Analysis of Tree Algorithms for RFID Arbitration

Radio Frequency Identiication (RFID) systems have emerged as a viable and aaordable alternative for tagging and/or labeling small to large quantities of items. The arbitration problem is that of identifying alìtags' present in the current environment (i.e reading the labels). This is a multiaccess communication problem in which contention for the communication medium is transient in nature. Important measures of performance include the time required to identify the tags, and the power consumed by the tags. The rst of these is proportional to the number of time slots needed to complete the arbitration process, and the second is proportional to the total number of times the tags are required to transmit during this process. We analyze a family of Tree Search algorithms with respect to these two measures. Previous analysis on the average number of time slots 1, 4, 5] reveals a linear dependence on the number of tags, m. We extend this work by developing expressions for the decomposition of these slots into three basic types: zero, one and multiple reply slots. Original expressions are also developed for the total number of tag replies. We show that on average this number is (m log m), and that this number can be reduced to (m) when prior knowledge of m is available.