Multiparty blind quantum computation protocol with deterministic mutual identity authentication
Introduction
Blind quantum computation (BQC) allows a client with limited quantum computation power to delegate a computing task to a quantum server confidentially. It is of critical importance nowadays due to the expensive cost and difficult maintenance of quantum computer. Owing to its excellent prospect, BQC attracts lots of attention and has been widely developed [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15]. The pioneering work was presented by Childs using the circuit-based quantum computing model [2]. In this protocol, the client is required to have a large quantum memory and the abilities of accessing the quantum channel. Subsequently, various BQC protocols were proposed based on a trusted third party [10] or more than one server [4]. However, it is impossible to achieve an ideal single-server BQC protocol with purely classical client without a trusted third party [11], [12], [13], [14], [15].
Authenticating the users’ identities is a prerequisite for quantum secure communications. Authenticating the identities of the client and the server is also a prerequisite for the security of BQC protocols. To resist attacks such as man-in-the-middle attack and denial-of-service attack, Li et al. [16] introduced the identity authentication to BQC, and proposed single-server and double-server BQC protocols based on a third party, respectively. However, the third party is required to be trusted. Subsequently, Shan et al. proposed a multiparty BQC protocol with mutual identity authentication [17]. In their protocol [17], there are four different roles: client, server, load balancer and certificate authority (CA). There are two load balancers, i.e., Load Balancer A and Load Balancer B. Load Balancer A takes charge of allocating m clients while Load Balancer B allocates n servers. CA is a semi-trusted third party which can help the generation of authentication keys in the registration phase by using measurement-device-independent quantum key distribution (MDI-QKD) [18] and the mutual identity authentication of the client and the server in the mutual identity authentication phase. The protocol in the third phase is similar to the single-server BQC protocol [12]. Unfortunately, the efficiencies of both the authentication key generation and the mutual identity authentication are low. In the registration phase, and only keep the bits under the same basis as the raw shared key when CA’s measurement result is . Here, . According to Table 1 in Ref. [17], this means that only a raw key bit is generated for every four pairs of single photons. In the mutual identity authentication phase, according to Table 2 in Ref. [17], only when CA’s state is or , there exists a quantum correlation between the measurement results of and . Here, . That implies that a half of the measurement results of and are useless for the mutual authentication thus leading to a lower authentication efficiency.
To solve these problems above, we present a variant of MDI-QKD and based on this, we further present a multiparty BQC protocol with deterministic mutual identity authentication where both the generation of the authentication keys of the registered client and the designated server and the authentication of their identities are performed in a deterministic way. Therefore, the protocol efficiency is improved significantly.
The rest of this article is organized as follows. In Section 2.1, a variant of MDI-QKD is presented. And then in Section 2.2, a concrete BQC protocol with deterministic mutual identity authentication is proposed based on the variant of MDI-QKD. The corresponding analyses are presented in terms of correctness, blindness and security of the proposed BQC protocol in Section 3. Discussion and conclusion are given in the last section.
Section snippets
Variant of MDI-QKD
The variant of MDI-QKD protocol can be implemented by subtly modifying the original MDI-QKD protocol [18]. The graph corresponding to the variant of MDI-QKD is given to make it more visible (see Fig. 1).
The variant of MDI-QKD is described as follows.
(1) Bob prepares a sequence of n photon pairs, each of which is randomly in one of the two Bell states . He extracts one photon from each pair to form the sequence , and the other photons form the sequence . Meanwhile, Alice prepares
Correctness, blindness and security analysis
In this section, we analyze the proposed BQC protocol from the aspects of the correctness, blindness and security. The correctness of Phases 1 and 2 has been shown in Section 2. Here, we focus on the analysis of the correctness of Phase 3. Blindness refers to keeping the client’s input, output and computation private when he delegates a quantum computing task to the server. So, the blindness analysis deals with only Phase 3. Finally, the security of the three phases is discussed under various
Discussion and conclusion
In this section, we make a comparison among the proposed BQC protocol and existing BQC protocols from multiple aspects, as shown in Table 3.
In fact, our BQC protocol with mutual identity authentication can be easily adapted for multi-party quantum network by introducing the load balancers. Our protocol can achieve higher authentication key generation rate and authentication efficiency than the one in Ref. [17]. In our BQC protocol with mutual identity authentication, the only requirement is
CRediT authorship contribution statement
Yu-Guang Yang: Designed research, Performed research, Wrote the paper, Reviewed the manuscript. Rui-Chen Huang: Designed research, Performed research, Reviewed the manuscript. Yi-Hua Zhou: Designed research, Reviewed the manuscript. Wei-Min Shi: Designed research, Reviewed the manuscript. Guang-Bao Xu: Designed research, Reviewed the manuscript. Dan Li: Designed research, Reviewed the manuscript.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Grant Nos. 62071015, 62171264).
Data availability statement
Data available on request from the authors.
References (40)
- et al.
Blind quantum computation with identity authentication
Phys. Lett. A
(2018) - et al.
Detector-device-independent quantum key agreement based on single-photon Bell state measurement
Internat. J. Theoret. Phys.
(2022) - et al.
New quantum key agreement protocols based on Bell states
Quantum Inf. Process.
(2019) - et al.
Two-party quantum key agreement over a collective noisy channel
Quantum Inf. Process.
(2019) - et al.
New quantum key agreement protocols based on cluster states
Quantum Inf. Process.
(2019) - et al.
Three-party quantum secret sharing against collective noise
Quantum Inf. Process.
(2019) - et al.
Multiparty anonymous quantum communication without multipartite entanglement
Quantum Inf. Process.
(2022) - et al.
Blind quantum computation
Int. J. Quantum Inform.
(2006) Secure assisted quantum computation
Quantum Inf. Comput.
(2005)- et al.
Universal blind quantum computation