This study presents the implementation and evaluation of the ASCON lightweight cryptographic
algorithm on Internet of Things (IoT) devices using Arduino Uno microcontrollers.
The research focuses on analyzing the performance of ASCON using different authentication
tag lengths (8, 12, and 16 bytes), with a detailed assessment of encryption time, decryption
time, and the avalanche effect measured by Hamming Distance. The system simulates
Machine-to-Machine (M2M) communication between two Arduino Uno devices using UART
serial communication. Given Arduino Uno ’s hardware limitations (2 KB SRAM and 32 KB
flash memory), the ASCON library was manually adapted to ensure compatibility and memory
efficiency, enabling secure authenticated encryption within a constrained embedded environment.
The results indicate that the 8-byte tag configuration offers the best balance between speed,
memory usage, and security. It achieved the lowest average encryption and decryption times
(2208–2236 ? s), with a strong avalanche percentage (46–50%, peaking at 50.96%), while also
consuming the least memory (22% flash and 42% SRAM). In comparison, the 12-byte and
16-byte tags provided more stable avalanche values (up to 50.83%) but required additional
program storage and processing time. This study concludes that the 8-byte tag is optimal for
ASCON implementation on resource-constrained IoT platforms, offering a practical trade-off
between cryptographic strength and hardware efficiency. The implementation can serve as a
lightweight cryptographic solution for secure data transmission in real-world IoT scenarios.
Key words: ASCON, lighweight algorithm, cryptographic, IoT.