Smart Attendance System Based on Frequency Distribution Algorithm with Passive RFID Tags

Staff attendance information has always been an important part of corporate management. However, some opportunistic employees may consign others to punch their time cards, which hampers the authenticity of attendance and effectiveness of record keeping. Hence, it is necessary to develop an innovative anti-cheating system for office attendance. Radio-Frequency IDentification (RFID) offers new solutions to solve such problems because of its strong anti-interference capability and non-intrusiveness. In this paper, we present a smart attendance system that extracts distinguishable phase characteristics of individuals to enable recognition of various targets. A frequency distribution histogram is extracted as a fingerprint for recognition and the K-means clustering method is utilized for more fine-grained recognition of targets with similar features. Compared with traditional attendance mechanisms, RFID-based attendance systems are based on living biological characteristics, which greatly reduces the possibility of false records. To evaluate the performance of our system, we conducted extensive experiments. The results of which demonstrate the efficiency and accuracy of our system with an average accuracy of $\mathrm{92}\mathrm{\%}$. Moreover, the system evaluation shows that our design is robust against differences in the clothing worn and time of day, which further verifies the successful performance of our system.