Neural network and fingerprinting-based localization in dynamic channels

In a harsh indoor environment, fingerprinting localization techniques perform better than the traditional ones, based on triangulation, because multipath is used as constructive information. However, this is generally true in static conditions as fingerprinting techniques suffer degradations in location accuracy in dynamic environments where the properties of the channel change in time. This is due to the fact that the technique needs a new database collection when a change of the channel´s state occurs. This paper proposes a method allowing an accurate mobile user´s location in time-varying channels when it is difficult or impossible to collect measurements. The system has the ability to generate, from a measured reference database, a new database corresponding to a new channel state. This is done by using measurements of few reference points in conjunction with a tree model data mining technique. The technique uses a regression analysis to learn the temporal predictive relationship between the received signal strength values of the mobile and the reference points in order to generate a new database at a different time state. After generating several databases, corresponding to several time states, an artificial neural network is used for location estimation. Results show low degradation, compared to a static channel, of approximately 7% and 11% at 3 meters in 2D and 3D dynamic environments, respectively.