Alessandro Copett,  J.C.B. Leite,  Orlando Loques.  VII Workshop de Testes e Tolerância a Falhas - Curitiba, PR, Brasil - 2006

The knowledge of a mobile device location is essential for the development of context-aware applications. This paper investigates solutions for the location problem, in indoor environments, that cater for low cost, precision, and fault tolerance requirements. In addition, the feasibility of execution in devices with restricted resources and the processing time required to execute the location function are also of concern. Specifically, we evaluate two solutions using Neural Networks and the k-nearest neighbor’s algorithm, which use radio signals generated by access points to mobile networks. A real physical scenario was used to implement and evaluate the techniques.

http://www.lbd.dcc.ufmg.br:8080/colecoes/wtf/2006/st4_2.pdf

Processando consulta.