A Probability-Based Anti-Collision Protocol for RFID Tag Identification

Abstract

Anti-collision protocols have long been an important research field in RFID systems, and the collision tree (CT) protocol is very typical. Some researchers have improved CT to achieve better performance. However, these schemes still encounter the problem of too many collisions occurring at the beginning of the tags identification. In order to reduce the initial redundant collision, we propose a Probability-based Query Tree protocol (PQT). PQT divides all tags into several small subsets before the tags to be identified, and the protocol is composed of three parts: Inverse Probability Function (IPF), Total Time Slot Function (TTSF), and Mapping Table. The inverse probability function quantifies the possibility that a subset contains tags. The total time slot function measures the number of total time slots, and is used to obtain an optimal position to divide tags in the beginning. The mapping table is presented as an adaptive method in practical applications. In performance analysis, the identification efficiency of PQT is close to the optimal value. Simulation results further show that PQT outperforms the other existing anti-collision protocols.