摘要

Dynamic frame slotted Aloha (DFSA) has been widely adopted to solve the anticollision problem in a radio frequency identification (RFID) system. In a DFSA procedure, the interrogator needs to continuously estimate tag backlog and select a new frame length for identifying the backlog. Intuitively, the accuracy of the tag estimator will affect the read performance. Hence, a considerable amount of research effort has been invested to improve the accuracy of backlog estimation. The improvement in general comes at the expense of large computation load and may lead to a serious challenge if one needs to implement such a kind of estimators in a real RFID system. This paper analyzes the influence of estimation error on read performance. Based on the analysis, we propose a feasible and easy-to-implement anticollision algorithm. Our proposed algorithm can achieve a normalized throughput of 35% that is very close to the theoretical maximum 36.1% for an EPCglobal UHF Class-1 Generation-2 system. The easy-to-implement advantage of our algorithm comes at the expense of only 1% reduction in normalized throughput as compared with the case where maximum throughput can be obtained. The results obtained are useful in designing fast and efficient interrogators.

  • 出版日期2014-4