Blind Quantum Computing with Decoy States

TL;DR

This paper introduces a modified remote blind qubit state preparation protocol using decoy states, significantly improving efficiency and speed for quantum computing over long-distance fiber links.

Contribution

A new RBSP protocol with decoy states that reduces pulse requirements from O(1/T^4) to nearly O(1/T), enhancing long-distance quantum computing efficiency.

Findings

Reduces pulse requirement for single qubit generation by six orders of magnitude over 100 km fiber.

Enables simultaneous generation of multiple qubits in a single protocol instance.

Significantly improves the practicality of blind quantum computing over long distances.

Abstract

In this paper, we study the Universal Blind Quantum Computing (UBQC) protocol, which allows a client to perform quantum computation on a remote quantum server and the Remote Blind qubit state Preparation (RBSP) protocol which allows the client to prepare single qubits on the server's location with week coherent pulses. This RBSP protocol is inefficient for small $ T $, because the required number of pulses for generating one single qubit scales as $ \mbox{O}(1/T^4) $, where $ T $ is the transmittance of the quantum channe between the client and the server. Our contribution in this paper is a modified RBSP protocol with decoy states. This new protocol allows us to generate $ S $ single qubits simultaneously in a single instance. With our new rotocol, we are able to reduce the number of weak coherent pulses required for generating one single qubit from $ \mbox{O}(1/T^4) $ to nearly $ \mbox{O}(1/T) $, significantly increasing the efficiency. For a 100 km standard telecom fiber of loss 0.2 dB/km, our method gives a six orders of magnitude improvement in the speed, which is very significant.