Educational institutions require security systems to account for their students in evacuations during disasters like fires and earthquakes. Additionally, schools seek to minimize the risk of intruders on campus by implementing appropriate security systems in campus buildings, but the high cost and impracticality of such systems often hinder schools from doing so. Current methods for tracking, such as GPS and camera surveillance, also violate student privacy.

Radio-frequency identification (RFID) technology is commonly used in corporate building security and product tracking. Despite numerous applications for RFID technology existing, these functions are seldom used together. My project combines the potential of RFID for security and tracking to create a network that simultaneously identifies the approximate location of students and affords them individualized clearance to school buildings.

I integrated passive MIFARE tags storing 1 kB of data into different forms of apparel that students use to access campus buildings. Assigning unique identifiers (UID) to each student's shoe or wristband and placing RFID checkpoints on the entrance and exits of each campus building allows schools to also receive data about the general building occupied by students instead of students' exact locations, which makes the process less intrusive.

The checkpoints' hardware consists of WiFi-enabled RFID readers on doors that function as both locks and tracking checkpoints. The checkpoints use ESP-12E microcontrollers, based on the ESP8266 WiFi module, and they communicate with the 13.56 MHz RFID readers using SPI. The microcontroller actuates a solenoid locking mechanism, and the zinc-plated door handles utilize capacitive sensing to identify the direction from which a student opens a door.

I programmed the ESP-12E in C/C++ to identify the student associated with a given RFID UID. The program then communicates the student's UID, name, contact information, and whether or not the student is inside a given building to a Firebase database. After unlocking, the checkpoint's updated information is transmitted to a dynamic webpage that creates a roster providing authorities with information vital for responding to natural disasters, missing person cases, and evacuations.

Examining the cycle time data from tests with simulated datasets of 10 to 100,000 students reveals that each checkpoint can process approximately 6 students per second. Whether the network compares a UID to 10 or 100,000 values has an insignificant effect on the processing time, which confirms the scalability of the model. Each checkpoint costs $9.68 to produce, which is less than 6% of the price of widely-used alternative IoT locks currently on the market, and the price of implementing a single RFID element into clothing is merely 8 cents.

My prototypes demonstrate that a system relying on RFID location logging is more cost-efficient and less invasive than techniques that use constant monitoring, making it a viable alternative to traditional security systems. By designing a seamless process for protecting students, I exemplified how RFID technology can be applied to security systems in any field at a low cost without compromising user privacy. In a world where privacy and safety are simultaneously prioritized, a security system like this one creates an innovative solution with convenience, affordability, and exceptional scalability.