Handling Device Heterogeneity for Deep Learning-based Localization
This addresses the challenge of deploying accurate localization systems across diverse mobile devices, which is incremental as it builds on existing deep learning fingerprinting methods.
The paper tackles the problem of device heterogeneity degrading accuracy in deep learning-based cellular localization by introducing techniques to map or transfer signal measurements between phones, resulting in over 220% improvement in localization accuracy across four testbeds compared to state-of-the-art systems.
Deep learning-based fingerprinting is one of the current promising technologies for outdoor localization in cellular networks. However, deploying such localization systems for heterogeneous phones affects their accuracy as the cellular received signal strength (RSS) readings vary for different types of phones. In this paper, we introduce a number of techniques for addressing the phones heterogeneity problem in the deep-learning based localization systems. The basic idea is either to approximate a function that maps the cellular RSS measurements between different devices or to transfer the knowledge across them. Evaluation of the proposed techniques using different Android phones on four independent testbeds shows that our techniques can improve the localization accuracy by more than 220% for the four testbeds as compared to the state-of-the-art systems. This highlights the promise of the proposed device heterogeneity handling techniques for enabling a wide deployment of deep learning-based localization systems over different devices.