An indoor item finder with active RFID tags

Abstract

Radio Frequency identification (RFID) is a wireless identification technology, used in a variety of applications. It has the potential for providing indoor tracking/location services, since both the identity and location information can be provided. However, radio signals in indoor environments are generally harshly impaired and tags have very limited capabilities which pose many challenges for positioning them. The existing RFID systems provide only coarse-grained location information and depend on an array of RFID readers placed in strategic positions. This paper aims at implementing a RFID-based indoor locating system. It is called indoor item finder where a novel model location fingerprinting will be adopted. This novel model is the opposite of the traditional indoor positioning system by using active reference tags to replace expensive RFID readers. Based on this model, three positioning algorithms are reviewed and simulated: the LANDMARC (Indoor Location Sensing Using Active RFID), together with two improved approaches: LANDMARC with virtual reference tags (VIRE) and LANDMARC with position convergence. Also, in this paper, novel interpolation schemes for VIRE: dynamic linear interpolation and Lagrange interpolation are presented. According to the simulation results, the proposed dynamic linear interpolation ensures much better performance. Also, a series of simulation is conducted to find and analyze the effect of different parameters: number of readers, placement of readers, density of reference tags and number of nearest neighbours, thus find out the best setting for the indoor item finder.