Bayesian sensor model for indoor localization in Ubiquitous Sensor Network

Ubiquitous sensor networks (USN) technology is one of the essential key for driving the next generation network (NGN) to realize secure and easy access from anyone, any thing, anywhere and anytime. The location information is one of the most important and frequently-used contexts in ubiquitous networking. However, a system can use the changes of location to adapt its behavior, such as computation and communication, without the user intervention. In this paper we introduce a Bayesian sensor framework for solving the location estimation errors problem in Radio Frequency Identification (RFID) environments. In our model the physical properties of the signal propagation are not taken into account directly. Instead, the location estimation is regarded as machine learning problem in which the task is to model how the location estimation error is distributed indoors based on a sample of measurements collected at several known locations and stored in RFID tags. Results obtained by simulations demonstrate the suitability of the proposed model to provide high performance level in terms of accuracy and scalability.