Analysis and improvement of verifiable blind quantum computation

Abstract

In blind quantum computation (BQC), a client with weak quantum computation capabilities is allowed to delegate its quantum computation tasks to a server with powerful quantum computation capabilities, and the inputs, algorithms and outputs of the quantum computation are confidential to the server. Verifiability refers to the ability of the client to verify with a certain probability whether the server has executed the protocol correctly and can be realized by introducing trap qubits into the computation graph state to detect server deception. The existing verifiable universal BQC protocols are analyzed and compared in detail. The XTH protocol (proposed by Xu Q S, Tan X Q, Huang R in 2020), a recent improvement protocol of verifiable universal BQC, uses a sandglass-like graph state to further decrease resource expenditure and enhance verification capability. However, the XTH protocol has two shortcomings: limitations in the coloring scheme and a high probability of accepting an incorrect computation result. In this paper, we present an improved version of the XTH protocol, which revises the limitations of the original coloring scheme and further improves the verification ability. The analysis demonstrates that the resource expenditure is the same as for the XTH protocol, while the probability of accepting the wrong computation result is reduced from the original minimum (0.866) d* to (0.819) d*, where d* is the number of repeated executions of the protocol.