Collection of a Continuous Long-Term Dataset for the Evaluation of Wi-Fi-Fingerprinting-Based Indoor Positioning Systems

Sensors (Basel). 2022 Nov 8;22(22):8585. doi: 10.3390/s22228585.

Abstract

Indoor positioning and navigation have been attracting interest from the research community for quite some time. Nowadays, new fields, such as the Internet of Things, Industry 4.0, and augmented reality, are increasing the demand for indoor positioning solutions capable of delivering specific positioning performances not only in simulation but also in the real world; hence, validation in real-world environments is essential. However, collecting real-world data is a time-consuming and costly endeavor, and many research teams lack the resources to perform experiments across different environments, which are required for high-quality validation. Publicly available datasets are a solution that provides the necessary resources to perform this type of validation and to promote research work reproducibility. Unfortunately, for different reasons, and despite some initiatives promoting data sharing, the number and diversity of datasets available are still very limited. In this paper, we introduce and describe a new public dataset which has the unique characteristic of being collected over a long period (2+ years), and it can be used for different Wi-Fi-based positioning studies. In addition, we also describe the solution (Wireless Sensor Network (WSN) + mobile unit) developed to collect this dataset, allowing researchers to replicate the method and collect similar datasets in other spaces.

Keywords: Wi-Fi dataset; Wi-Fi fingerprinting; Wireless Sensor Network; dataset; indoor positioning; long-term dataset; radio maps; validation.

MeSH terms

  • Computer Simulation
  • Reproducibility of Results*