The demand of localization services continuously rises with the advancing and developing of mobile and Internet of Things (IoT) devices in the last decade. Localization based on various wireless standards have been extensively studied to provide services in different applications. With the emergence of deep learning (DL) algorithms, many different DL models or architectures are utilized in localization tasks to increase the accuracy and robustness over conventional localization methods. However, there is a lack of survey paper that focuses on how different DL-related methodologies are adopted in corresponding localization scenarios. In this paper, we aim at presenting a comprehensive and up-to-date review of DL-based localization methods from the respects of those common wireless standards such as Wi-Fi, cellular, ultra-wideband (UWB) and Bluetooth low energy (BLE). We further highlight those factors that build up end-to-end DL-based methods including the datasets, preprocessing methods, and DL algorithms. We also discuss how DL can potentially address the challenges in terms of localization precision, latency, and data generation.
Andrews et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: