An intelligent homomorphic audio signal encryption algorithm for secure interacting

An intelligent homomorphic audio signal encryption algorithm for secure interacting

연구 분야: Cryptography

학회: Multimedia Tools and Applications

The present encryption algorithm has high computational complexity for the large data volume as audio and usually focuses on one audio format. Furthermore, the data in the cloud cannot be modified before decrypting, which means the interaction between the encrypted audio and the users is difficult. To address the above problems, a new homomorphic audio signal encryption algorithm has been developed for secure processing and interaction in the cloud. The original audio signal is encrypted without converting to binary to reduce the computational complexity. Adaptive parameters were generated to deal with varied audio formats and attributes. The users can choose the appropriate encryption level to balance the security and complexity. Meanwhile, this scheme supports additive and multiplicative homomorphism to perform some operations (e.g., volume adjustment and audio editing) correctly with the encrypted data. Security analysis and experiments show that the proposed algorithm has better encryption performance, stronger sensitivity to key, higher efficiency, less data expansion, and could be against attacks from statistical analysis. Overall, the proposed algorithm is secure and efficient to satisfy requirements for interaction with encrypted audio files in the cloud.