Clustering benefits in mobile-centric WiFi positioning in multi-floor buildings

Cramariuc, Andrei; Huttunen, Heikki; Lohan, Elena Simona

doi:10.1109/icl-gnss.2016.7533846

Cited by 52 publications

(65 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Clustering as described in [20]: This method is evaluated in two versions: on the one hand, the RSS clustering with affinity propagation and a modified log-Gaussian metric to match the RSS, and on the other hand, 3D coordinate clustering with the k-means method and the modified log-Gaussian metric. The final position, in both algorithms, is estimated by averaging over the three training positions that correspond to the three best matches.…”

Section: Benchmark Indoor Positioning Resultsmentioning

confidence: 99%

“…The log-Gaussian probabilistic approach described for example in [20,23]: This algorithm assumes normally distributed noise and evaluates the likelihood of the RSS measurements at the training positions and determines the position estimate from the highest likelihood value(s). Clustering as described in [20]: This method is evaluated in two versions: on the one hand, the RSS clustering with affinity propagation and a modified log-Gaussian metric to match the RSS, and on the other hand, 3D coordinate clustering with the k-means method and the modified log-Gaussian metric.…”

Section: Benchmark Indoor Positioning Resultsmentioning

confidence: 99%

“…The initial position estimate is based on a Gaussian likelihood location fingerprinting, described for example in [20]. Once an initial estimate is shown on the screen (as the three nearest neighbors plus a coverage area estimate shown via a blue circle), the user is asked for feedback as seen in Figure 1b.…”

Section: Crowdsourced Data Collection Proceduresmentioning

confidence: 99%

“…• reading of the 'csv' training-or test data, • different visualizations of the data, such as the location of the measurements, or examples of power maps in 2D and 3D views and • position estimation based on two clustering and positioning algorithms described in [20].…”

Section: Supporting Softwarementioning

confidence: 99%

“…• shows examples of power maps in 2D and 3D views • shows examples of position estimation via two basic algorithms: weighted centroid estimator [18,19] and the log-Gaussian likelihood estimator [19,20].…”

Section: Supporting Softwarementioning

confidence: 99%

See 4 more Smart Citations

Wi-Fi Crowdsourced Fingerprinting Dataset for Indoor Positioning

Lohan

Torres-Sospedra

Leppäkoski

et al. 2017

Data

122

View full text Add to dashboard Cite

Benchmark open-source Wi-Fi fingerprinting datasets for indoor positioning studies are still hard to find in the current literature and existing public repositories. This is unlike other research fields, such as the image processing field, where benchmark test images such as the Lenna image or Face Recognition Technology (FERET) databases exist, or the machine learning field, where huge datasets are available for example at the University of California Irvine (UCI) Machine Learning Repository. It is the purpose of this paper to present a new openly available Wi-Fi fingerprint dataset, comprised of 4648 fingerprints collected with 21 devices in a university building in Tampere, Finland, and to present some benchmark indoor positioning results using these data. The datasets and the benchmarking software are distributed under the open-source MIT license and can be found on the EU Zenodo repository.

show abstract

Section: Benchmark Indoor Positioning Resultsmentioning

confidence: 99%