Deconvolution-based indoor localization with WLAN signals and unknown access point locations

Torres-Sospedra

Leppäkoski

et al. 2017

Data

143

104

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.

Section: Benchmark Indoor Positioning Resultsmentioning

confidence: 99%

“…Furthermore, the MATLAB/Octave files for creating the more advanced algorithms, such as path loss-based estimators [23], Dempster-Shaffer-based estimators [24] or image-based processing [25,26], among others, are not made available due again to IP issues (e.g., calling some functions, which are proprietary code).…”

Section: Restrictions Of the Availabilitymentioning

confidence: 99%

Wi-Fi Crowdsourced Fingerprinting Dataset for Indoor Positioning

Torres-Sospedra

Leppäkoski

et al. 2017

Data

143

104

“…In order to estimate the unknown path-loss parameters 8 ap , the Least Squares (LS) solution can be employed [12]. We will obtain one parameter vector per transmitter (i.e., WLAN AP or RFID reader):…”

Section: Signal Strength Analysis and Path-loss Modelsmentioning

confidence: 99%

“…Global low-cost high accuracy indoor positioning is still hard to achieve and there is a significant research effort worldwide towards finding accurate indoor localization methods based on existing wireless signals. The most widespread low-cost indoor localization solutions nowadays are those based on Wireless Local Area Networks (WLAN) signals and using their Received Signal Strengths (RSS) [6] [7][8] [9][ 10] [ 12] [ 13]. Hybrid solutions with WLAN and sensors have also been studied [11].…”

Section: Imentioning

confidence: 99%

WLAN and RFID Propagation channels for hybrid indoor positioning

International Conference on Localization and GNSS 2014 (ICL-GNSS 2014)

Koski

Talvitie

et al. 2014

Self Cite

Indoor localization based on Received Signal Strengths (RSS) or on some form of power measurements is a low-cost and low-complexity solution gaining more and more interest in the research and commercial worlds. Typically, Wireless Local Area Network (WLAN) signals are employed for such purpose, due to the fact that they are widely spread in indoor environments. Nevertheless, any wireless signal available in indoor scenarios can be used for positioning based on similar power measurement approaches. One example is the radio frequency identification (RFID), which enable portable localization systems in the form of wearable RFID tags. Such RFID-enabled systems are highly demanded for health-state monitoring, object tracking, and security. These applications are mostly indoor applications, and thus, we can envision a near future where multiple RFID and WLAN signals will co-exist on multiple frequency bands. The existing signal diversity can offer a benefit in indoor positioning, providing that the signal propagation effects for both WLAN and RFID are well understood and taken into account. However, measurement-based studies on indoor channel modeling, including path loss and shadowing effects of RFID signals are still missing. A comparison between RFID and WLAN channel models for positioning purpose has yet to be made. It is the purpose of our paper to address the path-loss channel models and shadowing effects for WLAN and RFID signals based on extensive measurement campaigns in an office environment.

“…In a positioning context, these networks have been used mostly under fingerprinting solutions, offering a relatively good performance, 5 to 10 meters, in densely covered areas [14,15].…”

Section:  Digital Video Broadcasting -Terrestrial (Dvb-t)mentioning

confidence: 99%

Overview of positioning technologies from fitness-to-purpose point of view

Basiri

International Conference on Localization and GNSS 2014 (ICL-GNSS 2014)

Silva

et al. 2014

Self Cite

Abstract-Even though Location Based Services (LBSs) are being more and more widely-used and this shows a promising future, there are still many challenges to deal with, such as privacy, reliability, accuracy, cost of service, power consumption and availability. There is still no single low-cost positioning technology which provides position of its users seamlessly indoors and outdoors with an acceptable level of accuracy and low power consumption. For this reason, fitness of positioning service to the purpose of LBS application is an important parameter to be considered when choosing the most suitable positioning technology for an LBS. This should be done for any LBS application, since each application may need different requirements. Some location-based applications, such as locationbased advertisements or Location-Based Social Networking (LBSN), do not need very accurate positioning input data, while for some others, e.g. navigation and tracking services, highlyaccurate positioning is essential. This paper evaluates different positioning technologies from fitness-to-purpose point of view for two different applications, public transport information and family/friend tracking.