Emerging computational RFID (Radio Frequency IDentification) tags combine sensing and computation with traditional RFID technology. The combination allows small, long-lived sensors that are as unobtrusive as a sticker to be placed on everyday objects and read when they are in the proximity of an RFID reader. The technology is an enabler for ubiquitous computing, cyber-physical systems, and sensor networks. This project is developing computational RFIDs as a networked system, with an emphasis on operating system software and hardware support on the computational RFID tags and network protocols that will support new applications. One key research challenge is that these devices are extremely energy limited and dependent on the varying energy they can harvest; the tags must schedule tasks well to use energy effectively. A second research challenge is to design network protocols that are well suited to computational RFID, since traditional RFID protocols are intended for repetitive inventorying tasks and sensor network protocols assume that nodes have symmetric abilities. The approach taken is experimental. The project is building software support and hardware extensions on an existing computational RFID tag, and prototyping candidate applications. The impacts of this research will be to extend the coverage with which computing systems can instrument the physical world, which in turn enables new applications, such as automated activity detection for eldercare, that are beneficial to society. The software and hardware developed as part of this project will be made available to other researchers and for educational use.
