Fault tolerant scalable modular quantum computer architecture with an enhanced control of multi-mode couplings between trapped ion qubits

• US 9,858,531 B1
• Filed: 08/01/2014
• Issued: 01/02/2018
• Est. Priority Date: 08/02/2013
• Status: Active Grant

• Current Assignee: Duke University, University of British Columbia
• Original Assignee: Duke University, University of British Columbia, University of Maryland
• Inventors: Monroe, Christopher, Kim, Jungsang, Raussendorf, Robert
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): HANCHAK, WALTER P
• Application Number:
  US14/449,477
• Time in Patent Office: 1,250 Days
• Field of Search: None
• CPC Class Codes: G02F 1/113 Circuit or control arrangements
• CPC Class Codes: G06F 15/76 Architectures of general purpose stored program computers with program plugboard G06F15/08; multicomputers G06F15/16
• CPC Class Codes: G06F 2015/768 Gate array
• CPC Class Codes: G06N 10/00 Quantum computing, i.e. information processing based on quantum-mechanical phenomena

Abstract

A modular quantum computer architecture is developed with a hierarchy of interactions that can scale to very large numbers of qubits. Local entangling quantum gates between qubit memories within a single modular register are accomplished using natural interactions between the qubits, and entanglement between separate modular registers is completed via a probabilistic photonic interface between qubits in different registers, even over large distances. This architecture is suitable for the implementation of complex quantum circuits utilizing the flexible connectivity provided by a reconfigurable photonic interconnect network. The subject architecture is made fault-tolerant which is a prerequisite for scalability. An optimal quantum control of multimode couplings between qubits is accomplished via individual addressing the qubits with segmented optical pulses to suppress crosstalk in each register, thus enabling high-fidelity gates that can be scaled to larger qubit registers for quantum computation and simulation.

Claims

1. Large-scale modular quantum computer architecture, comprising:

a processor configured for receiving an input data to be computed and to control quantum computations in accordance with a quantum algorithm and to output results of the computations;

a plurality of modular elementary logic units (ELUs), each modular ELU housing a plurality of stationary matter qubits and constituting a high performance quantum memory register;

a photonic interconnect network propagating quantum degrees of freedom of said qubits and operatively coupled to said plurality of modular ELUs, said photonic interconnect network being configured for multiplexing, under control of said processor, said plurality of modular ELUs in a dynamically reconfigurable at least one multi-dimensional quantum computational structure supporting a scheduled realization of at last one of at least first quantum gate is being realized between qubits housed in respective distant modular ELUs arranged in said at least one multi-dimensional quantum computations structure, and wherein said at least second quantum gate is being realized between qubits housed in a respective at least one of said plurality of modular ELUs, and wherein said at least first and second quantum gates are being executed through application of predetermined qubit-state-dependent forces to respective qubits;

a control sub-system configured to control said qubit-state-dependent forces for application of optimal control parameters to multiple modes of motion of said qubits housed in said respective at least one ELU to suppress a mode crosstalk within said respective at least one ELU, thereby enabling high fidelity operation of said at least one second quantum gate in said respective at least one ELU for enhanced scalability of said quantum computer;

a photon detection sub-system operatively coupled to a respective at least one of said at least first and second quantum gates via said photonic interconnect network to detect realization of said respective at least one of said at least first and second quantum gates; and

a measurement sub-system operatively coupled to said detection sub-system and said at least on first and second quantum gates to measure, upon detection of photons produced thereat, the states of qubits resulting from realization of said respective at least one of said at least first and second quantum gates, said qubit'"'"'s states being supplied to said processor for processing and subsequent output in a form of computation results

wherein said stationary matter qubits include ion qubits, further comprising a laser sub-system generating laser beams with predetermined characteristics and operatively coupled to said respective at least one ELU in a controlled manner in accordance with said quantum algorithm to realize said at least one of said at least first and second quantum gates, wherein said control sub-system is operatively coupled to said laser sub-system for shaping at least one of said laser beams to apply said optimally controlled qubit state-dependent optical forces to a respective subset of said ion qubits,

wherein said laser sub-system includes;

a first laser configured for initialization of each of said ion qubits in said respective at least one ELU,

at least one second continuous-wave laser configured to simulate Raman transitions between said ion qubits and to produce said qubit state-dependent optical forces, wherein said at least one second laser generates a first laser beam for illuminating said plurality of ion qubits in said respective at least one ELU and a second laser beam for manipulating said respective subset of adjacent ion qubits of interest, and

a third resonant laser configured to participate in said ion qubits states measurement by applying said third laser'"'"'s beam to said ion qubits states,

wherein said multiple modes of motion of said ion qubits are caused by application of at least one of said first laser'"'"'s beam and said first laser beam of said at least one second laser, and

wherein said optimal control parameters include a predetermined pulse shape of said second laser beam designed for disentanglement of said multiple modes of motion of said ion qubits housed within said respective at least one ELU.

2. The large-scale modular quantum computer architecture of claim 1, wherein said pulse shape of said second laser beam is evenly partitioned into (2N_q+1) segments wherein N_q is the number the ion qubits in said respective at least one ELU.

3. The large-scale modular quantum computer architecture of claim 2, wherein said optimal control parameters further include spectral phase and amplitude of said second laser beam, and a predetermined addressing of said ion qubits of interest.

4. The large-scale modular quantum computer architecture of claim 1, wherein in said at least one multi-dimensional quantum computational structure, a probabilistic connection is realized in a fault-tolerant fashion between said respective distant modular ELUs by generation of entangled Bell pairs of communication qubits in accordance with said quantum algorithm, each communication qubit residing in a respective on of said respective distant modular ELUs to perform measurement-based quantum computations.

5. The large-scale modular quantum computer architecture of claim 4, wherein said respective at least one ELU is provided with a plurality of communication ports for coupling with said photonic interconnect network, further comprising a time division multiplexing (TDM) mechanism applied to said communication ports of said distant ELUs to reduce entanglement generation time therebetween.

6. The large-scale modular quantum computer architecture of claim 5, further comprising a fault-tolerance sub-system operatively coupled to at least one of said at least first and second quantum gates resulting in reduction of a mismatch between the entanglement generation time for said distant ELUs and the entanglement generation time within said respective at least one ELU, said fault-tolerance sub-system being configured to apply a Steane quantum error correction code to multiple levels of concatenation of the Steane code followed by at least one of said first and second quantum gates realization.

7. The large-scale modular quantum computer architecture of claim 1, wherein said at least one multi-dimensional quantum computational structure is configured as at least one multi-dimensional hypercell composed of said multiplexed modular ELUs, said at least one hypercell having a storage capacity for quantum information substantially corresponding to the storage capacity of each of said ELUs, and the ability of deterministical entanglement with a plurality of hypercells.

8. The large-scale modular quantum computer architecture of claim 7, wherein said at least one hypercell includes:

a snowflake contoured connectivity tree with each of said ELUs positioned at a respective node of said snowflake'"'"'s connectivity tree,

wherein said plurality of ELUs in said connectivity tree are interconnected to form a root ELU, and multiple layers of ELU branches coupled to said root ELU, and forming an extended surface area with a number of external ELU'"'"'s communication ports,

wherein at least two of said snowflake hypercells are entangled at said surface areas thereof via said external communication ports thereby attaining a photons link therebetween, and

wherein a Bell pair created between said at least two hypercells is teleported to a Bell pair between the root ELUs of said hypercells.

9. The large-scale modular quantum computer architecture of claim 7, wherein said plurality of ELUs are linked through said photonic interconnect network into at least one three-dimensional (3D) hypercell,

wherein said at least one 3D hypercell includes a 3-dimensional (3D) grid with said ELUs residing on the edges and on the faces thereof, wherein for each ELU containing four qubits and linked to four neighboring ELUs, said 3D grid probabilistically creates a 4-valent 3D cluster state by probabilistically generating Bell pairs between said each and said neighboring ELUs, and

wherein, when qubits in said each of said ELUs and said neighboring ELUs form near deterministic links between different ones of said 3D cluster states, each of said 3D cluster states transforming into a fault-tolerate star-shaped graph state via measurements in said measurement sub-system.

10. The large-scale modular quantum computer architecture of claim 1, wherein said at least one ELU includes at least one chain of physical qubits, refrigerator qubits interposed between said quantum gates, and communication qubits coupled to said dynamically reconfigurable photonic interconnect network to realize said at least first quantum gate between said distant ELUs, and wherein at least one logical qubit is formed with a plurality of physical qubits residing at the same ELU or in different ELUs.

11. The large-scale modular quantum computer architecture of claim 10, wherein said refrigerator qubits are selected from different species than said physical qubits.

12. The large scale modular quantum computer architecture of claim 10, further comprising an isolating mechanism for isolating said communication qubits from said physical qubits, said isolating mechanism including shuttling of said communication qubits away from said physical qubits, and utilizing species different than said physical qubits for said communication qubits.

13. The large-scale modular quantum computer architecture of claim 1, wherein said photon detection sub-system includes an array of N_ELU/2 Bell state detectors, each comprising a first and second optic fibers and a first and a second photon counting detectors, wherein said photonic interconnect network includes an optical cross-connect (OXC) switch having N_ELU input ports coupled to respective ELUs and N_ELU output ports, an a plurality N_ELU of beam splitters coupled, at an input line thereof, to a respective one of said output ports of said OXC switch, and, at an output end thereof, to a respective one of said first and second detectors,

wherein said OXC switch is configured to establish non-blocking connectivity between said plurality of ELUs, and wherein said photonic interconnect network provides entanglement generation between said distant ELUs in a respective pair thereof with up to N_ELU/2 entanglement operations performed in parallel, wherein N_ELU is the number of ELUs in said at least one multi-dimensional computation scheme.

14. The large-scale modular quantum computer architecture of claim 1,

wherein said processor is further configured to apply entanglement swapping protocols to coordinate the entanglement generation time with communication times scaling poly-logarithmically to a communication distance, or scaling independently of said communication distance, by employing quantum buses formed by qubits interpositioned between said respective distant modular ELUs.

15. The large-scale modular quantum computer architecture of claim 1, including a quantum adder circuit configured to compute a sum of two n-bit integers and employing 6n logical qubits placed on 1.5n ELUs to compute said sum of two n-bit integers at a first concatenation level of Steane code encoding,

wherein said quantum adder circuit includes X, CNOT, and Toffoli quantum gates, wherein Toffoli quantum gate is realized between said respective distant modular ELUs, and wherein each of said X and CNOT quantum gates are realized in a respective at least one modular ELUs,

wherein each of said distant modular ELUs supports at least three logical qubits, at least seven-qubit cat state, and a sufficient number of ancilla qubits to support qubit state preparation,

wherein each of said logical qubits includes at least four ancilla qubits to provide realization of four-qubit cat states,

wherein said Toffoli gate is realized by allocating at least one fresh ELU, preparing an initial |φ

₊ state, and teleporting said at least three logical qubits in at least one distant ELU into said initial |φ

₊ state via photon interference, and freeing and using said logical qubits from other ELUs for realizing of other Toffoli gates.

16. A method of operating a large-scale modular quantum computer, comprising:

(a) inputting data in a processor unit for being computed;

(b) interconnecting, under control of said processor unit, a plurality N_ELU of modular elementary logic units (ELUs) into at least one multi-dimensional quantum computational structure through a dynamically reconfigurable photonic interconnect network, wherein each modular ELU hosts a plurality N_d of stationary matter qubits,

(c) applying predetermined laser pulses to the qubits in at last one modular ELU to set said qubits in a controlled initial state representing a computational problem to be solved,

(d) controllably shaping a first laser beam,

(e) illuminating, by said first laser beam, a predetermined subset of qubits in said at least one modular ELUs to realize at least one first quantum gate between said qubits in said at least one modular ELU in accordance with a quantum algorithm, wherein said optimally shaped first laser beam supports a disentanglement of modes of motion of said qubits in said at least one modular ELU, thereby suppressing modes crosstalk in said at least one modular ELU, and thereby obtaining a high fidelity said at least one first quantum gate,

(f) illuminating, by a second laser beam, at least one pair of said qubits housed in at least a pair of distant modular ELUs in accordance with said quantum algorithm to realize at least one second quantum gate between said quantum algorithm to realize at least one second quantum gate between said distant modular ELUs, wherein said at least first and second quantum gates are realized in accordance with a predetermined schedule between qubits residing in said modular ELUs constitution said at least one multi-dimensional computational structure,

(g) detecting the realization of a respective at least one of said at least first and second quantum gates by registering a corresponding at least one photon event heralding an entangled quantum gate between said illuminated qubits,

(h) measuring the states of said qubits of said at least first and second quantum gates upon detection of the successful realization thereof, and

(i) transmitting quantum information corresponding to said measured qubits states to said processor unit for further processing and output in a form of computational results.

17. The method of claim 16, further comprising:

constructing an arithmetic circuit, including a quantum adder circuit, through realizing and operatively interconnecting said at least first and second quantum gates as X, CNOT, and Toffoli quantum gates in accordance with said quantum algorithm under control of said processor unit.

18. The method of claim 17, wherein a dedicated photonic communication bus is executed between said distant modular ELUs, wherein each of said distant modular ELUs hosts at least three logical qubits, at least 7-qubit cat state, and wherein each of said distant modular ELUs accommodates the generation of an initial |φ

₊ state, and a number of ancilla qubits to support said initial |φ

₊ state preparation, further comprising;

preparing said Toffoli quantum gate by;

allocating a fresh one of said plurality of modular ELU,

preparing the initial |φ

₊ state in said fresh modular ELU,

generating entangled states |φ

₊ in said distant modular ELUs via photon interference,

providing at least three optical ports in a cross connect switch of said dynamically reconfigurable photonic interconnect network to connect to said distant modular ELUs substantially in parallel, and operating said at least three optical ports in a multiplexing regime,

teleporting three logical qubits from the distant modular ELU in said |φ

₊ state, thus realizing the Toffoli gate, and

freeing said three logical qubits from said distant modular ELU for availability for realization of another Toffoli gate.

19. The method of claim 16, further comprising:

linking said plurality of modular ELUs into at least one three-dimensional (3D) hypercell supported by a 3D lattice, wherein said modular ELUs are located on the faces and edges of said 3D lattice, wherein each modular ELU contains at least four qubits and is linked to four neighboring modular ELUs in said at least one 3D hypercell,

scheduling the realization of said at least first and second quantum gates in said respective at least one modular ELU and between said distant modular ELUs in said at least one 3D hypercell through;

creating 4-valent 3D cluster state by probabilistically generating Bell pairs between said distant modular ELUs via said dynamically reconfigurable photonic interconnect network,

realizing CNOT gates within each said modular ELU, and

performing local measurements of three out of said at least four qubits in said modular ELU.

20. The method of claim 16, wherein said at least one multi-dimensional quantum computational circuit is reconfigured by multiplexing said plurality of modular ELUs into at least one hypercell having a storage capacity for quantum information as each of said modular ELUs, and the ability of deterministical entanglement with a plurality of said hypercells.

21. The method of claim 20, wherein said at least one hypercell includes:

a snowflake connectivity tree with each said modular ELU positioned at a respective node of said snowflake'"'"'s connectivity tree,

wherein said plurality of modular ELUs in said connectivity tree are interconnected to form a root modular ELU, multiple layers of modular ELU branches coupled to said root modular ELU, and forming an extended surface area with a number of external modular ELU'"'"'s communication ports,

wherein at least two of said snowflake hypercells are entangled at said surface areas thereof via said external communication ports thereby attaining a photons link, and

wherein a Bell pair created between said at least two hypercells is teleported to Bell pair between the root modular ELUs of said hypercells.

22. The method of claim 21, further comprising:

forming said at least one hypercell in stages, starting with external ELUs at said surface area, and iteratively coupling underlying layers of modular ELUs to superposed layers of modular ELUs.

23. The method of claim 16, wherein said stationary matter qubits include quantum systems featuring strong optical transitions.

24. The method of claim 23, wherein said quantum systems include trapped ions, quantum dots, neutral atoms, and nitrogen-vacancy (NV) color centers in diamond.