Design and Implementation of Indoor Positioning System Technology Lecture Attendance Application using Bluetooth Low Energy and K-NN Classification Model

Recording attendance of lecturers and students of a course is the responsibility of many universities. A study has been conducted in the Department of Informatics, Syiah Kuala University (USK) in a building belonged to the Faculty of Mathematics and Natural Sciences (FMIPA). An Android-based application was built to record the attendance of lecturers and students within this department. An Indoor Positioning System is required in order to locate users' location accurately when they are inside the building. Therefore, this study offers a solution that applies the Fingerprinting method by utilizing the Bluetooth Low Energy (BLE) signal strength to overcome this problem. In this study, the data collection was carried out to collect the signal strength by mapping reference points sequentially with distance of 2 meters. The application was tested for its accuracy of the reference point using the K-NN classification method. Based on the results of testing the accuracy of the reference point sequentially, it has the best accuracy of 78.60% compared to the reference point at random. It is found that the use of 6 Beacons has better accuracy than the use of 3 Beacons. All functionalities were tested using the Blackbox test, and they were all successful. The results for the usability test using SUS achieved a score of 78.5% for the lecturer's side of the application and 86.1% for the student's side of the application. Both results suggest that the application can be accepted by the users.