LocaToR: Locating Passive RFID Tags with the Relative Neighborhood Graph

Passive tags are widely used in many applications, for example, the supply chain, the food industry and ware-house management. In such applications, the location information of tags is very important. However, the uncertain proper-ties of Received Signal Strength Indicators (RSSI), various backscattering events on different power levels and the directivity of readers prevent traditional ranging-based approaches working well in passive RFID systems. In accordance with our observations during experiments, we create a novel approach to locate a specific tag among objects. Although absolute positions are difficult to obtain, we can estimate locations by building a relative relationship between tags. To reduce the effect of the above limitations, we propose a range-free approach named LocaToR to establish a relative neighborhood graph. We implement our method on a real passive system. Taking environmental factors into consideration, we look at two situations: a controlled chamber and a semi-open space. Experimental results show that our approach can obviously improve the accuracy of the localization system as well as save readers' energy.