Abstract. In order to calculate the number of tags in a radio frequency identification (RFID) system, several tag estimation methods have been investigated in literature and most of the available estimation methods need the overall knowledge of idle, success and collision slots of the previous frame to carry out the tag estimation process. In this article, we present three techniques to reduce the complexity of Vahedi's tag estimation for tag King Mongkut's University of Technology North Bangkokcollision resolution in RFID systems using dynamic frame slotted ALOHA. Our modified and useful approach considers the information about only the number of empty, successful or colliding slots in the previous frame for the tag estimation. Three decision rules were obtained by maximizing the likelihood of success, idle and collision which helps in the reduction of complexity substantially. However, the accuracy of estimation decreases for success-only and idle-only methods while the collision-only method gives a consistent and lower estimate error when the frame sizes and the number of tags increase.