||The growing importance of pervasive operations  such as location sensing   , object tracking , activity monitoring  and cardinality computing   has led to increasing interests in Radio Frequency Identification (RFID) systems. Counting the population of a tag set, one of the most crucial operations of dense RFID networks, seems to be a dilemma. Using tag identification protocols to recognize all tags is a straightforward approach, but the long latency is intolerable to most applications. Cardinality estimation protocols, designed to improve the efficiency and effectiveness, are not able to obtain a precise result. Ideally, we would like to have a protocol which can achieve the accuracy of former and the efficiency of latter. The two main contributions of my work are 1. the LOttery Frame protocol (LoF), a scheme that can quickly estimate the cardinality of a large tag set; and 2. the Precise and Anonymous Counting protocol (PAC), a scheme that remarkably reduces the processing time and energy cost in comparison to identification protocols, while still guaranteeing the result reliable and precise. The theoretical analysis shows that both the average latency and the energy cost can be significantly reduced by using my counting protocols. The simulation-based performance evaluation also shows LoF and PAC outperform existing possible schemes in tag estimation and precise counting respectively.