An Applied Security Model For Controlled Key-Release In Blowfish-Based Data Retrieval

Abstract

From a cybersecurity standpoint, the security of the data stored in these systems has become a crucial concern as cloud storage continues to proliferate in digital infrastructure. Credential compromise, brute-force attacks, and sophisticated, persistent infiltration tactics have demonstrated the vulnerability of traditional security techniques, especially those that rely on static encryption keys or password-based authentication. This study offers a multi-layered security approach that combines multi-stage authentication with a dynamic key-release method for Blowfish-encrypted file storage, in response to constantly evolving cyber threats. Before allowing the controlled recovery of the Blowfish decryption key, the model described here used sequential verification with user credentials, a time-based one-time password, and a registered device token validation procedure. By keeping encryption keys unreadable until identity is fully confirmed, the system architecture reduces the exposure risks associated with conventional key storage methods. FastAPI, Streamlit, PostgreSQL, and MinIO were used in the implementation; the configuration was simple but deployable in both academic and business environments. According to experimental evaluation, the framework significantly increases resistance to unauthorised access attempts while maintaining low latency for both authentication and decryption. The findings make it clear that the suggested architecture provides a safe and effective way to protect files in cloud environments, balancing computational efficiency and user accessibility while ensuring robust cryptographic security.