Quantum-computer architecture using nonlocal interactions

Abstract

Several authors have described the basic requirements essential to build a scalable quantum computer. Because many physical implementation schemes for quantum computing rely on nearest neighbor interactions, there is a hidden quantum communication overhead to connect distant nodes of the computer. In this paper we propose a physical solution to this problem which, together with the key building blocks, provides a pathway to a scalable quantum architecture using nonlocal interactions. Our solution involves the concept of a quantum bus that acts as a refreshable entanglement resource to connect distant memory nodes providing an architectural concept for quantum computers analogous to the von Neumann architecture for classical computers. Most modern computers share the same basic architecture first proposed by von Neumann in 1945. Von Neumann organized a computer into four basic components: memory, an input/output system, an arithmetic logic unit, and a control unit. The four units were interconnected by a bus that provided for the flow of classical bits or information between the various components [1]. Basic elements sufficient to build a scalable quantum computer have been described by DiVincenzo [2] and Preskill [3]. The five DiVincenzo criteria [2] for building a quantum computer are: a scalable physical system with well characterized qubits, the ability to initialize the state of the qubits to a simple fiducial state, long relevant decoherence times, a universal set of quantum gates, and a qubit specific measurement capability. In addition, Preskill lists other elements necessary for fault tolerant computation in order to maintain a reasonable accuracy threshold. Two of these are maximal parallelism and gates that can act on any pair of qubits.Although Preskill [3] communicates the need to interact arbitrary pairs of qubits, he provides no solution for this in a typical quantum computer restricted to nearest neighbor interactions. DiVincenzo [2] mentions two additional criteria essential for quantum communications namely: the ability to interconvert stationary and flying qubits, and the ability to faithfully transmit flying qubits between specified locations. Clearly, if such capabilities were engineered into the architecture, the above requirements of qubit interconnectivity and parallelism could simultaneously be satisfied. In this paper we show an alternative approach based on the concept of a quantum bus that consists of refreshable qubits that act as a resource for entanglement. This concept bears similarity to the classical bus, key to the von Neumann architecture.For concreteness we consider a lattice model of a quantum computer (e.g. a neutral atom optical lattice, quantum dot arrays, or 31 P embedded Si [4]) where qubits are fixed in position and interactions are with nearest neighbors (Fig. 1). One obvious way to connect distant qubits is to swap the states through intermediary qubits until the states are adjacent to each other, perform the requisite operations, and then swap back. This procedure s...

General bounds for sender-receiver capacities in multipoint quantum communications

“…More generally, we also consider the combination of these two cases, where many senders communicate with many receivers in a sort of all-in-all quantum communication or quantum interference channel. In practical implementations, this may represent a quantum bus [40,41] where quantum information is transmitted among an arbitrary number of qubit registers.…”

Controlled collisions for multi-particle entanglement of optically trapped atoms

“…The Mott state itself, with one atom per lattice site, could act as a huge quantum memory, in which information would be stored in atoms at different lattice sites. Going far beyond these ideas, it has been suggested that controlled interactions between atoms on neighbouring lattice sites could be used to realise a massively parallel array of neutral atom quantum gates 5,[11][12][13][14] , with which a large multi particle system could be highly entangled 6 in a single operational step. Furthermore, the repeated application of the quantum gate array could form the basis for a universal quantum simulator along the original ideas of Feynman for a quantum computer as a simulator of quantum dynamics [15][16][17] .…”

A quantum-walk-inspired adiabatic algorithm for solving graph isomorphism problems

“…In other physical implementations, such as crystal of trapped ions [39], the realization of the XY -Hamiltonian, together with its control and the preparation of its ground state, will be quite straightforward. In this setup, however, it is the realization of the ZZ interactions between distant qubits that may be very challenging, and would require some sort of quantum bus [40]. The defini-While the value ǫ can be usually bounded from above by a polynomial in the system size n, the spectral gap g min can happen to have an exponential dependence on the system size.…”