Abstract
Quantum k-means algorithm is widely used in solving clustering problems. However, the repeated calculation of the core subroutines overloads the quantum computer with mass computing tasks. In addition, due to the decoherence of the quantum state, the computing power of personal quantum computer is limited. Therefore, it is necessary to use high-performance quantum cloud computing to complete data processing. This paper proposes a quantum k-means algorithm in quantum cloud computing, which uploads the data to the server to execute the core subroutines SwapTest (calculating the similarity between ciphertext quantum states) and GroverOptim (quantum minimization algorithm) of quantum k-means algorithm and then sends the ciphertext to the client for further decryption using the decryption key, which reduces the load of the client. An improved quantum homomorphic encryption by t-gate updating in trusted server and data processing in semi-trusted server is introduced as well. In this scheme, the trusted server assists the semi-trusted server to execute the T-gate existing in the quantum circuit, which separates the key from the ciphertext and ensure the security of the ciphertext data processing in the quantum cloud. In our experiment, the two subroutines, SwapTest and GroverOptim, are conducted in ciphertext and plaintext conditions by using IBM Qiskit to evaluate the reliability and feasibility. The result shows that the decrypted ciphertext result is the same as the plaintext. We conclude that the proposed algorithm has a better performance in reducing the cost of load in client and protecting privacy data in cloud by contrast with the original quantum k-means algorithm.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have