AESPIE: Raspberry Pi AES Performance Evaluation Using Image Encryption in C and Python

Image encryption is crucial for safeguarding private and official documents online, while IoT devices and embedded systems handle large volumes of user data, necessitating cryptographic processing to ensure security. Key factors affecting IoT device performance include the chosen encryption algorithm, programming language, hardware acceleration, processor specifications, and data size. Evaluating these parameters for AES, a widely used encryption standard, reveals its effectiveness. AES is a 128-bit block cipher supporting key lengths of 128, 192, and 256 bits (AES128, AES192, and AES256), serving various computing applications for data security in industrial, commercial, and military domains. We employ a 700 × 875 RGB image dataset as a benchmark to evaluate the performance of various hardware and software tools using AES variants (AES128, AES192, and AES256). Encryption and decryption processes are executed in Python and C on Raspberry Pi, a widely adopted platform for IoT applications. Our focus lies in profiling the AES encryption and decryption stages and gathering execution time and overall performance data for the AES symmetric key algorithm. Results from the benchmark indicate that C outperforms Python significantly, with C being 46 times faster at encryption and 6 times faster at decryption. Additionally, in Python, AES decryption is 1.3 times slower than encryption, while in C, AES decryption is 10 times slower than encryption.