Abstract: For the radio frequency identification (RFID) systems, the tag collision brings about low efficiency in the process of tag identification. On the basis of the traditional adaptive multi-tree, it raises a new improved adaptive multi-tree (NAMS) anti-collision algorithm. Before calculating the collision factor and selecting fork number, in the algorithm it first estimated collided tag number N and the statistical collision digit capacity m, directly identified leaf nodes which satisfied the N=2^m to shorten the search time; simultaneously it introduced autosleep counting mechanism which decreased on average half of the number of executive commands, thereby shortening the communication time. Finally, the required total number of slots of the algorithm was analyzed and experiments were conducted. The results show that the improved multi-tree anti-collision algorithm has higher recognition speed and system throughput rate than the traditional anti-collision algorithm.

Key words: radio frequency identification; multi-tree; anti-collision algorithm; collision factor