Research Article
An Empirical Study of Trilateration and Clustering for Indoor Localization and Trend Prediction
@INPROCEEDINGS{10.1007/978-3-030-76063-2_6, author={Aarth Tandel and Anvesh Chennupati and Behnam Dezfouli}, title={An Empirical Study of Trilateration and Clustering for Indoor Localization and Trend Prediction}, proceedings={Science and Technologies for Smart Cities. 6th EAI International Conference, SmartCity360°, Virtual Event, December 2-4, 2020, Proceedings}, proceedings_a={SMARTCITY}, year={2021}, month={5}, keywords={Tracking Clustering Wireless Trilateration Gaussian mixture BLE}, doi={10.1007/978-3-030-76063-2_6} }- Aarth Tandel
Anvesh Chennupati
Behnam Dezfouli
Year: 2021
An Empirical Study of Trilateration and Clustering for Indoor Localization and Trend Prediction
SMARTCITY
Springer
DOI: 10.1007/978-3-030-76063-2_6
Abstract
Localization via trilateration determines the location of moving objects using the distances between each object and multiple stations. Since low-power wireless technologies are the primary enablers of these localization methods, the technology’s type and characteristics highly affect trilateration accuracy. In addition, pre-processing the collected data can also be used as an effective method to enhance system accuracy. This paper presents an effective way of tracking objects using trilateration in indoor environments. We analyze the data generated from the stations, including coordinates, timestamps, and identifiers. After running a clustering algorithm on the data, we infer information on the object’s behavior, frequently visited places, and predict objects’ location. Field testing results at Santa Clara University demonstrate that accuracy is increased in the range of 20 to 40% when applying the pre-processing method.
