Enhancing Data Security Using Rubik’s Encryption and CNN Model for Biometric Authentication

With the rising need for secure authentication and data protection, this paper proposes a multi-layered security framework integrating biometric authentication, encryption, and covert communication. It employs Convolutional Neural Networks (CNN) for iris and fingerprint recognition, fusing extracted feature maps to generate a zero-bit watermark as a cryptographic key. This key encrypts sensitive documents using Rubik’s encryption, ensuring robust security. A two-factor authentication (2FA) mechanism via Telegram Bot API sends an OTP for decryption access. Additionally, an audio steganographic approach embeds confidential messages within audio files, revealing the hidden text only to authorized users while playing a normal song otherwise. This system enhances security, offering resilience against unauthorized access, cryptographic attacks, and data breaches while maintaining high efficiency and scalability.