High-Accuracy and Efficient DV-Hop Localization for IoT Using Hop Loss
Abstract
Accurate localization is critical for Internet of Things (IoT) applications. Using hop loss in DV-Hop-based algorithms is a promising approach. Nevertheless, challenges lie in overcoming the computational complexity caused by re-calculating the predicted hop-counts, and how to further optimize the modeling for better accuracy. In this paper, a novel hop loss modeling, distance-based connectivity consistency (DCC), is proposed. By focusing on the first order connectivity, DCC avoids computing predicted hop-counts, and significantly reduces the time complexity. We also provide a proof to theoretically guarantee that this design achieves a full coverage of all hop errors. In addition, by computing a continuous loss function instead of the discrete hop-count errors, DCC further improves the localization accuracy. In the evaluations, DCC demonstrates notable improvements in accuracy over other highly regarded algorithms, and reduces 30% to 40% total computation time compared with the baseline algorithm using hop loss.
Cite
@article{arxiv.2412.19827,
title = {High-Accuracy and Efficient DV-Hop Localization for IoT Using Hop Loss},
author = {Zhengdi Shen and Qiran Wang},
journal= {arXiv preprint arXiv:2412.19827},
year = {2024}
}
Comments
This manuscript has been submitted to Cluster Computing and is under peer review