Symmetric and Asymmetic Development of Bluetooth-Based Indoor Localization Mechanisms

Citing these works

Citing Asymmetric with optimization: If you used assymetric code in your work, please cite the IEEE Access

bib @ARTICLE{8642816, author={Lovn-Melgarejo, Jess and Castillo-Cara, Manuel and Huarcaya-Canal, Oscar and Orozco-Barbosa, Luis and Garca-Varea, Ismael}, journal={IEEE Access}, title={Comparative Study of Supervised Learning and Metaheuristic Algorithms for the Development of Bluetooth-Based Indoor Localization Mechanisms}, year={2019}, volume={7}, number={}, pages={26123-26135}, doi={10.1109/ACCESS.2019.2899736} }

Abstract

The present papers constitute a seminal contribution to indoor localization utilizing Bluetooth signals and supervised learning algorithms. They elucidate pivotal scientific advancements that solidify research endeavors in this domain. The manuscripts scrutinize several pivotal facets:

• Bluetooth signal characterization: The investigation delves into the behavioral patterns of Bluetooth signals employing 12 distinct supervised learning algorithms. This characterization serves as a cornerstone for comprehending the intricacies of signal dynamics, laying the groundwork for fingerprinting-based localization mechanisms.
• Optimization of transmit power configuration: The utilization of metaheuristic algorithms for optimizing the configuration of transmit powers is extensively discussed. This endeavor is paramount for enhancing the efficacy of indoor localization systems by mitigating the computational complexities associated with brute-force search methodologies.
• Assessment of symmetrical and asymmetrical transmission power configurations: The paper thoroughly evaluates transmission power configurations aimed at mitigating the impacts of multipath fading. The analysis juxtaposes the performance of diverse classification models under varied configurations, elucidating the settings that yield optimal outcomes. These contributions epitomize significant strides towards the continual enhancement of indoor localization systems, bearing profound implications for their operational efficiency and reliability.

Getting Started

The project has the following folders - Data: Includes data obtained by the mobile phone and also by the Raspberry Pi. - Asymmetric Code: Includes Python code of the results with different transmission powers using evolutionary and genetic algorithms. - Asymmetric Code: It includes results obtained with classical machine learning algorithms with asymmetric transmission power. It is used for smartphone data and also for Raspberri Pi data.git

License

This code is available under the Apache License 2.0.

Authors

• Manuel Castillo-Cara

Institutions