Smart Attendance System: A Two-Factor BLE and Biometric Authentication Framework for Academic Integrity

Abstract

Attendance management in educational institutionsremainsachallengingtaskduetoissuessuch as proxy attendance, manual record maintenance, and insecure verification methods. Existing attendance systems based on QR codes, GPS tracking, RFID cards, or basic biometric authentication often fail to provide reliable protection against spoofing and unauthorized attendance marking. This paper presents a Smart AttendanceSystemthatcombinesBluetoothLowEnergy (BLE) proximity verification with AI-based facial authentication to improve security, accuracy, and automation in academic attendance management. The proposed system uses a two-factor verification mechanism. In the first stage, BLE Received Signal Strength Indicator (RSSI) analysis is used to verify whether the student is physically present within the classroom environment. In the second stage, facial recognitionandlivenessdetectiontechniquesareapplied using TensorFlow Lite and Google ML Kit to confirmtheidentityofthestudent.Theframeworkisimplementedas an Android application using Kotlin, Jetpack Compose,CameraX,FirebaseFirestore,andTensorFlow Lite. Most biometric operations are executed directly on the device to reduce latency and improve user privacy. The system is designed to minimize proxy attendance, reduce dependence on additional hardware, and provide secureattendancerecordsthroughcloudsynchronization. The proposed architecture offers a cost-effective and scalable solution suitable for modern educational institutions. This paper discusses the system design, implementation methodology, advantages, limitations, and possible future enhancements of the framework.

Introduction

This study proposes a Smart Attendance System (SAS) that combines Bluetooth Low Energy (BLE) proximity verification and AI-based facial recognition with liveness detection to create a secure, scalable, and cost-effective attendance management solution for educational institutions. Traditional attendance methods, such as manual roll calls and paper registers, are time-consuming and prone to proxy attendance and record manipulation. Existing digital systems based on QR codes, GPS, RFID, and basic facial recognition also suffer from security vulnerabilities, including screenshot sharing, location spoofing, card swapping, and biometric spoofing.

To overcome these limitations, the proposed system uses a Two-Factor Attendance Protocol (TFAP). The first factor verifies that a student is physically present in the classroom through BLE signal strength (RSSI) analysis. The second factor confirms identity using AI-powered facial recognition and liveness detection. This combination significantly reduces proxy attendance, location spoofing, and biometric fraud.

The system is implemented as an Android application using Kotlin, Jetpack Compose, CameraX, TensorFlow Lite, Google ML Kit, and Firebase Firestore. Most biometric processing occurs locally on the student’s smartphone, improving privacy and reducing cloud dependency. Attendance records are securely stored in Firebase Firestore.

The literature review highlights the shortcomings of existing attendance technologies. RFID systems require expensive infrastructure and can be bypassed through card sharing. QR-code systems are vulnerable to screenshot sharing, while GPS-based solutions suffer from poor indoor accuracy and location spoofing. Although facial recognition improves identity verification, many systems lack liveness detection and remain susceptible to spoofing attacks.

The proposed architecture consists of three major components:

1. Teacher Device – Acts as a BLE advertiser, broadcasting a unique classroom signal.
2. Student Device – Functions as a BLE scanner and biometric verification client.
3. Firebase Firestore Backend – Stores student data, attendance records, and session information.

The attendance workflow follows two steps:

• Proximity Verification: Students must be within the classroom and detect the teacher’s BLE signal above a predefined threshold.
• Biometric Verification: Students complete liveness checks (blinking, smiling, head movement) and facial recognition. Attendance is recorded only when both checks succeed.

The system aims to achieve a False Acceptance Rate below 0.1% and complete the entire attendance verification process in less than 15 seconds. Facial recognition uses a TensorFlow Lite FaceNet model that generates 128-dimensional facial embeddings, which are matched using cosine similarity.

Key advantages of the system include:

• Prevention of proxy attendance.
• Resistance to GPS spoofing and biometric spoofing attacks.
• Low-cost deployment without specialized hardware.
• Improved privacy through on-device biometric processing.
• Reduced administrative workload and faster attendance recording.

However, limitations include BLE signal fluctuations caused by environmental factors, dependency on smartphone hardware quality, and the need for internet connectivity to synchronize attendance data.

Future enhancements include cross-platform support using Flutter, offline attendance caching with later synchronization, adaptive BLE calibration for improved accuracy, and integration with university ERP systems.

Conclusion

This research presented a unified Smart Attendance System (SAS) engineered to resolve the persistent vulnerabilities of proxy reporting, location spoofing, and administrative inefficiency prevalent in modern educational institutions. By abandoning legacy singular-validation methods—such as vulnerable QR codes, inaccurate indoor GPS, and cost-prohibitive RFID infrastructure—this study successfully introduced a decentralized, Two-Factor Attendance Protocol(TFAP). The proposed architecture effectively bridges spatial and biological verification through edge-computing mobile devices. Utilizing native Bluetooth Low Energy (BLE) Application Programming Interfaces, the framework establishes a localized, hardware-independent classroom perimeter, enforcing a strict presence threshold of RSSI > -80 dBm. Following this spatial validation, the integration of TensorFlow Lite and Google ML Kit facilitates advanced liveness detection and local facial embedding extraction. By mathematically confirming student identities against cloud-registeredtemplatesusingCosineSimilarity(whereA•B/ ||A|| × ||B|| ? 0.85), the system practically eliminates biometricpresentation attacks. Implemented natively on Android with a secure Firebase Firestore backend, the SAS provides a highly scalable and cost-effective alternative to traditional attendance systems. While minor operational constraints exist regarding environmental BLE signal variability and mobile hardware dependencies, the foundational architecture demonstrates profoundimprovementsindataintegrityandprocessingspeed (consistentlyoperatingunder15seconds). Ultimately, this dual-layer verification framework illustrates the significant potential of merging short-range network telemetry with localized artificial intelligence. It delivers a robust, transparent, and privacy-compliant solution that minimizes administrative overhead while rigorously securingacademic integrity. { online version of the volume will be available in LNCS Online.MembersofinstitutessubscribingtotheLectureNotes inComputerScienceserieshaveaccesstoallthepdfsofallthe online publications. Non-subscribers can only read as far as the abstracts. If they try to go beyond this point, they are automatically asked, whether they would like to order the pdf, and are given instructions as to how to do so.

References

[1] M. A. Raza, M. Akhtar, S. Akhtar, and M. A. Khan, \"A review of intelligent attendance systems using machine learning and deep learning,\"Int.J.Adv.Comput.Sci.Appl.,vol.13,no.4,pp.1–12,2022. [2] S. B. Kotsiantis, C. Pierrakeas, and P. E. Pintelas, \"Predicting students\' performance in distance learning using machine learning techniques,\"Appl.Artif.Intell.,vol.18,no.5,pp.411–426,2004. [3] V. S. Dhaka and S. Sinhal, \"Biometric attendance system to preventproxyattendanceineducationalinstitutions,\"inProc.2nd Int.Conf.Comput.Commun.Technol.(ICCCT),Allahabad,India, 2011, pp. 756–759. [4] N. P. Patidar, R. Agrawal, and P. Sharma, \"Smart attendance management using biometric system and internet of things,\"in Proc. Int. Conf. Emerg. Technol. (ICET), Jaipur, India, 2020, pp.1–5. [5] A. El-Rabbany, Introduction to GPS: The Global Positioning System, 2nd ed. Norwood, MA, USA: Artech House, 2006. [6] K.Jain,A.Ross,andS.Prabhakar,\"Anintroductiontobiometric recognition,\" IEEE Trans. Circuits Syst. Video Technol., vol. 14, no. 1, pp. 4–20, Jan. 2004. [7] K. Nambiar, \"RFID technology: A review of its applications,\" in Proc. World Congr. Eng. Comput. Sci. (WCECS), San Francisco, CA, USA, 2009, vol. 2, pp. 1–5. [8] G.A.Akintola,O.A.Akinde,andA.S.Alatishe,\"Aweb-based students\' attendance management system,\" Int. J. Sci. Technol. Res., vol. 2, no. 7, pp. 164–168, Jul. 2013. [9] H. F. Lin, \"Attendance management system using QR code,\"in Proc. IEEE Int. Conf. Consum. Electron.-Taiwan (ICCE-TW), Nantou, Taiwan, 2017, pp. 193–194. [10] Y.JavedandM.Khan,\"GPSbasedstudentattendancesystem,\" in Proc. 10th Int. Conf. Digit. Inf. Manage. (ICDIM), Jeju, South Korea, 2015, pp. 100–103. [11] P. Davidson and R. Piché, \"A survey of selected indoor positioning methods for smartphones,\" IEEE Commun. Surv. Tutor., vol. 19, no. 2, pp. 1347–1370, 2nd quart. 2017. [12] H.Liu,H.Darabi,P.Banerjee,andJ.Liu,\"Surveyofwireless indoor positioning techniques and systems,\" IEEE Trans. Syst. Man Cybern. C Appl. Rev., vol. 37, no. 6, pp. 1067–1080, Nov.2007. [13] R. Harle, \"A survey of indoor inertial positioning systems for pedestrians,\" IEEE Commun. Surv. Tutor., vol. 15, no. 3, pp. 1281–1293, 3rd quart. 2013. [14] F. Zafari, A. Gkelias, and K. K. Leung, \"A survey of indoor localization systems and technologies,\" IEEE Commun. Surv. Tutor., vol. 21, no. 3, pp. 2568–2599, 3rd quart. 2019. [15] R. Faragher and R. Harle, \"Location fingerprinting with Bluetoothlowenergybeacons,\"IEEEJ.Sel.AreasCommun.,vol. 33, no. 11, pp. 2418–2428, Nov. 2015. [16] P. Spachos and K. N. Plataniotis, \"BLE beacons for indoor positioningataninteractiveIoT-basedsmartmuseum,\"IEEESyst. J., vol. 14, no. 3, pp. 3483–3493, Sep. 2020. [17] F. Schroff, D. Kalenichenko, and J. Philbin, \"FaceNet: A unified embedding for face recognition and clustering,\" in Proc. IEEE Conf. Comput. Vision Pattern Recognit. (CVPR), Boston, MA, USA, 2015, pp. 815–823. [18] A. G. Howard et al., \"MobileNets: Efficient convolutional neural networks for mobile vision applications,\" arXiv preprint arXiv:1704.04861, 2017. [19] TensorFlowLite,\"TensorFlowLite:On-deviceMLformobile and edge devices,\" Google LLC. [Online]. Available:https://www.tensorflow.org/lite. [Accessed: Apr. 2026]. [20] Z. Boulkenafet, J. Komulainen, and A. Hadid, \"Face anti-spoofing using speeded-up robust features and fisher vector encoding,\"IEEESignalProcess.Lett.,vol.24,no.2,pp.141–145, Feb. 2017. [21] A. Jourabloo, Y. Liu, and X. Liu, \"Face de-spoofing: Anti-spoofingvianoisemodeling,\"inProc.Eur.Conf.Comput.Vision (ECCV), Munich, Germany, 2018, pp. 290–306. [22] P. Mell and T. Grance, The NIST Definition of Cloud Computing, NIST Special Publication 800-145, Gaithersburg, MD, USA: National Institute of Standards and Technology, 2011. [23] S. Bhatt and R. Bhatt, \"Multi-factor authentication for smart campusattendance:Acasestudy,\"J.Comput.Educ.,vol.8,no.2,pp. 233–251, Jun. 2021. [24] M. A. Khan, F. Algarni, and M. T. Quasim, \"A decentralised IoT-based authentication system with multi-factor biometric verification for smart cities,\" IEEE Access, vol. 8, pp. 127298–127308, 2020. [25] T.PadmaandT.Kanimozhiselvan,\"Asurveyonmobilebased attendancemanagementsystems,\"inProc.3rdInt.Conf.Commun. Electron. Syst. (ICCES), Coimbatore, India, 2018, pp. 816–821. [26] EuropeanParliamentandCouncil,\"Regulation(EU)2016/679 of the European Parliament and of the Council (General Data Protection Regulation),\" Official Journal of the European Union,L 119, pp. 1–88, May 2016. [27] A. G. Howard et al., \"Searching for MobileNetV3,\" in Proc. IEEE/CVF Int. Conf. Comput. Vision (ICCV), Seoul, South Korea, 2019, pp. 1314–1324. [28] Google LLC, \"Firebase Firestore documentation: Security rules and data modeling,\" Google LLC. [Online]. Available: https://firebase.google.com/docs/firestore. [Accessed: Apr. 2026]. [29] Google LLC, \"ML Kit: Face detection API documentation,\"Google LLC. [Online]. Available: https://developers.google.com/ml-kit/vision/face-detection.[Accessed: Apr. 2026]. [30] Android Developers, \"Bluetooth low energy overview,\"Google LLC. [Online]. Available: https://developer.android.com/guide/topics/connectivity/bluetooth /ble-overview.[Accessed:Apr.2026].

Copyright

Copyright © 2026 Aditya Kiran Kumbhar, Tushar Parasharam Mane, Sanket Shrikant Shinde, Avishkar Anil Shinde, Raviraj Uddhav Patil, Jayan Juned Mulla, Ashok Sadavare. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.