An automatized contextual marketing system based on a wi-fi indoor positioning system

Journal ar
  • Volumen: 21
  • Número: 10
  • Fecha: 02 May 2021
  • ISSN: 14248220
  • Source Type: Journal
  • DOI: 10.3390/s21103495
  • Document Type: Article
  • Publisher: MDPI AG
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.A complete contextual marketing platform including an indoor positioning system (IPS) for smartphones is proposed and evaluated to later be deployed in large infrastructures, such as malls. To this end, we design and implement a novel methodology based on location-as-a-service (LAAS), comprising all the required phases of IPS generation: mall digital map creation, the tools/procedures for offline calibration fingerprint acquisition, the location algorithm, the smartphone app acquiring the fingerprint data, and a validation procedure. To select an appropriate fingerprint location algorithm, a comparison among K-nearest neighbors (KNN), support vector machine (SVM), and Freeloc is accomplished by employing a set of different smartphones in two malls and assessing different occupancy levels. We demonstrate that our solution can be quickly deployed at shop level accuracy in any new location, resulting in a robust and scalable proposal.

Author keywords

    Indexed keywords

      Funding details