In the setting of networked computation, data security can be a significant concern. Here we consider the problem of allowing a server to remotely manipulate client supplied data, in such a way that both the information obtained by the client about the server's operation and the information obtained by the server about the client's data are significantly limited. We present a protocol for achieving such functionality in two closely related models of restricted quantum computation-the boson sampling and quantum walk models. Because of the limited technological requirements of the boson scattering model, small scale implementations of this technique are feasible with present-day technology.
In the setting of networked computation, data security can be a significant concern. Here we consider the problem of allowing a server to remotely manipulate client supplied data, in such a way that both the information obtained by the client about the server's operation and the information obtained by the server about the client's data are significantly limited. We present a protocol for achieving such functionality in two closely related models of restricted quantum computation-the boson sampling and quantum walk models. Because of the limited technological requirements of the boson scattering model, small scale implementations of this technique are feasible with present-day technology.
Client-server models enable computations to be hosted remotely on quantum servers. One key requirement for these models is to maintain the privacy of client tasks. We present a novel protocol which meets this requirement, with practical advantages when using technology feasible in the near term. Client tasks are realized as linear combinations of operations implemented by the server, where the linear coefficients are kept secret from the server. We report on an experimental demonstration of our protocol using linear optics, which realizes linear combination of two single-qubit operations.
Abstract Client-server models enable computations to be hosted remotely on quantum servers. We present a novel protocol for realizing this task, with practical advantages when using technology feasible in the near term. Client tasks are realized as linear combinations of operations implemented by the server, where the linear coefficients are hidden from the server. We report on an experimental demonstration of our protocol using linear optics, which realizes linear combination of two single-qubit operations by a remote single-qubit control. In addition, we explain when our protocol can remain efficient for larger computations, as well as some ways in which privacy can be maintained using our protocol.
We give a protocol for the delegation of quantum computation on encrypted data. More specifically, we show that in a client–server scenario, where the client holds the encryption key for an encrypted quantum register held by the server, it is possible for the server to perform a universal set of quantum gates on the quantum data. All Clifford group gates are non-interactive, while the remaining non-Clifford group gate that we implement (the π/8 gate) requires the client to prepare and send a single random auxiliary qubit (chosen among four possibilities), and exchange classical communication. This construction improves on previous work, which requires either multiple auxiliary qubits or two-way quantum communication. Using a reduction to an entanglement-based protocol, we show privacy against any adversarial server according to a simulation-based security definition.
Abstract Quantum computing on encrypted data allows a client who has limited quantum capacity to delegate his or her private computation to an untrusted quantum server, while the input and output are encrypted by the quantum one-time pad and only the client can correctly decrypt them. Generally, the client is required to have ability to prepare some single qubits and perform some basic gates. In this work, we consider a further restricted situation where the client can only prepare one single qubit and perform one basic gate. Specifically, we show that as long as the client can prepare a fixed qubit and perform a fixed phase gate P , then he or she can still achieve the secure delegated quantum computation. Besides, our protocol can provide a more rigorous security for any quantum computation. For example, even if some encryption keys about the computation are leaked, it can still guarantee the privacy of the input and output. Finally, our protocol experimentally has a great significance in reducing the device complexity of the client's side.
We propose a protocol based on quantum teleportation and measurement-based computation allowing clients to conduct blind computation on the server with less information of input, output and computation revealed. Our protocol also requires less quantum resources on clients. There are two phases in our protocol, i.e., preparation and computation. Since we apply the small operation unit of an entanglement pair instead of cluster states or quantum gates, the server is able to serve more clients with limited resources, and provide flexible computation power and thus distribute his tasks with flexibility. The proposed protocol provides a new way to achieve blind computation with few quantum resources on clients and conceal clients' information from potential eavesdroppers. Moreover, to the best of our knowledge, the proposed protocol is the first to consider the reduction of the additional load on the server, and provides better scalability, flexibility and reliability for quantum blind computation.
Abstract An efficient technique of computing on encrypted data allows a client with limited capability to perform complex operations on a powerful remote server without leaking anything about the input or output. Quantum computing provides information-theoretic security to solve such a problem, and many such techniques have been proposed under the premises of half-blind quantum computation. However, these approaches are based on some universal combination of non-parametric gates ( H , S , T , C X , C Z , and C C X ). Hybrid quantum–classical algorithms require parametric gates, which, when decomposed into non-parametric gates, inadvertently increase the depth of the circuit and hence the communication rounds. We propose an approach for recursive decryption of any parametric gate, R z ( θ ) , without prior decomposition; exactly when θ = ± π / 2 m for m ∈ Z + , and approximately with arbitrary precision ϵ for any given θ . We also show that a blind algorithm based on such a technique needs at most O ( log 2 2 ( π / ϵ ) ) computation steps and communication rounds, while the techniques based on a non-parametric resource set require O ( ln 3.97 ( 1 / ϵ ) ) . We use this approach to propose a half-blind quantum computation protocol to enable efficient computation on encrypted data in the NISQ era.