Optimal Frame Tuning for Aloha Protocols in RFID Networks

We consider the problem of efficient tag identification in RFID networks. Among the solutions recently proposed for tag identification, a variant of the framed slotted ALOHA protocol, called tree slotted ALOHA (TSA), has been shown to achieve the lowest identification delay and highest time system efficiency. However, in case of a large number of tags, this protocol produces an estimate of the tag population size which still differs from the actual number of tags, and utilizes a non-optimal method for deciding the frame size. In this paper we address the issues of more precise tag number estimation and achieving optimal frame sizing by proposing a new protocol, the binary spitting TSA (BSTSA) protocol. Through thorough NS2-based simulations we show that BSTSA considerably outperforms all previous protocols, achieving 80% time system efficiency vs. the 50% achieved by TSA, and reduces the length of the tag identification process of up to 40% with respect to TSA.