Author(s): Moreira, Adriano ; Meneses, Filipe
Date: 2015
Persistent ID: http://hdl.handle.net/1822/38352
Origin: RepositóriUM - Universidade do Minho
Project/scholarship: info:eu-repo/grantAgreement/FCT/5876/147280/PT;
Subject(s): indoor positioning; Wi-Fi fingerprinting; organic; crowdsourcing; quality; Science & Technology; Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
In the past decade, the research community has been dedicating considerable effort into indoor positioning systems based on Wi-Fi fingerprinting techniques, mainly due to their capability to exploit existing infrastructures. Crowdsourcing approaches, also known as organic, have been proposed recently to address the problem of creating and maintaining the corresponding radio maps. In these organic systems, the users of the system build the radio map themselves while using it to estimate their own position/location. However, most of these collaborative methods, proposed by several authors, assume that all the users are honest and committed to contribute to a good quality radio map. In this paper we assess the quality of a radio map built collaboratively and propose a method to classify the credibility of individual contributions and the reputation of individual users. Experimental results are presented for an organic indoor location system that has been used by more than one hundred users over a period of around 12 months.
