© 2019 IEEE. Radio Frequency Identification (RFID) is an important communication technology for the Internet of Things (IoT). With the development of IoT, RFID technology is widely deployed in large-scaled city networks. Under the circumstances, how to effectively optimize RFID network to decrease the operational cost of IoT is an ongoing research direction. In this area, one of the problems is the redundant reader issue, which means multiple RFID readers cover and interact with one RFID tag or many same tags, thereby resulting in massive energy cost. In order to solve this problem, quite a lot redundant reader elimination algorithms were presented to reduce unnecessary RFID readers. In this paper, the author proposes and compares three existing redundant reader elimination algorithms, including Redundant Reader Elimination Algorithm (RRE), Three-Count Based Algorithm (TCBA) and Threshold Selection Algorithm (TSA), to observe their performances. In the simulations, the author designs three experiments to test RFID reader's detection radius, detection accuracy and algorithmic efficiency by observing the number of readers eliminated in different environmental sets. The simulation results show that TCBA can detect more redundant readers than RRE and TSA, but it costs more time. Compared with TCBA and RRE, the performance of TSA is more practical and efficient to satisfy real RFID environment and market's needs.