Blind quantum computation with completely classical client and a trusted center

TL;DR

This paper introduces a more practical blind quantum computation protocol that allows a completely classical client to delegate quantum tasks to one or two servers with a trusted center, simplifying previous multi-server protocols.

Contribution

It proposes a modified double-server BQC protocol with a trusted center, enabling a classical client and reducing quantum requirements compared to prior multi-server protocols.

Findings

Protocols are more practical for cloud-based quantum computing.

Client does not need quantum access or communication abilities.

Single-server protocol derived from double-server setup.

Abstract

Blind quantum computation (BQC) enables a client without enough quantum power to delegate his quantum computation to a quantum server, while keeping the input data, the algorithm and the result unknown to the server. In the studies of practical BQC protocol, an important problem is how to reduce the quantum requirement of the client. Multi-server BQC protocols have been proposed to solve this problem. We review the double-server and triple-server protocols [Li et al., Phys. Rev. A 89, 040302(R) (2014)], and propose a modified double-server BQC protocol with a trusted center. In our protocol, the servers are allowed to communicate mutually, and the client is completely classical. Furthermore, our double-server protocol can be modified into a single-server protocol by simply combining the two servers. Compared with the triple-server protocol, our double-server and single-server protocols are more simple and the client is not required to have the ability to access quantum channel. So our protocols are more practical when quantum computer is applied in the `cloud' model.