for automating restroom access, details the hardware and software architecture, reproduces key figures and

tables, reports empirical performance findings and situates the system within broader debates about

surveillance, consent, privacy and digital inclusion. Ultimately the work demonstrates that a thoughtfully

implemented biometric access system can enhance exam integrity while raising questions about equity,

autonomy and trust.

surveillance; digital inclusion.

Examinations are essential sites where educational institutions endeavour to reproduce meritocratic ideals.

Yet such ideals can be undermined by lax procedural controls. A seldom-examined aspect of exam

monitor the room and track restroom usage.

Biometric technologies promise to automate identity verification and record keeping. Fingerprint scanners

in particular are inexpensive, unobtrusive and widely understood. By coupling a fingerprint sensor to a

microcontroller and linking it with a database, it becomes possible to grant or deny restroom access based

on pre-registered templates, log entry and exit times automatically and enforce policies such as maximum

occupancy. In the original engineering report a system was developed for Universiti Teknikal Malaysia

Melaka and tested with a small cohort of volunteer students. This paper explores why such a system is

needed, how it was designed and what social implications it carries.

From a sociological perspective, toilets are more than functional necessities; they reflect cultural norms,

hierarchies and control. Studies of public facilities reveal that access is often a site where class, gender and

examination.

Biometric identification has been the subject of intensive research for decades. Jain et al. provide a

comprehensive introduction to the principles of biometrics, analysing modalities such as fingerprint, face,

iris and voice recognition [1]. They highlight that fingerprints remain the most mature modality due to their

uniqueness, stability and ease of acquisition [1]. Ahmed et al. examine failures in manual access control

and argue that such systems rely on human vigilance and therefore suffer from fatigue and bias [2]. Their

qualitative study of office buildings found that untrained security guards frequently overlook suspicious

behaviour. Ali et al. analyse the implementation of biometric access in educational institutions; their case

Zheng and Valli identify critical aspects of biometric security such as spoofing, template storage and

liveness detection [6]. Lin surveys sensor technologies and notes that improvements in optical and

capacitive sensors are making high-quality fingerprint scanning more accessible [7]. Das and Sengupta

explore the integration of biometric systems with the Internet of Things and highlight challenges around

network latency, encryption and interoperability [8]. These studies demonstrate that design choices—not

just algorithmic sophistication—affect the security and reliability of biometric systems.

Scholars in surveillance studies argue that biometric systems may normalise suspicion and expand

institutional reach. For example, Lyon describes surveillance as social sorting, where individuals are

categorised and treated differently based on data profiles [9]. Whitley and Kroener argue that consent in

biometric systems is often coerced when access to essential services depends on compliance [10].

integrated Wi-Fi and Bluetooth, large memory and low cost make it suitable for networked biometric

applications. The AS608 optical fingerprint sensor provides reliable enrolment and verification by

capturing high-resolution images and extracting minutiae features. On the software side, the Arduino IDE

was used for firmware development and XAMPP for hosting the web interface due to their ease of use and

cross-platform availability.

Protection Act (PDPA) classify fingerprint templates as sensitive personal data. Institutions using

biometrics must ensure informed consent, implement encryption, and define retention periods. The

ISO/IEC 24745 standard recommends separating biometric templates from personal identifiers and

employing liveness detection. Failure to follow these guidelines can lead to legal liability and erosion of

trust.

context [12].

The system’s core is an ESP32 microcontroller (Figure 1), selected for its processing power, memory and

For user feedback, it employs a 0.96-inch OLED display and a 16×2 LCD display. A push button triggers

the scanning process. Figure 3 illustrates the block diagram of the system, showing the flow between inputs

(fingerprint sensor, button), the ESP32 control unit and outputs (OLED, LCD), with communication to a

PHP server via Wi-Fi. Wiring is implemented on a breadboard (Figure 4), and components are housed in a

payload to the server. If no match is found, the OLED instructs the user to retry. Registrations require an

admin token: the device generates a new user ID, collects metadata (matrix number, full name) and enrols

two samples. The server side uses XAMPP (Apache, MySQL, PHP) to host a database and a web interface.

Endpoints support operations to register users, record logins/logouts and display logs. These design

decisions were informed by the report’s emphasis on low cost and ease of use.

under exam conditions. Participants completed at least five login/logout cycles. Observations and informal

were automatically stored in the MySQL database. Qualitative feedback was summarised to identify

Introducing biometric devices into exam halls could normalise surveillance. Foucault’s concept of

panopticism suggests that individuals internalise monitoring and discipline themselves accordingly [11].

While participants in this study largely accepted the device as fair, long-term use might foster a culture of

mistrust. Transparent explanation and open dialogues can help maintain trust. Institutions should avoid

extending biometric monitoring to other domains without clear justification.

The low cost (~US$50) makes the system accessible to universities but may still be prohibitive for poorly

funded schools. Unequal adoption could widen disparities in exam integrity. Policy interventions or

subsidies might be necessary. Cultural and religious objections to fingerprint scanning require

system in diverse conditions and measure long-term behavioural impacts. Multi-modal biometrics and

identity verification and logging. Empirical results showed high recognition accuracy and substantial time

savings. However, ethical issues around privacy, consent, surveillance culture and inclusivity require

careful consideration. Deploying such systems responsibly entails robust data protection, transparent

policies and accommodations for diverse users. Ongoing dialogue between engineers, educators,

policymakers and students is essential to ensure that technological solutions uphold educational values and

respect human dignity.

The authors would like to dedicate our appreciation to Universiti Teknikal Malaysia Melaka for supporting

this research. Our gratitude also goes to the Centre of Research and Innovation Management (CRIM)

UTeM and Centre for Telecommunication Research and Innovation (CeTRI) UTeM for providing the

facilities needed to conduct this study.

1. A. K. Jain, A. Ross and K. Nandakumar, Introduction to Biometrics. Springer, 2011,

doi:10.1007/978-0-387-77326-1.

International Journal of Computer Applications, vol. 182, no. 30, pp. 1–5, Oct. 2019.

5. Y. Choi, J. Lee and H. Kim, “Deep learning techniques in fingerprint recognition: An overview,”

IEEE Access, vol. 11, pp. 252–265, Jan. 2023.

6. Z. Zheng and C. Valli, “Critical aspects of security and recognition accuracy in fingerprint systems,”

Journal of Information Security and Applications, vol. 46, pp. 121–132, Mar. 2019,

doi:10.1016/j.jisa.2019.03.016.

7. D. Lin, “Review of fingerprint sensor technologies,” Sensors, vol. 23, no. 5, p. 4679, May 2023,

doi:10.3390/s23054679.

8. S. Das and S. Sengupta, “Biometric systems integration in IoT: Challenges and solutions,” IEEE

Internet of Things Journal, vol. 8, no. 11, pp. 9297–9308, Nov. 2021,