Handling Device Heterogeneity for Deep Learning-based Localization

TL;DR

This paper proposes techniques to address device heterogeneity in deep learning-based cellular localization, significantly improving accuracy across various phones and enabling wider deployment.

Contribution

It introduces methods to map or transfer RSS measurements between different devices, enhancing localization accuracy in heterogeneous environments.

Findings

Improved localization accuracy by over 220% across four testbeds.

Techniques effectively handle device heterogeneity in deep learning localization.

Demonstrated robustness across multiple Android devices.

Abstract

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.