SeCPlat: A Secure Computation Platform Based on Homomorphic Encryption in Cloud

To achieve diversified secure computations on encrypted data, in this paper, we propose and demonstrate a Secure Computation Platform (SeCPlat) based on homomorphic encryption in cloud environments. Using the Paillier cryptosystem as the foundation, we design two main sets of secure computation protocols, i.e., Basic Secure Protocols and Advanced Secure Protocols. Basic Secure Protocols support arithmetic operations, comparison operations, and bit operations in ciphertext, while Advanced Secure Protocols upon the basic protocols can provide secure computational operations (e.g., Euclidean distance) in more complex algorithms. For a better demonstration, we further developed a prototype for secure computation applications, in which several secure algorithms were implemented to show the high supportability to secure computational tasks and the user interfaces were designed to setup and illustrate the parameters, results, and workflows of SeCPlat. Moreover, we provided APIs to facilitate the implementation of secure computational applications.