally trained deep learning localization models, we propose the use of meta-learning-based approaches. By leveraging meta-learning, we aim to improve the models' ability to adapt to new environments without extensive retraining. Additionally, since meta-learning algorithms typically require diverse datasets from various scenarios, which can be difficult to collect specifically for localization tasks, we introduce a novel meta-learning algorithm called TB-MAML (Task Biased Model Agnostic Meta Learning). This algorithm is specifically designed to enhance generalization when dealing with limited datasets. Finally, we conduct an evaluation to compare the performance of TB-MAML-based localization with conventionally trained localization models and other meta-learning algorithms in the context of indoor localization.
Owfi, Ali, "Generalizable Deep-Learning-Based Wireless Indoor Localization" (2023). All Theses. 4135.