Adaptive Framework for Collisions in RFID Tag Identification

Radio Frequency Identification (RFID) is promising, as a technique, to enable tracking of essential information about objects as they pass through supply chains. Information thus tracked can be utilised to efficiently operate the supply chain. Effective management of the supply chain translates to huge competitive advantage for the firms involved. Among several issues that impede seamless integration of RFID tags in a supply chain, one of the problems encountered while reading RFID tags is that of collision, which occurs when multiple tags transmit data to the same receiver slot. Data loss due to collision necessitates re-transmission of lost data. We consider this problem when Framed Slotted ALOHA protocol is used. Using machine learning, we adaptively configure the number of slots per frame to reduce the number of collisions while improving throughput.