CN101088102A - Analog processor comprising quantum devices - Google Patents

Abstract

Analog processors for solving various computational problems are provided. Such analog processors comprise a plurality of quantum devices, arranged in a lattice, together with a plurality of coupling devices. The analog processors further comprise bias control systems each configured to apply a local effective bias on a corresponding quantum device. A set of coupling devices in the plurality of coupling devices is configured to couple nearest-neighbor quantum devices in the lattice. Another set of coupling devices is configured to couple next-nearest neighbor quantum devices. The analog processors further comprise a plurality of coupling control systems each configured to tune the coupling value of a corresponding coupling device in the plurality of coupling devices to a coupling. Such quantum processors further comprise a set of readout devices each configured to measure the information from a corresponding quantum device in the plurality of quantum devices.

Description

The analog processor that comprises quantum devices

Related application

According to 35U.S.C119 (e) joint, the application requires the right of priority of No. 60/638,600, the U.S. Provisional Patent Application submitted on Dec 23rd, 2004, and this patented claim is combined in this by reference in full.According to 35U.S.C119 (e) joint, the application also requires the right of priority of No. 60/705,503, the U.S. Provisional Patent Application submitted on August 23rd, 2005, and this patented claim also is combined in this by reference in full.

Technical field

This method, article and system relate to simulation process and quantum calculation device.

Background technology

2.1 analog computation

Analog computation utilizes physical phenomenon (machinery, electric, or the like) to represent the variable in this problem to simulate the problem of being paid close attention to by use physical quantity (pressure, voltage, position, or the like), and this problem is some abstract mathematical problem or some physical problem that relates to other physical quantitys here.In its simplest mode, a simulation system (for example analog computer) is found the solution this problem by one or more input variable that is taken into a problem, they is represented as physical quantity, and deduces out its state according to physical law then.Answer to this problem is produced as a physical descriptor, and this variable can be read then.

Simulation system has two advantages.First advantage is to carry out computing in real parallel mode.Because computing usually is subjected to the restriction of physical law, do not exist anything in a part of this simulation system, to forbid a computing in the physical rules of most simulation systems, and carry out another computing in another part of this simulation system simultaneously.Second advantage is simulation system domain operation when not relating to, thereby and do not require the use clock.A plurality of simulation systems are calculated in real time, and for most physical application, this is faster than carry out same calculating on a digital machine.

Traditionally, simulation system uses some physical quantitys (for example voltage, pressure, temperature or the like) to represent a continuous variable.This causes the problem of degree of accuracy, because the accuracy of problem answers is subjected to the restriction of the accuracy of the continuous variable that can quantize.This is because simulation system usually uses physical quantity to represent a variable in the problem, and the physical quantity internality that occurring in nature is found ground is continuous.On the other hand, digital machine relates to the differentiation between possible word bit value " 0 " and " 1 ", exists the accurate status that is easy to discern for this differentiation.Simulation system also often they restricted on the problem types that can solve.For example, a corona and a compass all are original analog computers.Yet the two all is merely able to carry out a computing, is respectively according to the position calculation time of the sun and the direction in calculating magnetic field of the earth.A digital machine can use the computing machine with generic to solve this two problems by reprogramming.Simulation system is often than digital machine complexity.And the number of calculations that can carry out of simulation system often is subjected to the restriction of the degree that circuits/devices can be replicated.

Although digital machine is useful for solving a plurality of general considerationss, yet there is the answer of some problems on the digital machine of a routine, to calculate effectively.In other words, find out problem answer time not on the degree of polynomial with being in proportion of problem.Can make the problem parallelization in some cases.Yet this parallelization is often unactual on the angle of cost.Digital machine uses a kind of approach of finite state machine.Although the computational problem finite state machine approach work for numerous kinds is good, there is a bottom line in it for the complex nature of the problem that can find the solution.This be because the finite state machine approach use a kind of clock in other words timer carry out computing.Clock with the CMOS technology implementation in the prior art level has a maximum clock speed (frequency) that is about 5GHz.On the contrary, multiple simulation system does not require clock.Thereby the answer that can must go wrong with a kind of natural mode in simulation system often draws answer than the corresponding system of digital computation with faster speed far away, possibility even by exponential faster speed.

The shown practicality that goes out of digital machine be they low-power consumption, be easy to the discrete binary nature of the state of distinguishing and they can solve the ability of versatility computational problem widely.Yet a plurality of particular problems in complete (NP-complete) problem of quantum simulation, optimization, a NP difficult problem (NP-hard) and other NP are still thorny on digital machine.If can overcome the shortcoming such as the restriction of its limited accuracy of simulation system, just can easily surpass classical digital machine finding the solution on the important computational problem simulation system.

2.2 complicacy classification

The computer scientists that pay close attention to complicacy use different complicacy classification definition routinely.Other number of complexity class is always changing, because new complicacy classification is defined and existing complicacy classification is merging in the computer science progress.Be known as polynomial time (polynomial-time) (P), non-definition polynomial time (non-deterministicpolynomial-time) (NP), NP-fully the complicacy classification of (NPC) and a NP-difficult problem (NPH) all be the classification of decision problem.Decision problem has the result of binary.

Problem among the NP is the computational problem that it is existed the checking of polynomial time.In other words in order to verify that a potential Xie Buhui surpasses the polynomial time (classification P) of this problem size.Produce potential separating and to surpass polynomial time.Spend the longer time possibly for a NP-difficult problem and verify potential separating.

It is problem among the NP that problem among the NPC can be defined as, and these problems have shown compared with a known problem among the NPC and are equal or more are difficult to answer.Say that equivalently the problem among the NPC also is the problem of NPH among NP.This can be expressed as NPC=NP ∩ NPH.

If a problem is existed the reduction of a kind of polynomial time of the known problem from NPC, then this problem be exactly with NPC in a known problem of equal value or than more being difficult to answer.Reduction can be regarded a kind of extensive of mapping as.This mapping can be man-to-man function, many-to-one function or utilize prophesy or the like.The thorny property how other notion of complexity class and complicacy classification define some computational problem sees also, for example, what Garey and Johnson showed in 1979: Computer and Intractability:A Guideto the Theory of NP-Completeness, Freeman, SanFrancisco, ISBN:0716710455 (hereafter is " Garey and Johnson ").Can also consult Cormen, Leiserson, and Rivest, nineteen ninety: Introduction to Alrorithms, MIT Press, Cambridge, ISBN:0262530910.

Decision problem often has a kind of relevant optimization problem, finds the solution this relevant optimization problem with the definition right judgement.Find the solution a efficient and can cause finding the solution corresponding efficient based on the problem of optimizing based on the NP-complete problem of judging.This usually sets up any problem among the NP.The problem of often finding the solution for it is based on the problem of optimization.

2.3 quantum devices

Quantum calculation is a kind of new relatively computing method, and this computing method use quantum devices so that utilize quantum effect, such as the involving of the stack of ground state and quantum devices, and carries out some calculating faster than the digital machine of classics with this.In digital machine, information is to be stored in the word bit, and this word bit both can be that " 0 " also can be " 1 ".For example, a word bit can be represented logical zero and represent logical one with a high voltage with a low-voltage.Opposite with the word bit of digital machine, a quantum computer is information stores in the quantum bit (qubit), and this is a kind of quantum devices, and data wherein both can be " 0 " state or one state, or any stack of these states,

α|0>+β|1>. (1)

According to the term of formula (1), " 0 " state class of a digital machine is similar to a quantum bit | and 0〉ground state.Similarly, the one state of a digital machine is similar to a quantum bit | and 1〉ground state.According to formula (1), the stack of the fair a plurality of quantum bit ground state of quantum bit, wherein this quantum bit has certain probability or is to be in | and 0〉or be in | 1〉state.| α | ²Be to be in | 0〉state probability and | β | ²Be to be in | 1〉state probability, wherein | α | ²+ | β | ²=1.Obviously continuous variable α and β comprise much bigger information than a word bit in a digital machine, and the word bit in a digital machine is simple 0 or 1.The state of a quantum bit can be expressed as vector,

$\left[\begin{array}{c}\mathrm{\α}\\ \mathrm{\β}\end{array}\right].---\left(2\right)$

Although this quantum bit can be that it only can be used as and is among a kind of linear combination (perhaps stack) at a plurality of states | 0〉or | 1〉state is read out or measures.Quantum devices has represented the quantum performance, such as the involving of the stack of the quantum tunneling effect between the quantum ground state, ground state, quantum bit, coherence and show wave characteristic and particle characteristics simultaneously.In a master pattern (being also referred to as the circuit model of quantum calculation) of quantum calculation, the running of quantum door is to carry out in time domain on the quantum bit in quantum calculation device.In other words, each independent door moves a preset time cycle to carry out a kind of quantum calculation on the state on one or more quantum bit in a quantum calculation device.A plurality of doors are expressed as matrix and multiply by the state vector of moving on quantum bits.The most basic single quantum bit door is Poly matrix (Pauli matrices):

$X\≡\left[\begin{array}{cc}0& 1\\ 1& 0\end{array}\right],$ $Y\≡\left[\begin{array}{cc}0& -i\\ i& 0\end{array}\right],$ $Z\≡\left[\begin{array}{cc}1& 0\\ 0& -1\end{array}\right].---\left(3\right)$

Other single quantum bit doors comprise Hadamard door, phase gate, and π/8.For example, referring to Nielson and Chuang, 2000, Quantum Computation andQuantum Information, Cambridge University Press, Cambridge, pp.174-177.

Two quantum bits that are coupled together are also followed stack:

α ₀₀|00>+α ₀₁|01>+α ₁₀|10>+α ₁₁|11>. (4)

The state of two quantum bit systems is represented by a four vector, and the computing of two quantum bit doors is represented by 4 * 4 matrixes.Therefore n quantum bit system is by one 2 of continuous variable ⁿVector is represented.A subclass of basic simple gate computing, for example those subclass represented in (3) and the computing of one or more pair quantum bit door have formed and it is said a general door set of quantum computing.A universal set of quantum computing is the set that allows any quantum computing of all possible quantum computing.

2.4 the requirement of quantum computing

In general, a quantum bit is a kind of physical arrangement of abundant definition, this physical arrangement (i) has a plurality of quantum states, (ii) can be coherently from its environment isolate and (iii) allow with two or more quantum states that this quantum bit is associated between the quantum tunneling effect.For example referring to Mooji et al., 1999, Science 285, p.1036 (hereinafter are called " Mooji "), and this article is combined in this by reference in full.A summary for the current physical system that therefrom can form quantum bit is consulted Braunstein and Lo (eds.), and 2001, Scalable Quantum Computers, Wiley-VCH, Berlin (hereinafter referred to as " Braunstein and Lo ").

For showing as a kind of quantum bit, the system that makes a physics must satisfy several requirements.Referring to DiVincenzo in Braunstein and Lo, Chapter 1, chapter 1.These requirements comprise that this physical system (quantum bit) need be prolongable.In other words, it must be with the quantum bit of a kind of relevant mode in conjunction with a fair amount.What be associated with expansibility is the laddering that needs to eliminate quantum bit.Quantum bit can be used in the quantum calculation computing that also requirement can make quantum bit carry out initialization, control and be coupled.The control of a quantum bit comprises carries out single quantum bit computing and the computing on two or more quantum bits.In order to support quantum calculation, the set of this computing need be a universal set.A plurality of set is general, for example consults Barenco et al, 1995, and PhysicalReview A 52, p.3457, this article is combined in this by reference in full.Another requirement to quantum calculation is to need to measure the state of this quantum bit so that calculate running and information extraction.Circuit model for quantum calculation has developed these requirements, and can relax requirement to other models.

2.5 superconduction quantum bit

Several quantum calculation hardware suggestions have been proposed.In these hardware suggestions, it seems that the most prolongable physical system be those superconducting structures.Superconductor is the material that does not have resistance under critical current, magnetic field and temperature.The Josephson knot is the example of this structure.

Two kinds of main means that realize the superconduction quantum bit are arranged.Means are corresponding to the restriction at the electric charge (electric charge quantum bit) of abundant definition.Another means are corresponding to the restriction at the phase place (phase place/quanta of energy position) of abundant definition.Phase place is relevant variable with electric charge, and according to basic quantum principle, they are standard conjugation each other.The division of this two classes device both had been set forth in Makhlin et al., and 2001, Reviews of Modern Physics 73, among the pp.357-400 (hereinafter referred to as " Makhlin "), this article is combined in this by reference in full.Abundant known device in the field under the superconduction quantum bit comprises, such as Josephson knot quantum bit.For example consult Barone and Patern ò, 1982, Physics and Applications ofthe Josephson Effect, John Wiley and Sons, New York; Martinis et al., 2002, Physical Review Letters 89,117901, this article are combined in this by reference in full; With Han et al., 2001, Science 293, and p.1457, this article is combined in this by reference in full.

2.5.1 flux qubit

One type flux qubit is the steady current quantum bit.Consult Mooji andOrlando et al., 1999, Physical Review B 60,15398-15413 (hereinafter being called " Orlando "), this article are combined in this by reference in full.The superconduction phase qubits is fully known and has showed long coherence time.For example, referring to Orlando andIl ' ichev etal., 2003, Physical Review Letters 91,097906 (hereinafter referred to as " Il ' ichev "), this article is combined in this by reference in full.The phase qubits of some other types comprise have more than or be less than the superconducting circuit of three Josephson knot.For example consult G.Blatter et al., 2001, Physical Review B, 63,174511 and Friedmanet al., 2000, Nature 406,43 (hereinafter referred to as " Friedman 2000 "), every piece of article all is combined in this by reference in full.More details for flux qubit, see also following United States Patent (USP): be entitled as " Resonant controlled qubit system " 6,960, No. 780, be entitled as " Resonant controlled qubit system " 6,897, No. 468, be entitled as " Multi-junction phase qubit " 6,784, No. 451, be entitled as " Sub-fluxquantum generator " 6,885, No. 325, be entitled as " Quantum phase-chargecoupled device " 6,670, No. 630, be entitled as " Finger squid qubit device " 6,822, No. 255, be entitled as " Superconducting low inductance qubit " 6,979, No. 836; The U.S. Patent application of following discloses: be entitled as " Extra-substratecontrol system " 2004-0140537 number, be entitled as " Methods for single qubitgate teleportation " 2004-0119061 number, be entitled as " 2004-0016918 number of System and method forcontrolling superconducting qubits; be entitled as " Encoding and error suppression for superconducting quantumcomputers " 2004-0000666 number; be entitled as " Quantum phase-charge coupleddevice " 2003-0173498 number; be entitled as " Quantum computing integrateddevelopment environment " 2003-0169041 number; be entitled as " Quantumcomputing integrated development environment " 2003-0121028 number; be entitled as " Trilayer heterostructure iunctions " 2003-0107033 number, with be entitled as " Quantum bit with a multi-terminal junction and loop with a phaseshift " 2002-0121636 number, they each all be combined in this in full by reference.

Figure 1A illustrates a superconduction phase qubits 100.Phase qubits 100 comprises a loop 103 by the superconductor of Josephson knot 101-1,102-2 and 103-3 interruption.The Josephson knot typically uses standard manufacture technology to form, relate generally to material deposition and photoetching stage.For example, consult Madou, 2002, Fundamentals ofMicrofabrication, Second Edition, CRC Press, Van Zant, 2000, Microchip Fabrication, Fourth Edition, McGraw-Hill, New York, Levinson, 2001, Principles of Lithography, The International Societyfor Optical Engineering, Bellingham Washington, and Choudhury, 1997, Handbook of Microlithography, Micromachining and MicrofabricationVolume 1:Microlithography, The International Society for OpticalEngineering, Bellingham Washington..Make the method for Josephson knot, for example be illustrated in Ramos et al., 2001, IEEE Transactions on AppliedSuperconductivity 11, p.998.Common matrix for example comprises silicon, monox, perhaps sapphire.Josephson knot 101 can also comprise insulating material, such as aluminium oxide.The example that can be used for forming the superconductor of superconducting circuit 103 is aluminium and niobium.Josephson knot 101 has the size of scope from 10 nanometers (nm) to about 10 microns (μ m).One or more Josephson knot 101 have with phase qubits 100 in other Josephson tie 101 different parameters, for example size of knot, knot surface area, Josephson energy or charging flux.Difference in phase qubits 100 between any two Josephson knot 101 is characterized by a coefficient of representing with α, and this coefficient typically scope has the knot of equivalent parameters in this α=1 representative between about 0.5 to about 1.3.In some example, the condition α of a pair of Josephson knot is the ratio of the critical current of corresponding Josephson knot in this phase qubits.The critical current of Josephson knot is that to make this knot no longer be the electric current of this knot of flowing through of superconduction.In other words, in this superconduction that this knot is below critical current, and at this more than critical current, this knot is not superconduction.Thereby for example, the condition α of knot 101-10 and 101-2 is defined as the critical current of knot 101-1 and the ratio of the critical current of knot 101-2.

Referring to Figure 1A, be coupled to bias source 110 irritability phase qubits 100.Bias source 110 is used for passing a magnetic flux through phase qubits 100 _xSo that the control to the state of this phase qubits to be provided.Phase qubits 100 typically by a scope at about 0.2 φ ₀To about 0.8 φ ₀Between magnetic flux biasing φ _xCarry out computing, here φ ₀It is the flux quantum.

Phase qubits 100 has a kind of two-dimentional potential energy with respect to the phase place of striding Josephson knot 101 of simplification.Typically make phase qubits φ 100 biasings with a magnetic flux x, thereby family curve that should two dimension potential energy comprises the zone of local energy-minimum, at this, this this locality energy-minimum is to be separated from each other and to be separated by the zone of big energy barrier with other by little energy barrier.This potential energy is a kind of pair of trap gesture 150 (Figure 1B), and this pair trap gesture comprises a left trap 160-0 and a right trap 160-1, represents clockwise 102-1 and 102-1 round-robin supercurrent counterclockwise in the phase qubits of Figure 1A respectively.When applying about 0.5 φ ₀The flux biasing time can form a two trap gesture 150.

When two trap 160-0 and 160-1 are degraded or during near sharpening, mean that they are in identical energy gesture or approaching identical energy gesture, as as shown in Figure 1B, at this moment the quantum state of phase qubits 100 becomes a coherence stack of phase place or basic status, and this device can be used as a kind of phase qubits and carries out computing.Be in or be called the calculating operating point of phase qubits 100 at this near the point of sharpening.In the process of the calculating of phase qubits 100 operation, can use controllable quantum effect according to the rule treatments of quantum calculation with this phase state canned data.Because the quantum information of storing and handling in this phase qubits is with the phase bits basis, it is insensitive to the noise based on electric charge.Il`ichev etc. (Illichev) have used a kind of three Josephson knot flux qubit that is coupled to high-quality tank circuit (tankcircuit) to carry out the continuous observation of Rabi vibration.

The quantum calculation model of standard has a plurality of problems, and this makes it to become a kind of challenge heroic undertaking of science and technology.Quantum calculation relates to coherently handles quantum information.This requirement has decoherencing the time of abundant length in quantum bit, will avoid noise and error simultaneously.Laddering makes that the master pattern quantum calculation of time domain door level is very difficult.Therefore wish to drive quantum effect, for example irrelevant tunnel effect is found the solution useful problem with this, thereby is overcome the challenge of master pattern quantum calculation.

Summary of the invention

(i) this method, article and system aspect provides a kind of computing system that comprises a kind of simulation (quantum) processor.This quantum processor comprises a plurality of quantum devices of a plurality of nodes that form a dot matrix, and these quantum devices have first and second base states and comprise the loop of the superconductor that is interrupted by the Josephson knot.This quantum processor further comprises a plurality of devices that are coupled, and this is coupled device and in the mode of a nearest-neighbors and/or inferior nearest-neighbors these quantum devices is coupled together.

(ii) another aspect of this method, article and system provides a kind of method of the result who uses a computational problem of quantum processor definition, and this quantum processor comprises a plurality of quantum devices and a plurality ofly is coupled device to what this quantum devices was coupled together.This method comprises by a state setting each quantum devices in these a plurality of quantum devices and sets the intensity that is coupled that these a plurality of each that are coupled in the device are coupled device makes this quantum processor be initialised to a kind of original state, makes this quantum processor to calculate out to approach the end-state of ground state of a nature of this computational problem; Thereby and read the result that an end-state defines this computational problem in one or more quantum devices from these a plurality of quantum devices.

(iii) another aspect of this method, article and system provides a kind of computer system that comprises a central processor unit and be coupled to a storer of this central processor unit, is used to define the result of a computational problem.This storer comprises a user interface module, comprising the instruction that is used to define this computational problem, a mapper module, comprising instruction and analog processor NIM of a kind of mapping that is used to produce this computational problem.This analog processor NIM comprises and this mapping is transferred to the instruction of an analog processor and is used to respond this mapping receives a result from this analog processor instruction.This analog processor comprises a plurality of quantum devices and a plurality ofly is coupled device and this mapping comprises the initialization value of each quantum devices in these a plurality of quantum devices and the initialization value that this a plurality of each that are coupled in the device are coupled device.This is coupled device this quantum devices is coupled to device their nearest-neighbors and/or next nearest-neighbors.

(iv) another aspect of this method, article and system provides a kind of computer program that same digital machine is used in combination that is used for.This computer program comprises a kind of computer readable storage medium and a computer program mechanism of implanting therein, and this computer program mechanism comprises a user interface module, comprising the instruction that is used to limit a computational problem, a mapper module, instruction comprising a mapping that is used to produce this computational problem, with an analog processor NIM, be used for from a result's of this analog processor reception instruction comprising the instruction that this mapping is transferred to an analog processor and in response to this mapping.This analog processor comprises a plurality of quantum devices and a plurality of device that is coupled, and this mapping comprises the initialization value of each quantum devices in these a plurality of quantum devices and the initialization value that this a plurality of each that are coupled in the device are coupled device, and this is coupled device this quantum devices is coupled to their nearest-neighbors and/or nearest-neighbors secondly.

(v) another aspect of this method, article and system provides a kind of quantum processor.This quantum processor comprises that a plurality of quantum devices that are arranged to a dot matrix, more than first are coupled more than device and second and are coupled device.First be coupled in the device one and be coupled device and will in this dot matrix, be coupled by the most contiguous one first quantum devices and one second quantum devices at this, will in this dot matrix, be coupled by time the most contiguous one the 3rd quantum devices and one the 4th quantum devices and be coupled device these more than second of being coupled in the device.

(vi) another aspect of this method, article and system provides a kind of quantum processor that comprises a plurality of quantum devices, also have more than first to be coupled device, more than second and to be coupled device, to be coupled to a readout device of at least one quantum devices and to be coupled to a local bias unit of at least one quantum devices.These a plurality of quantum devices and this more than first individual device that is coupled form a planar rectangular array, this two-way array has a diagonal line, and these more than first at least one that are coupled in the device are coupled device and are coupled intensity with one one first quantum devices and second quantum devices are coupled, this is coupled that intensity has a minimum negative be coupled an intensity and a maximum just be coupled the interior value of scope between the intensity.Being coupled device these more than second at least one that are coupled in the device is coupled the intensity that is coupled that intensity and one zero is coupled a kind of value in the scope between the intensity a minimum negative one the 3rd quantum devices and one the 4th quantum devices along the diagonal line arrangement of this array is coupled to have.

(vii) another aspect of this method, article and system provides a kind of computing system that comprises a kind of quantum processor.This quantum processor comprises a plurality of qubit devices and a plurality of device that is coupled of a plurality of nodes that form a dot matrix.Be coupled device these a plurality of first of being coupled in the device one first qubit devices is coupled to one second qubit devices, this first qubit devices and this second qubit devices are to be in a kind of nearest-neighbors or in the configuration of a kind of nearest-neighbors.

(viii) another aspect of this method, article and system provides a kind of quantum processor, comprising a plurality of qubit devices that are arranged to a kind of dot matrix, a kind of more than first individual device and a kind of more than the second individual devices that are coupled of being coupled.Being coupled device these more than first first of being coupled in the device is coupled one first qubit devices and one second qubit devices, this first qubit devices and second qubit devices are configured to the nearest-neighbors in this dot matrix, and be coupled device these more than second first of being coupled in the device one the 3rd qubit devices and one the 4th qubit devices are coupled, the 3rd qubit devices and the 4th qubit devices are configured to the inferior nearest-neighbors in this dot matrix.

(ix) another aspect of this method, article and system provides a kind of method of the result who uses a computational problem of a quantum processor definition.This quantum processor comprises a plurality of quantum devices and a plurality of device that is coupled, and each is coupled device and is coupled a pair of quantum devices.This method comprises by state setting each quantum devices and sets each and is coupled of device and is coupled intensity and makes this quantum processor be initialized to a kind of original state, allow this quantum processor to calculate to approach a kind of end-state of ground state of a nature of this computational problem, thereby an end-state of reading at least one quantum devices defines the result of this computational problem, produce a kind of carrier wave, wherein embodied the result's who comprises this computational problem a data-signal.

(x) this method, another aspect of article and system provides a kind of computer system, comprising being used to import a P that will find the solution, NP, the device of the computational problem of a NP-difficult problem and NP-complete problem, be used for this computational problem is mapped to a quantum processor, be coupled comprising qubit devices and being used to nearest neighbor and time nearest neighbor qubit devices be coupled device, use this quantum processor to obtain the device of separating of this computational problem, be used to export the device of separating of this computational problem, and be used for this is separated as a device that data-signal transmits that is embodied in a carrier wave.

(xi) another aspect of this method, article and system provides a kind of digital signal on the carrier wave that is embodied in, comprising the respective value of each node in a plurality of nodes.These a plurality of nodes are at least two nodes in the node dot matrix in a quantum processor.Each node in this node dot matrix is a quantum devices.The value of at least one node in a plurality of nodes individually or is collectively represented one by be mapped in later at least one part of this dot matrix a time separating by a computational problem calculating this quantum processor and solved by a figure representing this computational problem.。

(xii) another aspect of this method, article and system provides a kind of digital signal on the carrier wave that is embodied in, the answer of a computational problem that defines comprising the value of each node in a plurality of nodes is assessed.These a plurality of nodes are at least two nodes in the dot matrix of a node in the quantum processor, and each node all is a quantum devices in the dot matrix of this node.The value of at least one node in these a plurality of nodes is to represent a figure of this computational problem to be mapped in behind later at least one part of this dot matrix this quantum processor of temporal calculation defined.

(xiii) another aspect of this method, article and system provides a kind of digital signal on the carrier wave that is embodied in, comprising the figure of a computational problem will finding the solution by this quantum processor, wherein this quantum processor comprises the dot matrix of a quantum devices.The figure of the computational problem of finding the solution comprises a plurality of nodes, and for each node corresponding of this a plurality of nodes, comprise between initial value of respective nodes and the corresponding node in these a plurality of nodes and another node one corresponding be coupled constant.The figure of the computational problem that this will be found the solution is configured to and can be mapped to it on this dot matrix of this quantum processor.

(xiv) another aspect of this method, article and system provides a kind of digital signal on the carrier wave that is embodied in, comprising a computational problem will finding the solution by a quantum processor.This quantum processor comprises the dot matrix of a quantum devices.This computational problem of finding the solution is converted into a kind of figure, this figure comprises a plurality of nodes, and, for each node corresponding in these a plurality of nodes, comprise the constant that is coupled of initial value of respective nodes and a node corresponding in these a plurality of nodes and a correspondence between another node.The figure of the computational problem that this will be found the solution is configured to and can be mapped to it on this dot matrix of this quantum processor.

(xv) another aspect of this method, article and system provides a kind of graphical user interface, and this graphical user interface is to be used for that a computational problem is obtained one to separate and comprise one first viewing area and one second viewing area.One of a corresponding the value when indication of this first viewing area has received each node that comprises in a plurality of nodes is embodied in a digital signal on the carrier wave.These a plurality of nodes are at least two nodes in the dot matrix of a node in a quantum processor, and each node in this node dot matrix is a quantum devices.A value of at least one node in these a plurality of nodes individually or is collectively represented by a figure representing this computational problem and has been mapped to later at least one part of this dot matrix a time separating by this computational problem of calculating this quantum processor and having solved.

(xvi) another aspect of this method, article and system provides a kind of graphical user interface, and this graphical user interface is used for that a computational problem is obtained one and separates and comprise one first viewing area and one second viewing area.When the indication of this first viewing area has received the digital signal that is embodied on the carrier wave of the answer of this computational problem.Answer to this computational problem is to define by the value of assessing at least one node in a plurality of nodes.These a plurality of nodes are at least two nodes in the dot matrix of a node in a quantum processor, and each node in the dot matrix of this node is a quantum devices.Be mapped in later at least one part at this dot matrix time at a figure representing this computational problem and determined the value of at least one node in these a plurality of nodes by this quantum processor of calculation later on.This second viewing area shows separating this computational problem.

(xvii) another aspect of this method, article and system provides a kind of graphical user interface, and this graphical user interface is to be used for that a computational problem is obtained one to separate and comprise one first viewing area and one second viewing area.When the indication of this first viewing area has produced the digital signal that is embodied on the carrier wave that comprises the computational problem that will find the solution by a quantum processor.This quantum processor comprises the dot matrix of a quantum devices.This computational problem of finding the solution comprise a plurality of nodes and, for each node corresponding in these a plurality of nodes, corresponding one between the respective nodes that comprises an initial value being used for respective nodes and these a plurality of nodes and another node is coupled constant.This computational problem that will find the solution is configured to and can be mapped to it on this dot matrix of this quantum processor.This second viewing area shows separating it after receiving this computational problem.

(xviii) another aspect of this method, article and system provides a kind of computing system.This computing system comprises a local computer, remote computer and a long-range quantum processor of communicating by letter with this remote computer.This quantum processor comprises a plurality of quantum devices, wherein each quantum devices in these a plurality of quantum devices all is the node of a dot matrix, and wherein one first quantum devices in these a plurality of quantum devices has one first base state and one second base state.This quantum processor further comprises a plurality of devices that are coupled, first quantum devices in a plurality of quantum devices is coupled to one second quantum devices in these a plurality of quantum devices this to be coupled device these a plurality of first of being coupled in the device, and wherein the configuration of first quantum devices of this in this dot matrix and second quantum devices is to be selected from by a nearest neighbor configuration and one nearest neighbor to dispose the group that is constituted.This local computer is configured to send a computational problem to this remote computer.This remote computer is configured to the answer of this local computer transmission to this computational problem.

(xix) another aspect of this method, article and system provides a kind of computer system, and this computer system is used to define a result of a computational problem.This computer system comprises a local computer, a remote computer and an analog processor.This local computer comprises a CPU (central processing unit) and a storer that is coupled to this CPU (central processing unit).User interface module of the memory stores of this local computer, comprising the instruction that is used to define this computational problem, a mapper module, instruction comprising a mapping that is used to produce this computational problem, with a transmission module, comprising the instruction that sends this mapping to this remote computer.This remote computer comprises a CPU (central processing unit) and a storer that is coupled to this CPU (central processing unit).The memory stores of this remote computer one receive module, comprising instruction and an analog processor NIM of being used for receiving this mapping, comprising the instruction that sends this mapping to this analog processor from this local computer.This analog processor comprises a plurality of quantum devices and a plurality of device that is coupled.This mapping comprises the initialization value of at least one quantum devices that is used for these a plurality of quantum devices and is used for the initialization value that these a plurality of at least one that are coupled device are coupled device.These a plurality of of being coupled in the device are coupled device the relevant quantum devices of a correspondence in these a plurality of quantum devices are coupled on nearest-neighbors of this relevant quantum devices and in time nearest-neighbors at least one.

(xx) another aspect of this method, article and system provides a kind of computer system, and this computer system is used to define a result of a computational problem.This computer system comprises a local computer, a remote computer and an analog processor.This local computer comprises a CPU (central processing unit) and a storer that is coupled to this CPU (central processing unit).The storer of this local computer comprises instruction and transmission module that is used to limit this computational problem, comprising the instruction that is used for sending to this remote computer this computational problem.This remote computer comprises a CPU (central processing unit) and a storer that is coupled to this CPU (central processing unit).The memory stores of this remote computer one receive module, comprising the instruction that receives this computational problem from this local computer.A mapper module is comprising instruction and analog processor NIM of a mapping that is used to produce this computational problem, comprising the instruction that is used for sending to this analog processor this mapping.This analog processor comprises a plurality of quantum devices and a plurality of device that is coupled.This mapping comprises the initialization value of at least one quantum devices that is used for these a plurality of quantum devices and is used for the initialization value that these a plurality of at least one that are coupled device are coupled device, and wherein this a plurality of of being coupled in the device are coupled device the relevant quantum devices of a correspondence in these a plurality of quantum devices is coupled on a nearest-neighbors of this relevant quantum devices and in the inferior nearest-neighbors at least one.

Description of drawings

Figure 1A and 1B illustrate the two trap potential energy curves according to a flux qubit and a correspondence of prior art.

Fig. 2 A illustrates the dot matrix according to an embodiment of this method, article and system, and the quadrature that has between the node is coupled.

Fig. 2 B illustrates a dot matrix according to an embodiment of this method, article and system, and the quadrature and the diagonal angle that have between the node are coupled.

Fig. 2 C illustrates the dot matrix according to another embodiment of this method, article and system.

The embodiment that Fig. 2 D illustrates according to this method, article and system has rotated 45 ° to the dot matrix of Fig. 2 B.

Fig. 3 A and 3B illustrate an embodiment of this method, article and system, are used for a planimetric map of five nodes is mapped to the corresponding analogue body based on dot matrix.

Fig. 4 A and 4B illustrate an embodiment of this method, article and system, are used for a planimetric map with six nodes and are mapped to the analogue body based on dot matrix with correspondence that inferior nearest neighbor is coupled.

Fig. 5 illustrates an embodiment of this method, article and system, is used to make multiplely be coupled device and node is equivalent to a single device that is coupled.

Fig. 6 A and 6B illustrate an embodiment of this method, article and system, in order to the planar graph of analogue body mapping based on dot matrix to correspondence.

Fig. 7 illustrates according to first group five of prior art figure K completely ₁To K ₅

Fig. 8 illustrates a K according to prior art _3,3Two-part figure.

Fig. 9 A and 9B illustrate an embodiment of this method, article and system, in order to the nonplanar figure of analogue body mapping based on dot matrix with correspondence that inferior nearest neighbor is coupled.

Figure 10 A and 10B illustrate an embodiment of this method, article and system, are used for to the nonplanar graph of analogue body mapping based on dot matrix with correspondence that inferior nearest neighbor is coupled.

Figure 11 illustrates the system that an embodiment according to this method, article and system carries out computing.

Figure 12 A and 12B illustrate an embodiment of this method, article and system, in order to shine upon the figure based on dot matrix to an integrated circuit.

Figure 13 A and 13B illustrate another embodiment of this method, article and system, in order to shine upon the figure based on dot matrix to an integrated circuit.

Figure 14 A and 14B illustrate another embodiment of this method, article and system, in order to shine upon the figure based on dot matrix to an integrated circuit.

Figure 15 illustrates the photo according to the set of four quantum devices that are coupled mutually of an embodiment of this method, article and system.

Figure 16 illustrates a kind of layout according to an analog processor of an embodiment of this method, article and system.

Figure 17 A and 17B illustrate the several embodiments of this method, article and system, are used to control a kind of pair of trap potential energy.

Figure 18 illustrates a steady current quantum bit according to prior art.

In these accompanying drawings, identical label is represented similar element or action.Each size of component and relative position in the accompanying drawing there is no need to draw in proportion.For example, the shape and the angle of various elements are not drawn in proportion, and in these elements certain some amplified arbitrarily and ranking with the raising legibility.In addition, the given shape of these elements that drawn not is the information that is intended to pass on the true form of any relevant particular element, and just discerns in the drawings for convenience and choose.And although accompanying drawing can illustrate specific layout, will appreciate that in design, layout and manufacturing of one of ordinary skill in the art can change, thus shown in layout never constitute restriction to the layout of this method, article and system.

Embodiment

In the following description, narrated some specific details to provide comprehensive understanding to various embodiments of the present invention.Yet one of ordinary skill in the art will appreciate that implementing the present invention can not have these details.Under other situation, the abundant known structure that is associated with analog processor, such as quantum devices, be coupled device and comprise microprocessor and the control system of driving circuit is not shown specifically or illustrates to avoid the unnecessarily explanation of indistinct embodiments of the present invention.Unless context has requirement in addition, in whole instructions and appended claims, " comprising " speech and various displacement thereof should be interpreted as implication a kind of opening, containing property, are the meaning of " comprise, yet be not limited to " in other words." a kind of embodiment ", " executing mode for one " or " a kind of replacement scheme ", " replacement scheme " mentioned in the whole text at this instructions refer to such an extent that be that specific feature, structure or a characteristic of being narrated is included at least one embodiment of the present invention.Thereby occurring such statement in the whole text everywhere at this instructions is not to refer to same embodiment.And then this special characteristic, structure or characteristic can be combined in one or more embodiment in any suitable manner.Subhead herein is for convenience rather than explains the category and the meaning of claimed invention.

According to this method, article and system specialization analog processor.In some embodiments, this analog processor comprises a plurality of quantum devices and a plurality of a plurality of devices that are coupled that this quantum devices is coupled together with the dot matrix arrangement.In some embodiments, this is coupled the device that device is coupled to the indivedual quantum devices in these a plurality of quantum devices its nearest neighbor and/or next nearest neighbor.In some embodiments, this analog processor can approach separating of some problems of falling into NP (non-definition polynomial time) problem category.

This NP class problem is those problems that can verify by polynomial time with a non-definition Turing machine.The example of NP classification problem comprises, yet (ISG) problem, maximum independent set close that (MaximumIndependent Set), Clique (Max Clique), maximum cutting (Max Cut), summit cover (Vertex Cover), flow sales force (Traveling Salesperson) (TSP) problem, k-SAT, integer linear programming and seek the ground state of a tunnel effect spin glass that do not setover, non-to be not limited to Yi Xin spin glass (Ising Spin Glass).These problems can the surface on a figure, wherein their projections for constituting with summit and the edge that is associated with this summit.On the whole, each in these summits and the edge can have different values or weight, and this makes different characteristics is arranged aspect the relation of this figure between different summits.

A computational problem can finding the solution with an analog processor is that maximum independent set closes problem.Garey is as follows relevant independent set problem definition with Johnston:

Example: figure G=(V, E), positive integer K≤| V|.

Whether problem: G comprises that one or more sizes are the independent set of K, promptly whether has the subclass of a V,

Possess

Do not link together by an edge among the E so that do not have two summits among the V`?

What wherein should be emphasized is to show that this maximum independent set closes and be called a difference of another problem of (Clique), and this is in following explanation.When expanding this definition, the figure of direction-free edge weighting of a positive integer K of number of considering to have the set at the set on a summit and an edge and being less than or equal to the summit of this figure.This independence set problem that is expressed as a computational problem is hereinafter asked the subclass that whether a size is arranged is the summit of K and make do not have two summits to be connected by an edge of this figure in this subclass.This problem exist a plurality of other distortion and comprise optimization problem based on this computational problem.A kind of example of optimization problem is the independent set of marking the peaked figure of the K that does not go out to produce.This is called as maximum independent set and closes.

On mathematics, solved independent set and just can solve another problem that is called as group.This problem is searched this group in a figure.A group is the set on all interconnective summits.A given figure and a positive integer K, whether the problem that is proposed in this group for existing interconnective all K summit (these summits say that also becoming is to be " neighbours " mutually).Similar to independent set problem, this clique problem can convert an optimization problem to.The calculating of group is worked in economy and cryptography.At figure G ₁=(V finds the solution an independent set on E) and is equivalent at G ₁' benefit G ₂=(V, (this edge for example, for by all summits that the edge connected among the E, is removed, at G by the group of finding the solution on the V * V)-E) ₂The middle insertion makes at G ₁In the edge that couples together, unconnected summit.Garey and Johnston are defined as group:

Example: figure G=(V, E), positive integer K≤| V|.

Whether problem: G comprises that one or more sizes are the group of K, and whether the subclass of a V is promptly arranged, Possess

So that per two summits are all linked together by an edge in E in V`?

At this, what should be emphasized is to show this group and the independent difference of gathering of setting forth above.Can also show that how relevant with vertex cover problem group is.Equally, all problems in NP-complete problem can be brief mutually in polynomial time, makes the device that can find the solution a np complete problem effectively might be used to find the solution other np complete problem equally.

(V, E), (V, maximum independent set E) close the maximal subset that M is V to G=, and wherein any one all do not have by an edge among the E and connect with being connected each figure G=that the set of the edge E on summit is formed for the set by a summit V.A maximum independent set closes M  V and can define by following object is minimized:

$E(x_1,...,x_N)=-\underset{i\∈V}{\mathrm{\Σ}}{x}_i+\mathrm{\λ}\underset{(i,j)\∈E}{\mathrm{\Σ}}{x}_i{x}_j,---\left(5\right)$

In following formula, N is the quantity on the summit among the V, each summit of i mark, (i j) is marked at edge between i among the E and the j fixed point, and x or 0 or 1.If node i is in M, this indieating variable x _iEqual 1, otherwise it equals 0.In the formula (5) first support big set M, and second can be regarded as a kind of compensation, it has strengthened in M not the restriction by interconnective summit, an edge.Factor λ plays a kind of glug Lang Ri and takes advantage of item and this compensation term of weighting.For enough big λ, we can be sure of to satisfy this restriction.In some examples, glug Lang Ri takes advantage of a λ to equal 2.

Each summit among the figure G can be by the physics spin s of value for-l and+1 _iRepresentative.Yet,, need a kind of x in order to accomplish this point _iTo spin s _iMapping.The summit that in this figure G, exists be defined as having spin+1 and in maximum independent set close separate among the M the node among the non-existent G be defined as having spin-1.This mapping is defined as follows:

s _i＝2x _i-1 (6)

Formula (6) substitution formula (5) is drawn following energy function

$E({\mathrm{<figure-callout\; id="1"\; label="s"\; filenames="A20058004434800811.png,A20058004434800821.png"\; state="\{\{state\}\}">s</figure-callout>}}_1,...,s_N)=-\frac{1}{2}\left|N\right|+\frac{\mathrm{\&lambda;}}{4}\left|E\right|-\frac{1}{2}\underset{i\&Element;V}{\mathrm{\&Sigma;}}{s}_i(1-\frac{\mathrm{\&lambda;}}{2}{d}_i)+\frac{\mathrm{\&lambda;}}{4}\underset{(i,j)\&Element;E}{\mathrm{\&Sigma;}}{s}_i{s}_j---\left(7\right),$

Wherein N is the sum on summit in G, and E is the sum at edge in G, and d _iIt is the sum that is connected to the edge of summit i.This maximum independent set closes separating of problem and can draw by minimizing formula (7).

Another example of NP class problem is Yi Xin spin glass (ISG) model, and this model definition is:

$E({\mathrm{<figure-callout\; id="1"\; label="s"\; filenames="A20058004434800811.png,A20058004434800821.png"\; state="\{\{state\}\}">s</figure-callout>}}_1,...,s_N)=-\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{h}_i{s}_i+\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j>i}{\overset{N}{\mathrm{\&Sigma;}}}{J}_{\mathrm{ij}}{s}_i{s}_j,---\left(8\right)$

S wherein ₁To s _NBe the value of respective nodes s, J _IjRepresentative is at s _iAnd s _iThe value that is coupled between the node, and h _iRepresentative is at corresponding node n _iOn biasing.Close separating of problem in order to draw maximum independent set, formula (8) is restricted to: if having an edge then this is coupled (J between node i and j _Ij) just have+λ/4 value, and if do not have the edge between node i and the j then this is coupled (J _Ij) just have 0 value, and node bias h _iHave+value of a, wherein should be from formula (8) definition a

An example by a NP class problem of a figure representative is sales force (TSP) problem that flows.In this TSP problem, different cities is represented by the summit, and the road between these cities is represented by the edge.Separating of any special case of TSP all is just to pass by all these cities shortest paths once.

This TSP problem provides the good explanation of limitation of the digital machine of prior art level.In this TSP problem, personnel are sold in distinguished personages' untie-sell must visit N city once and be only once, return to origin when End Of Tour.The best route that determining of must making will be taked.At this, " the best " depends on given priority condition, but the minimizing of briefly best total distance that can be meant travelling.Say that more realistically " the best " can refer to the minimizing of certain combination of flight time and cost.Say that with physical term what pursued is that the ground state of the system of a complexity is separated or " minimum value ".In other words, this TSP problem is sought minimum energy configuration (stroke of minimum energy in other words in this case).The number of possible stroke depends on the number in the city of existence.For N city, comprise this sales force's base camp, (N-1) arranged! Each city of individual visiting is once possible path only: first city N-1 kind is selected, and next city N-2 kind is selected, and the rest may be inferred.For N=10 city, this does not also include too bad: have only 362,880 kinds of selections.Allow in these strokes of digital computation computer each cost and determine the minimum perhaps effort too not very of the cost of which stroke.This technology is called " power of enforcement " " limit search method " in other words conj.or perhaps.Yet along with the growth of independent variable N, this factorial function increases very fastly.In fact the factorial growth will be faster than exponential growth.For N=20, N! ≈ 2 * 10 ¹⁸For drawing very much the large-scale parallel digital computer of the speed computing of bat with 100, calculate the problem of this size and also want a few hours.For N=40, N! ≈ 8 * 10 ⁴⁷, use and take the existing digital computation chance of limit search method method then can not solve this problem.Comprise that a plurality of quantum devices and an a plurality of analog processor that is coupled device can be used to the problems referred to above are minimized.

5.1 mapping

In some embodiment of this method, article and system, user utilizes a picture specification (a for example set at the summit set and edge) to define a problem, NP class problem for example, and an interface computer is handled this input with the mapping of definition to a dot matrix then.At this, dot matrix is made up of a quantum devices and the set that is coupled and can is a kind of grid.Under the situation of Shi Yonging, a dot matrix is the periodic arrangement of quantum devices rule in this article.On the basis of this mapping, analog processor is initialised, calculates, and reads the result and return this interface computer.This interface computer can be a digital machine.A computer cluster that the example of digital machine includes, but are not limited to a supercomputer, connect by computer network and a table face computing machine.

Minimized this ISG problem that is defined as above-mentioned formula (8) is the example of the np problem classification that can define on a figure and fall into.For example, referring to the paper of 200 years Lidar, New Journal of Physics 6, p.167, this article is combined in this by reference in full.Verified other NP class problems can be mapped to this ISG problem in the polynomial expression step.For example, referring to " the Treating the IndependentSet Problens by 2D Ising Interactions with Adiabatic of Wocjan in 2003 etc.; Quantum Computing, " arXiv.org:quant-ph/0302027 (hereinafter being called " Wocjan "), this article is combined in this by reference in full.According to this method, article and system, a kind of analog processor with Quantum Properties has been described, this analog processor designed to be used and approaches separating of ISG problem, and by extension approach other NP classification problems shine upon separate.

This ISG problem is projected on the two-dimensional lattice that contains summit (being also referred to as node).Many lines (being also referred to as the edge) connect these nodes.Can stipulate the original state of each node in the dot matrix, the weight of each node and the weight at each edge for any given situation of this problem.In these nodes each all has a kind of information state.Be any given edge weights on the dot matrix of N * M and the configuration of node weights for size, wherein N and M representative is along the quantity of node on the side of this dot matrix, and this ISG problem relates to the ground state of determining this node system.In some cases, any edge in this problem all may have one and be about 0 weight, means between respective nodes not connect.Edge weights can be set to scope from J _C ^FTo J _C ^AFValue, amplitude J wherein _C ^FBe ferromagnetic maximum value of being coupled that is coupled possible between the node, and amplitude J _C ^AFIt is antiferromagnetic maximum value of being coupled that is coupled possible between the node.In replacement scheme, J _C ^FCan be less than zero, and J _CA ^FGreater than zero.In another replacement scheme | J _C ^F| greater than | J _C ^AF|.In another replacement scheme | J _C ^F| equal or no better than | J _C ^AF|.For example, consult the U.S. Patent application 60/640 that is entitled as " Coupling Schemes for InformationProcessing ", No. 420, with the U.S. Patent application 11/247 that is entitled as " CouplingMethods and Architectures for Information Processing ", 857, they each all be combined in this in full by reference.

Fig. 2 A illustrates an embodiment of this method, article and system, is used for one four and takes advantage of four rectangular lattices 200, has node N1 to N16 and is coupled J1-2 to J15-16, and 24 are coupled altogether.Be coupled Ji-j node Ni is connected to node Nj.For example, being coupled J3-4 is connected node N3 with N4.These nodes can be represented the summit of a figure problem, and these are coupled the edge that can represent this figure problem.In order to know and, in Fig. 2 A, only to have marked whole nodes and a subclass that is coupled in the dot matrix 200 in order to emphasize numbering convention.Subclass 280 be dot matrix 200 comprise five nodes and four subclass that are coupled.Centroid in subclass 280 has four nearest neighbors and is coupled, and this is the highest number that the nearest neighbor of any node in dot matrix 200 is coupled.

Node on the periphery of dot matrix 200 has only two or three nearest neighbours.The connectivity that dot matrix 200 has is four, is coupled because the node of each non-periphery has four nearest neighbors.In some dot matrix that uses in this method, article and system, the connectivity that this dot matrix has is three, and the quantum devices that means each non-periphery has three nearest neighbors and is coupled.

Fig. 2 B illustrates one four a kind of embodiment of taking advantage of four rectangular lattices 202 according to this method, article and system, have quantum devices N1 to N16 and altogether 42 be coupled be coupled device J1-2 to J15-16.Each quantum devices in the dot matrix 202 is corresponding to a node N in the dot matrix 202.In order to know and, in Fig. 2 B, only to have marked whole quantum devices and a subclass that is coupled device in the dot matrix 202 in order to emphasize numbering convention.Subclass 282 is subclass of dot matrix 202, comprises that nine quantum devices and 20 are coupled device.Center quantum devices in subclass 282 has four nearest neighbors and is coupled, such as J14-15 and four time nearest neighbors are coupled, such as J1-6 and J8-11, and this is the highest number that the nearest neighbor of any quantum devices in dot matrix 202 is coupled.Quantum devices on the periphery of dot matrix 202 has only two or three nearest-neighbors and one or two times nearest-neighbors, is used for altogether three or five and is coupled.The connectivity that dot matrix 202 has is eight, because the quantum devices of non-periphery is coupled to eight neighbours.

Fig. 2 C illustrates the another kind of embodiment according to a dot matrix of this method, article and system.Having connectivity shown in Fig. 2 C is two rectangular lattices of four, a black dot matrix 204 and another is in 205 of white.They are linked together by edge, diagonal angle such as J2-17, and this edge is connected to the node N2 of dot matrix 205 the node N17 of dot matrix 204.Therefore, in a kind of like this structure, each node in each dot matrix 204,205 is connected on another node in another dot matrix diagonally.In other words, this similar is in having two rectangular lattices, and one and each node in each dot matrix is connected on the node of correspondence of another dot matrix on another, and then passes a dot matrix diagonally.Fig. 2 D illustrates the another kind of embodiment that a connectivity with subclass 286 is eight dot matrix 206.Except being rotated 45 °, it is identical with shown in Fig. 2 B structurally.In some cases, the orientation of this dot matrix can be rotated any angle and not loss of function property.The dot matrix 204,205 of Fig. 2 C is not mapped on the dot matrix 206 of Fig. 2 D with can having hell and high water.

Can use connectivity is not 4 and 8 dot matrix, for example uses to have connectivity 2,3,5,6, perhaps 7 dot matrix.Connectivity can be by not using some to be coupled be to simulate on four the dot matrix a connectivity less than 4 dot matrix.For example by not using among Fig. 2 A any one vertically to be coupled, it is two dot matrix that dot matrix 200 becomes a connectivity.Similarly, connectivity can be to simulate on eight the dot matrix by not using some to be coupled in a connectivity less than 8 dot matrix.For example by not using the streaky diagonal angle among Fig. 2 B to be coupled, it is six dot matrix that sub-dot matrix 282 becomes a connectivity.In some cases, can by be coupled accordingly device adjust make this be coupled device to be coupled intensity be zero or realize not using some to be coupled near zero.

Each quantum devices in dot matrix 200 and 202 has a binary value and effectively setover in a this locality, and this biasing drops on about 100 * J _C ^FWith about+100 * J _C ^AFBetween.In addition, each in dot matrix 202 is coupled device and has one from J _C ^FTo J _C ^AFThe value of scope.J _C ^FAnd J _C ^AFAbsolute value can be between about 30 milli Kai Erwen (mK) and about 10 Kai Erwen (K), perhaps, alternately, J _C ^FAnd J _C ^AFAbsolute value can be between about 100mK and about 1.5K.Although the real unit of J is an energy, this unit can be according to formula E=k _BT converts the estimating of a kind of equivalence of the temperature that for example with Kai Erwen is unit, wherein k to _BIt is Boltzmann constant.Can apply this this locality simultaneously to each quantum devices in dot matrix 200 and 202 and effectively setover, the feasible more than one quantum devices of setovering simultaneously.

Fig. 3 A and 3B illustrate an embodiment of this method, article and system, are used for being coupled (J1-3, J2-3, J3-4, J3-5) one planar graph 300 (Fig. 3 A) and connectivity based on dot matrix being conversion between four the layout 301 (Fig. 3 B) arbitrarily having five node N1-N5 and four.The node of Fig. 3 A is corresponding to the quantum devices of Fig. 3 B with same numeral.Fig. 3 B illustrates the embodiment of one nine quantum devices, five quantum devices wherein, and promptly N1 to N5 activates, and four quantum devices do not activate.Quantum devices by dotted line definition among Fig. 3 B is the quantum devices that does not activate, and one of them represents with N ', all the other quantum devices isolation of they and system.A quantum devices that does not activate is to be made as zero by the value of being coupled that is coupled that the quantum devices that does not activate is coupled to adjacent quantum devices, thereby keeps apart with the quantum devices that activates.Should be pointed out that in order to know and, move to the lower right of Fig. 3 B that the label of Fig. 3 A is retained among Fig. 3 B from the upper left side of Fig. 3 B in order to keep state how much.Generally, from one arbitrarily planar graph be that the mapping of four dot matrix is known and effective to a connectivity.For example referring to Wocjan.

Fig. 4 A and 4B illustrate an aspect of this method, article and system, are used for having planar graph 400 (Fig. 4 A) that six node N1-N6 and five are coupled (J1-4, J2-4, J3-4, J4-5, J4-6) to having that nearest neighbor is coupled (J2-4, J4-5, J3-4) and inferior nearest neighbor is coupled the conversion of a dot matrix 402 (Fig. 4 B) of (J1-4, J4-6) at one.The node of Fig. 4 A is corresponding to the quantum devices of the identical label of having of Fig. 4 B.One is utilized nearest neighbor to be coupled and dot matrix that inferior nearest neighbor is coupled is that a connectivity based on dot matrix is eight layout.Fig. 4 B illustrates one six quantum devices embodiment, all six quantum devices wherein, and promptly N1 to N6 activates.Only to have connectivity that nearest neighbor is coupled be the quantum devices that seven activation need be arranged in the dot matrix of one nine quantum devices among four the dot matrix for the same figure shown in Fig. 4 A (connectivity is five) being embedded one.Obviously, have that inferior nearest neighbor is coupled and nearest neighbor is coupled and causes more effectively and simple mapping.

Can be considered to an adjacent quantum devices of this given quantum devices with a given quantum devices each quantum devices in same figure.Alternately, nearest adjacent quantum devices can be defined as with this said quantum devices in same figure and with any quantum devices at the shared edge of this said quantum devices.In another replacement scheme, inferior nearest neighbor quantum devices can be defined as with this said quantum devices in same figure and the edge by two quadratures and another quantum devices be connected to any quantum devices of this said quantum devices.In another replacement scheme, inferior nearest neighbor quantum devices can be defined as any quantum devices that leaves two steps according to manhatton distance.The distance of opening by a single edge separation between two nodes that manhatton distance 1 is a quadrature X-Y scheme.For example the N5 of figure 402 and N6 leave a step mutually when measuring according to manhatton distance.In another example, N4 and N5 leave two steps mutually, and first step is from N5 to N6 and second step is from N6 to N4.In figure 402, this nearest neighbor is coupled and is painted as vertically and horizontal line, for example, is coupled J3-4, and inferior nearest neighbor is coupled with 45 jiaos and draws, and for example, is coupled J1-4.This with nearest neighbor be coupled give for vertically and level inferior nearest neighbor is coupled to give be arbitrarily for the diagonal angle.This time nearest neighbor is coupled and can be drawn as vertically and the edge of level and nearest neighbor are coupled and are drawn as the edge, diagonal angle.For example, the N1 of figure 402 and N4 can be to leave a step according to manhatton distance in this case, and node N1 and N3 can be to leave two steps according to manhatton distance.It can be to intersect that corresponding a pair of nearest neighbor is coupled, for example, figure 402 be coupled J1-4, and J2-3, and nearest neighbor is coupled and does not intersect.Carve alternatively, each time nearest neighbor is coupled to be coupled with another time nearest neighbor and intersects.In another alternative, corresponding a pair of nearest neighbor be coupled can be intersect and inferior nearest neighbor is coupled and does not intersect.

Single between two quantum devices is coupled and can be mapped between three or more quantum devices one or more and be coupled.Under situation about can not arrange these quantum devices adjacent to each other, a kind of like this to be mapped in the layout based on dot matrix be useful.Fig. 5 illustrates a kind of one first figure 500 that simply is coupled Ji-j that is included between node Ni and the Nj.Figure 502 illustrates by being coupled a series of Ji-1 to Jn-j of being coupled that intermediate node N1 to Nn is coupled endpoint node Ni and Nj.Node N1 to Nn is called convenient device node, and is to be used to make things convenient for being coupled between endpoint node Ni and the Nj when these endpoint nodes can not be arranged in the position adjacent in a dot matrix.Be coupled one of Ji-1 to Jn-j and may it seems that is-symbol is coupled.In planar graph 500 arbitrarily, this symbol is got the symbol identical with being coupled Ji-j, and remaining is coupled the ferromagnetic state that is coupled that is fixed in.For example, consideration wherein is coupled Ji-j in figure 500 symbol is positive or antiferromagnetic, and the Ji-1 that is coupled in the figure 502 has been considered to the situation that this symbol is coupled.So, if figure 502 want in the representative of graphics 500 be coupled Ji-j just be coupled Ji-1 be made as positive or antiferromagnetic, and in the figure 502 all the other between node Ni and Nj be coupled be set at negative or ferromagnetic.Equally, the symbol of considering in figure 500, to be coupled Ji-j be that bear or ferromagnetic and figure 502 in the Ji-1 that is coupled still be considered to the situation that this symbol is coupled.In the case, the symbol of the Ji-1 in the figure 502 is set as negative or ferromagnetic, and all the other be coupled also be set at negative or ferromagnetic.Therefore, the Ji-1 is-symbol is coupled, and that J1-1 to Jn-j is set as is negative or ferromagnetic.For the ease of interacting, node N1 to Nn is set to local biasing with zero effect, thereby makes them become passive node and interference-free transmission information between node Ni and Nj.In these two examples of being considered, make in the figure 502 one of be coupled with 500 in to be coupled Ji-j identical, and in the figure 502 all remaining be coupled be set to negative or ferromagnetic.

In making figure 502 one of be coupled with 500 in to be coupled Ji-j identical, and in the figure 502 all remaining be coupled and be set at when negative or ferromagnetic, can be coupled by using fi-SQUIDs or dc-SQUIDs (the two is explanation hereinafter all) reach.Alternately, being coupled in the figure 502 can all be that direct electric current connection makes node Ni be connected electrically to node Nj, in the case all other to be coupled all be ferromagnetic, and therefore always to be coupled Ji-j be ferromagnetic, and node Ni has identical quantum state with Nj.In another replacement scheme, being coupled in the figure 502 can comprise the mixture that an electric current is coupled, rf-SQUID is coupled and dc-SQUID is coupled, fi-SQUID or dc-SQUID one of are coupled be equal to the Ji-j that is coupled in 500, and in figure 502 all the other to be coupled be that bear or ferromagnetic.

Fig. 6 A and 6B show another aspect, in order to comprise that five node N1-N5 and five are coupled the example that the planar graph 600 (Fig. 6 A) of (J1-3, J2-3, J3-4, J4-5) and connectivity based on dot matrix are conversion between four the layout 602 (Fig. 6 B).The node of Fig. 6 A is corresponding to the quantum devices with same numeral of Fig. 6 B.Fig. 6 A illustrates the J4-5 that is coupled between node N4 and the N5.It is the embodiment of a kind of mapping of four layout that Fig. 6 B illustrates for a connectivity based on dot matrix, and illustrate one the 6th quantum devices N6, as a convenient device node (quantum devices) with realize between quantum devices N4 and the N5 between be coupled J4-5.In Fig. 6 B, N4 is connected to N5 by effectively being coupled J4-5.Effectively being coupled J4-5 comprises quantum devices N6 and is coupled J4-6 and J5-6.

When in the figure 600 to be coupled J4-5 be antiferromagnetic the time, to being coupled J4-6 and can giving an amplitude that positive sign is arranged in the dot matrix 602, represent antiferromagnetic being coupled at this this plus sign.Then, will give a suitable amplitude that negative value is arranged, in ferromagnetic being coupled of this this negative value indication to being coupled J5-6.This will make in the spin at N6 place and follow spin at the N5 place.In other words, the spin at the N5 place is copied to N6.Alternately, being coupled J5-6 and can being selected as symbol and being coupled of dot matrix 602, thereby obtain with figure 600 in be coupled the identical symbol of J4-5 (antiferromagnetic being coupled of positive sign indication in this example), so and can be fixed into a kind of ferromagnetic being coupled to J4-6.This will make in the spin at N6 place and follow spin at the N4 place.In other words, the spin at the N4 place is copied to N6.In these two examples, quantum devices N6 is the sub-device of a kind of convenient tolerance and applies one zero local bias-field of imitating, thereby makes the spin states that can follow the tracks of its ferromagnetic with it quantum devices that is coupled in the spin states at N6 place.

This method, article and system by in layout, utilize based on grid nearest neighbor and inferior nearest neighbor be coupled the summit provide nonplanar figure embed into a two dimension based on grid layout in.As known in the affiliated technical field, one has the K of using _nThe complete figure on n summit of expression is the figure on a kind of n of having summit, wherein by an edge between every pair of summit each summit all is connected with during other are fixed a point each.The first five is figure, i.e. K completely ₁To K ₅, be shown among Fig. 7.Such as in this article definition, a nonplanar graph is a kind of complete figure K that comprises this as spirte ₅Perhaps two parts figure K ₃₃Figure.If can being divided into the subclass U of two disconnections and V, its summit make each edge be connected to a summit among the V, figure two parts to a summit among the U.If each summit in U all is connected to each summit among the V, two parts summit is exactly two parts figure completely.V has m unit if U has n unit, just the complete two parts figure K that draws _{N, m}Expression.Fig. 8 illustrates a K _3,3Two parts figure.Any nonplanar graph all is K ₅Perhaps K _3,3One of expansion.By adding edge and node expansion figure.Planar array can be a rectangle.Use for example and comprise the example of finding the solution a problem that on embedding a nonplanar graph that has in the planar array that nearest neighbor and time nearest neighbor be coupled, defines.

Fig. 9 A and 9B illustrate according to system of the present invention, apparatus and method have five node N1-N5 and ten nonplanar graphs 901 that are coupled and have nearest neighbor and a layout 951 (Fig. 9 B) of inferior nearest neighbor idol (that addresses specially below only illustrating in Fig. 9 A and 9B is coupled) in succession based on dot matrix between conversion.It is that those that be miter angle in Fig. 9 B are coupled that inferior nearest neighbor among Fig. 9 B is coupled, and for example is coupled 970.Dot matrix 951 comprises 16 quantum devices, and 12 (N1-N12) wherein activate, and promptly is connected at least one other quantum devices.Be called K at the special pattern shown in Fig. 9 A ₅Figure.It is the figure that connects fully on five nodes, means that each node in the figure all is connected to each other the node in this figure.K5 is minimum nonplanar graph.Any K that comprises ₅Figure as a spirte also will be an on-plane surface.K shown in Fig. 9 A ₅Figure 901 can embed in the dot matrix, for example embeds in the rectangular lattice 951 shown in Fig. 9 B.Similarly, any have a K ₅Can embed in the dot matrix picture point battle array 951 as the figure of spirte.The node that does not activate among Fig. 9 B, promptly N ' is shown in broken lines, and keeps apart from the remainder of this system.In the practice,, for example, be coupled 971 by adjacent being coupled, the value of being coupled be made as zero or more generally be made as an insignificant value, a quantum devices that does not activate is promptly kept apart with the quantum devices that activates.

The sub-device of convenient tolerance is used to carry out from the conversion of a planar graph to a two-dimensional lattice layout.Is that quantum devices, antiferromagnetic being coupled add sub-device of tolerance easy to use and the ferromagnetic a kind of isomorphism that is coupled from nonplanar graph 901 to the conversion of two-dimensional lattice 951.Node N1-N5 in the figure 901 is corresponding to the quantum devices N1-N5 in the dot matrix 951.Symbol coupling portion ground between the node is corresponding to the edge in the figure 901.Edge in the figure 901 is represented by being coupled with ferromagnetic being coupled of symbol that have in the dot matrix 951.Symbol is coupled by the runic solid line between a pair of quantum devices in the dot matrix 951 to be represented, for example is coupled 973.Ferromagnetic is coupled, and for example is coupled 972, is represented by the bold dashed lines in the dot matrix 951.Ferromagnetic being coupled of in dot matrix 951 each passed a sub-device of convenient tolerance.The sub-device of convenient tolerance in the dot matrix 951 comprises quantum devices N6 to N12, and because they are elaborated into lattice, so can pick out at dot matrix 951.Local biasing of the sub-device of each convenient tolerance can be made as zero.The sub-device of convenient tolerance with this ferromagnetic be coupled to cooperate to propagate a kind of symbol be coupled.For example, the sub-device N12 of convenient tolerance is coupled 980 and propagates symbol and be coupled 973 with ferromagnetic.

Each symbol in a two-dimensional lattice layout (for example, layout 951) be coupled both can be one ferromagnetic also can be antiferromagnetic being coupled.Alternately, to be coupled can be a kind of antiferromagnetic being coupled to each symbol in two-dimensional lattice layout.

Figure 10 A, 10B illustrate an example according to this method, article and system, and one of them has nonplanar K of six node N1-N6 _3,3Figure 1001 (Figure 10 A) is embedded in the layout 1051 based on dot matrix (Figure 10 B), should have that nearest neighbor is coupled and inferior nearest neighbor is coupled based on layout of dot matrix.Dot matrix 1051 is a kind of one three K that take advantage of in four arrays that are embedded in _3,3Figure.Node N1 to N6 in the figure 1001 is corresponding to the quantum devices N1 to N6 of dot matrix 1051.Being coupled in the dot matrix 1051 (N1, N2), (N1, N4), (N3, N2), (N3, N4), (N3, N6), (N5, N4) and (N5 is that the antiferromagnetic of edge corresponding to figure 1001 is coupled N6), and by the extrabold wire tag.For example, antiferromagnetic in the dot matrix 1051 is coupled 1070 and is coupled together with the N1 in the dot matrix 1051 and N2 and corresponding to the edge in the figure 1,001 1020.

By convenient device node, be coupled (N1, N6) and (N5, N2) separately by propagating by ferromagnetic being coupled of thick dashed line (being respectively (N7, N2) and (N8, N10), (N10, N9), (N9, N11) and (N11, N6)) expression by antiferromagnetic being coupled with one group with extrabold line (being respectively (N1, N8) and (N5, N7)) expression by a kind of chequer (being respectively N7 and N8, N10, N9 and N11) expression.The sub-device of convenient tolerance, for example, N7, N8, N9, N10 and N11 are the quantum devices with zero local biasing, they cooperate with ferromagnetic being coupled propagates a kind of antiferromagnetic being coupled.For dot matrix 1051, use 11 quantum devices embedding figure 1001, yet, if with the diagonal angle be coupled (N8, N9) and (N9, N6) walks around quantum devices N9 and N11, just can use as few as nine quantum devices embedding figures 1001.Quantum devices N ' with dashed lines the contour representation that does not activate.

As for the K shown in Fig. 7 ₅Figure, any tool K _3,3Nonplanar graph as a spirte can be embedded into as among 1051 the dot matrix,

Figure 11 illustrates a system 1100, and this system is according to an embodiment operation of this method, article and system.System 1100 comprises a digital machine 1102, and this digital machine comprises

● at least one CPU1110;

● a main nonvolatile memory cell 1120 by controller 1125 controls;

● a system storage 1126, preferably high-speed random access memory (RAM) is used for the storage system control program, such as operating system 1130, the data and the application program that are written into from nonvolatile memory cell 1120; System storage 1126 can also comprise ROM (read-only memory) (ROM);

● a user interface 1114, comprise one or more input media (for example, keyboard 1118, mouse 1116) and display 1112, and other peripheral units of Gong selecting for use;

● a network interface unit (NIC) 1124 or other telecommunication circuit; With

● an internal bus 1106 is used for the said elements of interconnection system 1100.

System 1100 further comprises an analog processor 1150.Analog processor 1150 comprises a plurality of quantum devices nodes 1172 and a plurality of device 1174 that is coupled.Although in Figure 11, do not illustrate, quantum devices node 1172 and be coupled device 1174 can be arranged in a connectivity based on dot matrix be in four layouts, for example, shown in Fig. 2 A, 3B, 6B, 12A, 12B, 13A and 13B.Alternately, quantum devices node 1172 and be coupled device 1174 can be arranged in a connectivity based on dot matrix be in eight layouts, for example, as shown in Fig. 2 B, 4B, 9B, 10B and the 14B.So, node 1172 and be coupled that device 1174 is equivalent to diagram in all respects and with respect to any node of these description of drawings or be coupled device.

Analog processor 1150 further comprises a readout device 1160.Readout device 1160 can comprise a plurality of dc-SQUID magnetometers, wherein each dc-SQUID magnetometer all is connected to irritability a different quantum devices node 1172, and according to by each measurement of dc-SQUID magnetometer in readout device 1160, NIC 1124 receives voltage or electric current from readout device 1160.

Analog processor 1150 comprises that further is coupled an apparatus control system 1164, and this is coupled apparatus control system and comprises that being used for each is coupled of device 1174 and is coupled controller.Be coupled in the apparatus control system 1164 each be coupled controller accordingly can be at J _C ^FTo J _C ^AFScope in adjust the intensity that is coupled that is coupled device 1174 of a correspondence, amplitude J wherein _C ^FBe the ferromagnetic possible maximum value of being coupled that is coupled between the node, and amount J _C ^AFIt is antiferromagnetic possible maximum value of being coupled that is coupled between the node.Analog processor 1150 further comprises a kind of quantum devices control system 1162, and this quantum devices control system comprises a controller that is used for each quantum devices node 1172.

Several modules and data structure can be stored and be handled by system 1100.Typically, all or part of various feature and advantage that are stored in the storer 1126 and show this method, article and system for convenience of such data structure, this type of data structure and program module are drawn as the parts of storer 1126.Yet be to be understood that and be stored in the non-volatile storage unit 1120 in this program shown in the system storage 1126 and data structure in any given time.In addition, all or part of of such data result and program module can be stored on the remote computer that does not illustrate among Figure 11, as long as this remote computer can pass through digital machine 1102 addressing.Addressable refers to such an extent that be between this remote computer and digital machine 1102 certain means of communication to be arranged, thereby can pass through a kind of data network (for example, Internet, connected in series, parallel a connection.Ethernet or the like) use a kind of communication protocol (for example, FTP, telnet, SSH, IP, or the like) swap data between these two computing machines.This point is taken into account, be the following describes such data structure and program module.

Computing machine 1102 can be a kind ofly to be used to handle various systems, such as file service and be used to carry out the operating system 1130 of hardware based task.Known a plurality of operating systems that can act on operating system 1130 in the affiliated technical field include, but are not limited to UNIX, Windows NT, Windows XP, DOS, LINUX and VMX.Alternately, can not have operating system, and instruction can be carried out by a kind of mode of daisy chaining.

User interface module 1132 is used to help a user definition and carries out the problem that will find the solution on analog processor 1150.Especially, user interface module 1132 allows a user by setting the J that is coupled between the node _IjValue and this locality biasing h of such node _i, and regulate working time controlled variable such as the annealing scheme define a problem that will solve.User interface 1132 also is provided for dispatching an instruction of separating of calculating and obtaining this problem.Especially, separating this calculating as collecting from the output of analog processor 1150.Subscriber module 1132 can comprise the user interface (GUI) that also can not comprise a figure.When not comprising a GUI, user interface module 1132 receives an instruction sequence that has defined the problem that will find the solution.This instruction sequence can be the form of a kind of macrolanguage of being decomposed by user interface NIM 1132.These instructions can be that XML instruction and this user interface module 1132 can be a kind of XML interpretive routines.

Mapper module 1136 is mapped to the computational problem that will find the solution of user interface module 1132 definition the explanation of the correspondence problem that can find the solution by analog processor 1150.Mapper module 1136 can be the customized configuration desired desirable graphic representation of problem from the statement mapping of a set onto another analog processor 1150 of a tablet pattern.Mapper module 1136 can comprise in disconnected property being eight the defined problem of graphic representation is mapped to an equivalence that defines in a connectivity is eight graphic representation problem.Mapper module 1136 can be represented the equivalence that some np problem (for example, maximum independent set closes, Clique, maximum cutting, TSP problem, the programming of k-SAT integer linear, or the like) is mapped in the ISG model.

In case find the solution after the graphic representation of the needed hope of desirable problem shone upon by a mapper module 1136, analog processor NIM 1138 promptly is used to set up the corresponding value of being coupled and the local bias that is coupled device 1174 and quantum devices node 1172 of analog processor 1150.The function of analog processor NIM 1138 can be divided into three discrete program modules: initialization module 1140, a calculation module 1142 and an output module 1144.

1140 decisions of initialization module are coupled the suitable value of the being coupled J of device 1174 _IjValue h with this locality biasing of the quantum devices node 1172 of analog processor 1150 _i Initialization module 1140 can comprise the instruction that the each side of problem definition is converted to physical values, for example is coupled intensity level and node bias value, and this instruction can be programmed in the analog processor 1150.Initialization module 1140 sends this appropriate signals among the NIC1124 along internal bus 1106 then.NIC 1124 is again to quantum apparatus control system 1162 be coupled apparatus control system 1164 and send such instruction.

For any given problem, each time point in the calculation process, calculation module 1142 is identified for the J that is coupled that is coupled device 1174 of analog processor 1150 _IjThis locality biasing h with quantum devices node 1172 _iAppropriate value, to finish some predetermined calculation scheme.In case calculation module 1142 pairs of calculation schemes have been determined suitable device value and the local bias of being coupled, such signal are just along bus 1106 transmissions and enter NIC1124.NIC 1124 is again to quantum apparatus control system 1162 be coupled apparatus control system 1164 and send such order.

The calculation of analog processor 1150 can be a kind of calculation of thermal insulation or a kind of calculation of annealing.Adiabatic calculation is the calculation of using in adiabatic quantum computation, and calculation module 1142 can comprise the instruction of calculating the state of processor 1150 according to the calculation of using in adiabatic quantum computation.For example, referring to U.S. Patent Publication No. 2005-0256007,2005-0250651 and 2005-0224784, each all is to be entitled as " Adiabatic QuantumComputation with Superconducting Qubits, " they each all in full is combined in this by reference.Annealing is the calculation form that another kind can be used for some analog processor 1150, and calculation module 1142 can comprise the instruction of calculating the state of analog processor 1150 according to the annealing calculation.

Analog processor 1150 signal that is provided by initialization module 1140 and calculation module 1142 is provided finds the solution a quantum problem.In case this problem is found the solution, just can measure separating from state quantum devices node 1172 by readout device 1160 to this problem.Output module 1144 combines work and separates to read this with the readout device 1160 of quantum processor 1150.

System storage 1126 can also comprise a driver module 1146 that is used for to analog processor 1150 output signals.NIC 1124 can comprise with the quantum devices node 1172 of analog processor 1150 and be coupled device 1174 and carries out the desired suitable hardware of interface, no matter this is directly or by readout device 1160, quantum devices control system 1162, and/or is coupled apparatus control system 1164.Alternately, NIC 1124 can comprise from driver module 1146 and order is transformed into software and/or the hardware that directly puts on quantum devices node 1172 and be coupled the signal (for example, voltage, electric current) of device 1174.In another replacement scheme, NIC 1124 can comprise that team is from quantum devices node 1172 be coupled device 1174 switching signals (represent separating or some other forms of feedback of a problem) and change so that make it can be by the software and/or the hardware of output module 1144 decipherings.Therefore, in some cases, initialization module 1140, calculation module 1142 and/or output module 1144 are communicated by letter rather than are directly communicated by letter with NIC 1124 with driving module 1146, so that transmission signal and from analog processor 1150 received signals.

The function of NIC 1124 can be divided into two functional categories: data acquisition and control.Can use dissimilar chips to handle each these discrete functional category.Data acquisition is used for having finished the physical characteristics that this quantum devices node 1172 is measured in a calculating later at analog processor 1150.Can use any amount of customization or commercially available data acquisition microcontroller, comprise, but be not limited to (Fareham by Elan Digital Systems, UK) data collecting card of company's manufacturing comprises AD 132, AD 136, MF 232, MF 236, AD 142, AD 218 and CF 241 cards.Alternately, data acquisition and control can be handled by the microprocessor of single type, are for example handled by Elan D403C or D480C.A plurality of NIC 1124 can be arranged providing quantum devices node 1172 and to be coupled the sufficient control of device 1174, and in order to measure the result of the quantum calculation on analog processor 1150.

Digital machine 1102 can also comprise and being used for to the device separated of other system transmission by a computational problem of analog processor 1150 processing.The device of realizing these devices includes, but are not limited to a telephone modem, a radio modem, a LAN (Local Area Network) connection or a wide area network connection.Digital machine 1102 can produce and embody a carrier for digital signal, and wherein this digital signal is to being encoded by the answer of analog processor 1150 handled these computational problems.

Analog processor 1150 can be a superconduction quantum computer, and the example comprises quantum bit register, readout device, and servicing unit.The superconduction quantum computer is usually worked under the temperature of milli Kai Erwen, and works in a kind of dilution refrigerator usually.An example of dilution refrigerator is a model (Holland, Galgewater No.21,2311 VZ Leiden) of MNK 126 series of Leiden Cryogenics company.The part parts of analog processor 1150 or whole parts can be packed within this dilution refrigerator.For example, quantum devices control system 1162 and be coupled apparatus control system 1164 and can be contained in outside this dilution refrigerator, and the remaining part of analog processor 1150 is installed within this dilution refrigerator.

User interface module 1132, analog processor NIM 1138 and driver module 1146, perhaps its any combination can be implemented with existing software package.The software package that is suitable for include, but are not limited to MATLAB (Massachusetts MathWorks, Natick, Massachusetts) and LabVIEW (Dezhou National Instruments, Austin).

This method, article and system may be embodied to a kind of computer program, and this computer program comprises and is embedded in the computer program mechanism in the computer-readable medium.For example this computer program can contain the program module shown in Figure 11.These modules can be stored in CD-ROM, a DVD, disk storage product, perhaps on any other the computer-readable data and program storage product.These software modules in this computer program are also through the Internet or other modes, and the computer data signal (this software module embeds in this signal) that is embodied in a kind of carrier wave by transmission carries out electronic distribution.

5.2 processor and quantum devices

According to an embodiment of this method, article and system, a kind of machine simulation that can approach the ISG problem that a ground state separates can be provided by the form of a kind of analog processor (for example, the analog processor 1150 of Figure 11).This analog processor comprises a hardware configuration, and this hardware configuration comprises one group of quantum devices (for example, the quantum devices node 1172 of Figure 11).Each this type of quantum devices is defined as at least two base states and can stores binary information with these ground state.This analog processor further comprises and is used for a kind of readout device of this quantum devices (for example, the readout device 1160 of Figure 11), can the binary message of detection of stored in the quantum devices of correspondence.This analog processor comprises that further one group (for example is coupled device, Figure 11 is coupled device 1174), this is coupled device each node is connected to its nearest neighbor node and/or next nearest neighbor node (s), as the above explanation of being done with reference to Fig. 2 A, 2B, 3A, 3B, 4A, 4B, 6A, 6B, 9A, 9B, 10A and 10B.This analog processor comprises that further being used for each is coupled of device and is coupled controller (for example, be installed in Figure 11 be coupled within the apparatus control system 1164).Each is coupled controller accordingly can put in order value J to the intensity J that is coupled that is coupled device of a correspondence _C ^FTo J _C ^AFScope, J wherein _C ^FBeing that this maximum is ferromagnetic is coupled intensity and is negative value and J _C ^AFBe maximum antiferromagnetic be coupled intensity and be on the occasion of.Given of being coupled between two nodes zero J value means that these two nodes are not coupled mutually.

This analog processor further comprises the Node Controller that is used for each quantum devices (for example, being installed in the quantum devices control system shown in Figure 11 1162).Each this type of Node Controller can be controlled the effective biasing on the quantum devices that puts on a correspondence.Effective biasing like this from approximately-100 * | J| changes to approximately+100 * | J| is average maximum value of being coupled of this corresponding node at this J.

Quantum devices in this quantum processor can have different information ground state conveniently to read and initialization.This quantum devices can the utilization sub-feature, the relevant quantum tunneling effect between the incoherent quantum tunneling effect between the ground state, the ground state for example, the Quantum Properties of perhaps involving between the state of different quantum devices, and this quantum devices can be strengthened the computing power of this analog processor.

This analog processor carries out a kind of calculating to approach the ground state of mapped system.This information state is across an energy form (energy landscape), and it depends on the condition of the example defined of this problem.In this energy form, ground state energy is minimum energy point, is called global minimum.This energy form contains local minimum value, and this this locality minimum value can be caught the state (be included in all quantum devices of this dot matrix inside and be coupled) of this system and be prevented that it from shifting to lower energy-minimum.Introducing Quantum Properties makes the state of this analog processor to break away from this local minimum value by tunnel effect, thereby this state can more easily move to lower energy-minimum, in other words can be than there not being quantum tunneling effect to move to lower flux minimum with bigger probability.A kind of like this analog processor can go processing information with the restriction that has substantially reduced than a digital processing unit.

5.2.1 superconducting device

In some embodiment of this method, article and system, the quantum devices of this analog processor (for example, the quantum devices node 1172 of Figure 11) is a plurality of superconduction quantum bits.In such embodiment, this analog processor can comprise the superconduction quantum bit of any number, for example four or more, ten or more, 20 or more, 100 or more, perhaps 1,000 and 1,000, the superconduction quantum bit between 000.

The superconduction quantum bit has two kinds with respect to the operational pattern of the possessing state of canned data therein.When this quantum bit carries out initialization or when measured, this information is classical, promptly 0 or 1, and represent these states of this classical information also be classical so that reliably state prepare.Like this, a quantum bit measurement that first operational pattern is enable state preparation and classical information.This first operational pattern is useful for each embodiment of this method, article and system.

Second operational pattern of a quantum bit occurs in during the quantum calculation.In such quantum calculation process, the information state of this device is leading by quantum effect, thereby this quantum bit performs calculations as the relevant superposition may command of these states, and, in some cases, become with this quantum computer in other quantum bits involve mutually.Yet this second operational pattern is difficult to realize to carry out general quantum calculation with sufficiently high quality.

The superconduction quantum bit can be used as node.It is desirable for reading that computing in first pattern makes them, and has significantly reduced to exist in second operational pattern and limited the difficulty when for example reading this quantum bit, relevant time requirement or the like.A superconduction quantum bit can and rest in this first operational pattern as a node in this analog processor, thereby this quantum bit is retained in this first operational pattern and calculates when not reading.In this way, Zui Xiao Quantum Properties is significantly and to the interference of the state of this quantum bit to be minimum.

The superconduction quantum bit has the characteristic that falls into two types usually: phase qubits and electric charge quantum bit.Phase qubits is with the phase state storage of this device and the quantum bit of process information.In other words, phase qubits is as the information-bearing degree of freedom with phase place.The electric charge quantum bit is state of charge storage and the process information with this device.In other words, the electric charge quantum bit uses electric charge as the information-bearing degree of freedom.In superconductor, between the difference of superconductor, have phase differential, and elementary charge is represented by a storehouse right electron pair of uncle that is called that flows in this superconductor.This device is divided into two classifications in the literary composition of Makhlin, has done general introduction.Phase place and electric charge be in the superconductor relevant value and, on the dominant energy rank of quantum effect, phase qubits has the phase state of the abundant definition that is used to store quantum information, and the electric charge quantum bit has the state of charge of the abundant definition that is used to store quantum information.In this method, article and system, the superconduction quantum bit of the mixture of phase qubits, electric charge quantum bit or phase place and electric charge quantum bit can be used in the analog processor.

Superconducting device equals Nature 398 in 1999 as the experimental realization of quantum bit by Nakamura, p.786 make in, this article is combined in this by reference in full, they have developed a kind of electric charge quantum bit that the fundamental operation that shows a quantum bit requires, but that this quantum bit has is poor (short) decoherence time and strict controlled variable.

5.3 shine upon to superconducting integrated circuit

According to the embodiment of this method, article and system, directly be mapped to an integrated circuit based on the layout of ISG dot matrix, this circuit has satisfied all requirements of the calculating of the ground state of approaching or define exactly this system.This analog processor can comprise:

(i) group node, each node comprise a superconductor loop that is interrupted by one or more Josephson knot;

(ii) one group is coupled device, and this is coupled in the device group each and is coupled device and all is coupled in two nodes in this group node;

(iii) one group of readout device, each readout device in this group readout device is configured to read out in the state of one or more corresponding node in this groups of nodes; With

The (iv) bias unit of one group of this locality, wherein the device of each this locality in the local bias unit of this group is configured to apply on one or more corresponding node at this group node the bias-field an of this locality.

This group is coupled in the device one or more and is coupled device and can comprises a superconductor loop that is interrupted by one or more Josephson knot separately.This parameter that is coupled device is to set on the basis of loop size and Josephson junction characteristic.This be coupled device typically by by or the control system of a correspondence of the biasing that applies magnetic or apply electricity adjust.

Figure 12 A illustrates a figure 1200, have two node N1 and N2 and one single be coupled device J1-2, this be coupled device be coupled mark node N1 and N2.Figure 12 B illustrates the N2 of figure 1200 node N1 and is coupled the conversion of device J1-2 to an integrated circuit 1202.Integrated circuit 1202 comprises superconduction node N1 and N2, and these superconduction nodes are corresponding to the node N1 and the N2 of figure 1200.Integrated circuit 1202 also comprises bias unit 110-1 and 110-2 and readout device 120-1 and 120-2 respectively, and the single device J1-2 that is coupled.In Figure 12 B, node N1 and N2 respectively are a rf-SQUIDs, can comprise 130, one compound Josephson knots 131 of a single Josephson knot.This compound Josephson knot 131 can also be stated a kind of dc-SQUID that is interrupted a superconduction loop as.Can apply magnetic flux so that a kind of modulation to this node parameter of extra degree to be provided to this compound Josephson knot 131 then.Especially, can be by changing the tunnel effect rate of regulating this quantum devices (superconduction node N1) by device 11 electric currents that apply.Equivalently, can regulate the height of the energy barrier 1700 of this system (shown in Figure 17 and explanation hereinafter).

Node N1 and N2 can be three Josephson knot quantum bits.Such structure comprises a superconduction loop that is interrupted by three Josephson knots.Node N1 in the integrated circuit 1202 and N2 respectively have corresponding to the electric current that flows in its corresponding superconduction loop or, two states of two possible directions of supercurrent.For example, represented by the electric current of counter-clockwise circulation in its corresponding superconduction loop by one second state by the electric current representative of clockwise circulation in its corresponding superconduction loop for node N1 and N2 first state.Corresponding to this loop current band table of each this state the different magnetic field that produces by such loop current.

Readout device 120-1 is with 120-2 and be coupled device J1-2 and illustrate with the identical square frame that adds shade at Figure 12 B, because they are devices of same type in some embodiments, have similar structure and parts, yet be configured in integrated circuit 1202, finish different functions.For example, being coupled device J1-2 can be a kind of dc-SQUID, and it is configured to be coupled node N1 and N2 adjustablely.It can be monostable being coupled device J1-2, this means that it has only a potential energy minimum.Readout device 120-1 and 120-2 can be dc-SQUIDs, are coupled to their irritability corresponding node and are configured to controllably detect electric current in these nodes.Alternately, readout device 120-1 and 120-2 can be any devices that can detect the state of corresponding node N1 and N2.

Bias unit 110-1 and 110-2 shown in Figure 12 B are being the loop of metal.Can be by driving an electric current that flows through this bias unit applies a this locality to the node of correspondence from a bias unit 110 magnetic field.Bias unit 110 can be used metal such as the aluminium and the niobium manufacturing of low-temperature superconducting.This bias unit can not be a loop, thereby and can be to be coupled magnetic flux near the electric wire that passes the corresponding node N in this loop simply.Each bias unit 110 can comprise a lead, and this electric wire for example is connected on the base plate passing near the node of correspondence then using a path to be connected on another metal level on the chip.Integrated circuit such as the circuit 1202 of Figure 12 B can be directly from this ISG dot matrix mapping and the control that comprises all needed degree with process information.

Figure 13 A illustrates a kind of embodiment based on 1300 groups of the nodes of dot matrix, comprises a figure, and this figure has five node N1 to N5 and four and is coupled device J1-3, J2-3, J3-4 and J3-5.Figure 13 B illustrates a kind of conversion of dot matrix 1300 to integrated circuit 1302.Integrated circuit 1302 comprises five quantum devices N1 to N5, corresponding to five nodes of dot matrix 1000, is coupled device J1-3, J2-3, J3-4 and J3-5 with four that are connected these five quantum devices, corresponding to the device that is coupled of dot matrix 1000.Integrated circuit 1302 further comprises local bias unit 110-1,110-2,110-4 and 110-5 and readout device 120-1,120-2,120-4 and 120-5.For the sake of clarity, Figure 13 B does not obviously illustrate the bias unit or the readout device of this locality that is used for node N3.The various aspects of integrated circuit 1302 can place each layer separately to go up to optimize space constraint.In the case, the top or the below that can be placed in the layer of making this N3 in order to local bias unit or the readout device of node N3.Each parts of integrated circuit 1302 can with integrated circuit 1002 (mistakes! Do not find Reference source.Corresponding component 10B) is identical, and exception is to share four at the node N3 at center and adjacent node N1, N2, N4 and N5 to be coupled device.

On N1, N2, N4 and N5 side obsolete quantum devices can be arranged in the integrated circuit 1302.Yet for the sake of clarity, so obsolete quantum devices does not illustrate in Figure 13 B.Each figure of coding can utilize any amount of quantum bit that is present in this integrated circuit in integrated circuit 1302.

One or more quantum devices N1 to N5 of integrated circuit 1302 can be configured to a kind of gradometer loop, and this magnetic field just influences this gradiometer loops when this loop is crossed in magnetic field unevenly thereby have only.Gradiometer is useful for assisting to be coupled and being used to reduce this system to the susceptibility of external magnetic noise.Can become vertical angle or be coupled (for example crosstalking) with the parasitism that reduces between the adjacent node with subvertical angular arrangement nearest neighbor node.When one second main axis of one first main axis of this first node and this Section Point alignd mutual vertically, one first and second node can be considered to orthogonal angular arrangement.

Figure 14 A illustrates node N1 to N9 with nine activation and the corresponding a kind of embodiment that is coupled of device based on 1400 groups of the nodes of dot matrix, and Figure 14 B illustrates a dot matrix 1400 and has nine node N1 to N9 and 20 conversions that are coupled the integrated circuit 1402 of device to one.For the sake of clarity, in Figure 14 A and 14B, only mark node N1, N2, N4 and N5, and be coupled J1-4, JN1-5, JN2-4 and J4-5.Local bias unit 110-1,110-7,110-8 and 110-9, and readout device 120-3,120-6 and 120-9 also mark in integrated circuit 1402.Figure 14 B does not at large comprise the local bias unit of all nodes.The each side of integrated circuit 1402 can place layer separately to go up to optimize space constraint.In the case not the bias unit of this locality of the node of the bias unit an of this locality shown in Figure 14 B or readout device can be placed in the layer of wherein making these these nodes above or below.These bias units can not be loops, but near the simple electric wire that node N, passes and magnetic flux is coupled in this loop.This bias unit can be connected to a path then by an electric wire that passes at same or different layer near this quantum bit, this path is connected to another metal level on this chip, for example be connected on the base plate.

Each parts of integrated circuit 1402 can be identical with the parts of integrated circuit 1202 and 1302.Such parts are illustrating above with reference to Figure 12 B and 13B.A difference of integrated circuit 1402 and other circuit is to have added inferior nearest neighbor to be coupled device JN in integrated circuit 1402, for example JN2-4 and JN1-5.As shown in FIG., inferior nearest neighbor is coupled device JN2-4 leap time nearest neighbor and is coupled device JN1-5.Be coupled device JN1-5 and JN2-4 one or both electric wires in all can be on a plurality of layers.

Inferior nearest neighbor is coupled device, for example is coupled device JN2-4 and JN1-5, can be dc-SQUID, perhaps alternately can be rf-SQUIDs.What they can be equivalent to Figure 12 B is coupled device J, and only different on its structure.Three readout device 120-3,120-6,120-9 only are shown in Figure 14 B, are used for reading respectively corresponding node N3, N6 and N9.Every other node can have corresponding readout device 120.Alternately, can only use the minority readout device, and the state reproduction technology that can use classics is duplicated the state of internal node to all mid-side node N3, N6, N9, for example, as be entitled as " Methods of Ferromagnetic and Adiabatic Classical Qubit StateCopying; " U.S. Patent Application Serial 60/675,139 illustrated, this application in full is combined in this by reference.

Although do not illustrate in Figure 14 B, can there be the quantum devices that there is not use on quantum devices N1, N2, N3, N4, N6, N7, N8 and N9 side at periphery in the integrated circuit 1402.One or more quantum devices N1 to N9 in integrated circuit 1402 can be configured to a kind of gradometer loop, and this magnetic field just influences this gradiometer loops when this loop is crossed in magnetic field unevenly thereby have only.Gradiometer loops is useful for helping to be coupled and being used to reduce this system to the susceptibility of external magnetic noise.Can become vertical angle or be coupled (for example crosstalking) with the parasitism that reduces between the neighbors with the nearest node of subvertical angular arrangement.

Figure 15 illustrates the photo as an example of a physical layout of this method, article and system.Four quantum devices based on flux, 1501-1 to 1501-4 have been manufactured on the superconducting integrated circuit.Each quantum devices uses nearest neighbor and time nearest neighbor to be coupled and is connected on each other the quantum devices in the photo.For example, being coupled device J1-3 is one and is used for the nearest neighbor that quantum devices 1501-1 and 1501-3 are coupled together is coupled device.Nearest neighbor is coupled and also is present between quantum devices 1501-1 and the 1501-2, between 1501-2 and the 1501-4, and between 1501-3 and the 1501-4, does not at large indicate although these are coupled device.Being coupled device J2-3 is the example that time nearest neighbor is coupled, and quantum devices 1501-2 and 1501-3 are coupled together.Another time nearest neighbor is coupled and is present between quantum devices 1501-1 and the 1501-4, although at large do not indicate it.Have readout device and local bias unit on this circuit, but they do not illustrate in Figure 15 yet.

Figure 16 illustrates another alternative arrangement as this method, article and system.Six quantum devices are arranged in the drawings, and three in them are labeled as 1601-1,1601-2, and 1601-3.Yet the layout shown in the figure can easily expand to any amount of quantum devices.Quantum devices 1601-1 and 1601-2 link together by being coupled device J1-2.Being coupled device J1-2 can be a rf-SQUID, perhaps alternately is a dc-SQUID.Quantum devices 1601-1 and 1601-3 are coupled together by being coupled device J1-3, and this a kind of direct electric current in Figure 16 is coupled.Thereby, quantum devices 1601-1 and 1601-3 be coupled ferromagneticly and have an identical quantum state.That implements is coupled device J1-3 and can comprises and utilize a plurality of paths to create a path that is coupled device of using a plurality of metal levels.Example is the intersection J1-3-A among Figure 16, is manufactured on another metal level and uses two paths to be connected on the original layers in this part that is coupled device J1-3.Such technology is known in the affiliated technical field.

5.4 simulation process

5.4.1 system level

An aspect of this method, article and system provides the method for finding out the minimum energy configuration or approach the minimum energy configuration under given one group of starting condition.These methods generally include the problem that will find the solution are mapped on the dot matrix layout topology.The topology of this dot matrix layout is mapped on the circuit of the dot matrix that comprises a quantum devices, arranges to be coupled between the dot matrix of this quantum devices.This quantum devices and be coupled individually initialization and by at this quantum devices be coupled and use local biasing control or by using a kind of global bias field to bring out control working time.In this way, the dot matrix layout topology of the problem that will find the solution of representative is mapped on the dot matrix of a physics of quantum devices.Then the end-state of the dot matrix of this quantum devices as separating of this problem read.This separates the form that binary number can be arranged.

5.4.2 initialization

Initialization with analog processor of quantum behavior comprises that each that will be used to show the problem found the solution with the state initialization at each quantum devices place and with each is coupled the state initialization of device.The potential energy curve of representing the quantum devices of a node in the figure that will find the solution can be an a kind of pair of trap potential energy, be similar to that Friedman in 2000 etc. set forth " Detection of a Schr  dinger ' s Cat State in an fi-SQUID; " arXiv.org:cond-mat/0004293v2, this article are combined in this by reference in full.Figure 17 A and 17B respectively illustrate the figure of two trap potential energy.Energy meter is shown on the y-axle, and some other the subordinate variable that is associated with this device, such as the inside flux of this quantum devices is illustrated on the x-axle.This system is by an a kind of particle explanation of moving in this power curve.If this particle is in left trap, it just is in | L state, and if this particle be in the right trap, it just is in | R state.These two states can be denoted as respectively | and 0〉and | 1 〉, perhaps be denoted as respectively | 1〉and | 0 〉.In a superconduction flux qubit or steady current quantum bit, these two states are corresponding to two of loop current different directions, left circulation and right circulation.The initialization of the state at each node place can be adjusted the biasing at each node place by this locality, or by using a kind of global bias field to carry out.Alternatively, such adjustment can also be undertaken by the barrier height between the reduction state.If this potential energy curve is to lopsidedness, as shown in Figure 17 A, this particle will have bigger probability to move in the lower flux trap.Under the situation of Figure 17 A, this can be | R〉state 160-1.If this potential energy well curve tilts at opposite side, this particle will have bigger probability to move in the opposed trap.Under the situation of Figure 17 A, this can be | L〉state 160-0.

Its state of initialization comprises that by a quantum devices of the position description of particle in two trap gesture this locality by adjusting the node place is biased in this potential energy of lopsidedness, and waits for that the sufficiently long time makes this particle move to lower potential energy with certain high probability.Local field biasing can be a kind of magnetic field, and the field biasing of adjusting this this locality at the node place can be included in closely and apply an electric current near this quantum devices place to a superconduction loop or coil, thereby produces the magnetic field biasing an of this locality in this quantum devices.Through after the sufficiently long time, the state of this device will relaxation enter this pair trap gesture than in the small throughput trap, desirable original state that Here it is.The state of this device can lose by the heat ease and fall into lower energy wells, and perhaps, the state of this device can reach lower energy by the tunnel effect process of passing at this barrier 1700.In some cases, the heat ease is become estranged the tunnel effect process all to the initialisation effect.

Field biasing that should this locality in each quantum devices place initialization comprises the field biasing of an overall situation setting up a whole dot matrix of striding this quantum devices and waits for time of certain-length.The field biasing that applies an overall situation causes that all quantum devices are initialized to same state.This global bias can be a kind of magnetic field.Represent each quantum devices of a node can comprise the loop of the superconductor that is interrupted by one or more Josephson, wherein initialization can be striden all quantum devices and undertaken by the magnetic field that applies an overall situation, and this will cause that each quantum devices all is initialized to identical steady current state.

It at the quantum devices of an integrated circuit that can be used for finding the solution a computational problem loop of a superconductor being interrupted by one or more Josephson knot.Can suitably constitute a kind of like this loop, make it to have the illustrated potential energy family curve of a kind of pair of trap gesture that is similar to described in Figure 17 A or the 17B.Two traps of in this pair trap gesture this are corresponding to two different directions of the steady current (for example, the electric current 102-0 and 102-1 of Figure 1A) of this superconductor loop.Can become desirable state to this loop initialization by this pair trap gesture that tilts, as shown in Figure 17 A.For example, can be by apply this inclination of flux biasing causing of an outside through this superconduction loop.In some cases, be initialised to minimum energy state, just can have removed this outside flux biasing in case determine the state of this quantum devices.By the loop of an electric wire or coil are placed closely near this superconduction loop and through the loop of this electric wire in other words coil apply a bias current, can be applied to the flux of an outside on the superconduction loop.This bias current is through the change in this superconducting ring pass magnetizing field, and this changes the potential energy that influences quantum devices.

Can change the height of barrier 1700 by the critical current that changes the Josephson knot that is interrupted this superconduction loop.In the rf-SQUID of a standard, in a single day can in manufacture process, carry out this change, but constitute this device, the critical current of a knot is generally fixed.Yet, if the single Josephson that replaces among the rf-SQUID with a compound Josephson knot ties, even just might after manufacturing, also can adjust effective critical current.This finishes by rimala knot loop is applied a magnetic field, and by adjusting effective critical current that this magnetic field changes this rf-SQUID.

One or more quantum devices that plays the node effect in an integrated circuit can be rf-SQUIDs.A rf-SQUID is the loop of a superconductor, is interrupted this loop with one or more Josephson knot.The device that has three Josephson knots in this loop is called as a 3JJ quantum bit.Can dispose a such rf-SQUID type device so that its potential energy family curve by the explanation of a kind of pair of trap gesture.Two traps in this pair trap potential energy are corresponding to two of the steady current in the loop of superconductor different directions.The device of the rf-SQUIDs that represents the quantum performance that has is set forth in the literary composition of Friedman in 2000.By to placing to such an extent that closely apply a bias current, can apply the flux of an outside on the superconduction loop of this rf-SQUID near the loop or the coil of an electric wire of the superconduction loop of this rf-SQUID.

Each quantum devices (for example, quantum processor) that is used for finding the solution an integrated circuit of a quantum problem can be a kind of superconductor loop that is interrupted with three Josephson knots.The initialized method of the quantum bit of these types can be identical with those of above situation explanation at the rf-SQUID quantum devices.The device of these types does not require big loop inductance, and does not therefore require that big loop area is to have two trap potential energy family curves.There is the device of three Josephson knots to be set forth among the Orlando.One or more quantum devices can be a kind of steady current quantum bit, and for example shown in Figure 180 the sort of, this figure duplicates from the Orlando literary composition.A kind of like this device can be used as the quantum devices in this method, article and the system.Each Josephson in Figure 18 ties by an X mark, and by a desirable Josephson knot and a capacitor C _iParallel connection combination simulate.Resistive passage supposition in parallel is negligible.Desirable Josephson knot has a kind of current phase relations I _i=I ₀Sin  _j,  in the formula _jIt is the gauge invariance phase place of knot i.A characteristic of the Josephson energy of each knot X among Figure 18 is that it is the function of two phase places.These two the phase place  of scope that suppress f for a magnetic ₁And  ₂Allow two stable configuration, the DC electric current that these two two stable configuration flow corresponding to reverse direction.Such as in Orlando discussion, by considering the rechargeable energy (capacitive character energy) in this knot and mechanically consider this circuit quantum, can be adjusted to this circuit parameter and make near two minimum states of this system the f=1/2 will be for two classical states of opposite loop current.

The all or part of quantum devices of an integrated circuit that is used for finding the solution a computational problem can be compound Josephson knot rf-SQUID.Except the dc-SQUID that is connected to this rf-SQUID loop by substituted single Josephson knot, a compound Josephson knot rf-SQUID was similar to a rf-SQUID, and this dc-SQUID is also referred to as compound Josephson knot.A dc-SQUID is become by the Josephson structure that two or more and two electric contacts that are formed between this knot are connected in parallel.The performance of this device is similar to a rf-SQUID, exception be in loop, can change that it has a kind of extra control degree on the meaning of critical current by the flow through flux of this dc-SQUID loop of adjustment.Adjust this critical current and changed separately left trap and the right trap state of this pair trap gesture | L and | R〉the barrier height.The flux that passes big rf-SQUID loop is still adjusted the inclination of this pair trap gesture, as among the rf-SQUID of a standard.This quantum devices initialization both can have been comprised by apply the biasing of flux to this rf-SQUID this pair trap gesture is tilted, also can reduce this barrier height by apply a biasing to this dc-SQUID loop, or both dual-purposes, and wait for that then this device is initialized to ground state.The σ of the flux representative of this dc-SQUID loop to the state of this quantum devices passed in adjustment ^XControl.

Each quantum devices that is used for finding the solution the integrated circuit of a computational problem can be a kind of gradometer quantum bit.The initialization of gradiometer quantum bit is with the method that is similar to initialization rf-SQUIDs.The initialized method of gradiometer quantum bit comprised apply the biasing of flux and wait for certain section time span.Apply a bias current applies flux from an outside a to loop by coil or loop to the electric wire of closely placing near this loop.The gradient quantum bit is led to by mutual Electricity Federation and has two superconduction leaves of opposite direction of current to constitute.Initialization can relate to one of these two leaves or apply a flux jointly to the two setovers.

The initialized method of quantum devices has been discussed hereinbefore.Being coupled device also is initialised.In some cases, be set to a desirable original state and wait for then as this certain period that is coupled the feature of device and be coupled device and in fact be set in desirable original state by this being coupled device to guarantee this.Initialized result as such is coupled the state that device is initialised to a kind of J=-1 or J=1 to this, and wherein this is coupled intensity J and is standardized as and makes for a given problem, and the desirable intensity that is coupled is corresponding to J=|1|.

It can be a kind of quantum superconducting device that in integrated circuit at least one is coupled device.For example, being coupled device can be rf-SQUID in this integrated circuit.Under these circumstances, one can comprise to this as the initialization of rf-SQUID that is coupled device and to be coupled the flux biasing that device applies a this locality.This can be by closely finishing near this superconduction loop or bias current of coil arrangement that is coupled device through one.As the Rf-SQUID that is coupled device can be monostable, and this refers to such an extent that be that its potential-energy function has only a minimum value.The all or part of device that is coupled in this integrated circuit (for example, quantum processor) can be dc-SQUID, and such initialization that is coupled device comprises to such device that is coupled and directly applies a bias current.

The all or part of device that is coupled in this integrated circuit (for example, quantum processor) can be that gradiometer is coupled.To be coupled initialized method as the gradiometer that is coupled device in the integrated circuit and comprise that whole two foliar spraies to a leaf of this gradiometer or this gradiometer add a flux biasing.

5.4.3 control working time

According to the embodiment of this method, article and system, the method for carrying out control working time of an analog processor comprises that changing this quantum devices effectively setovers.This can by separately local biasing of each quantum devices place adjustment of this analog processor, adjust in this analog processor quantum devices between the intensity that is coupled that is coupled carry out, or undertaken by the barrier height of adjusting each quantum devices, the adjustment of this barrier height is equivalent to the effective temperature that changes this system, and wherein this system is made up of a quantum devices and the dot matrix that is coupled element.

The barrier height that correspondingly reduces or improve this quantum devices just is enough to improve or reduce the effective temperature of this system.The barrier height of a quantum devices is the potential energy barrier between two potential wells of this energy form, is depicted as barrier 1700 at Figure 17 A and 17B.If this quantum devices comprises a compound knot, the barrier height of this quantum devices can change by the external magnetic field that the loop of this composite junction is passed in adjustment.

If reach the whole last current state of an analog processor with this effective temperature, at first reduce the potential energy barrier of all quantum devices, this just is easy to carry out the mistake of heat ease from local minimum value by the quantum state that makes this analog processor increases this effective temperature.Improve the potential energy barrier of this quantum devices then lentamente, thereby reduce effective temperature, make the quantum state of this analog processor can find lower minimum value.

The annealing of losing by the heat ease is called as classical annealing purely, because it does not utilize the quantum effect of this system.The method that finds the whole last current state of this analog processor can be classical fully.Alternately, can be additional to classical annealing and carry out quantum annealing.A kind of form of quantum annealing is a quantum tunneling effect, the tunnel effect by passing this potential barrier rather than lose wherein by the heat ease, and the quantum state of this analog processor finds a minimum value lower than its present located state.Thereby when the probability that is lost by its existing minimum value heat ease was little statistically, quantum annealing can help this quantum state to find lower minimum value.

Finding the whole last current state of an analog processor to calculate by the quantum of thermal insulation carries out.In the quantum of thermal insulation develops, this analog processor is initialised to a kind of ground state of quantum state of known Hamiltonian (Hamiltonian).Allow this quantum state can develop a kind of Hamiltonian at end eventually then adiabaticly.This thermal insulation develops must be enough to prevent that this quantum state from moving to a kind of excited state from ground state usually slowly.Adiabatic differentiation can be by adjusting being coupled intensity or by adjusting each other biasing of this quantum devices, or being undertaken by adjusting the global bias that influences all quantum devices between the quantum devices in this processor.This ground state representative is separated by this Hamiltonian computational problem of encoding end last end last.U.S. Patent Application Publication No. 2005-0256007,2005-0250651 that the more information of this process for example can be mentioned referring to preamble, and 2005-0224784.

The method of carrying out control working time of an analog processor comprises the method that improves the actual temperature of this analog processor by a kind of thermal anneal process.This thermal annealing process can comprise brings up to one to the temperature of this system in the temperature between the 30mK to 3K from basal temperature, and then the temperature of this system is reduced to this basal temperature.

5.4.4 read

The method of reading the state of a quantum devices (for example, quantum processor) in the integrated circuit can comprise a readout device initialization and measure a physical characteristics of this readout device.A quantum devices has two possible states of reading, promptly | 0〉state and | 1〉state.Reading a quantum devices disintegrates the quantum state of this device to one of two classical states.Wherein the barrier height on this quantum devices is adjustable, can improve this barrier height before reading the state of this quantum devices.Improve this barrier, the barrier 1700 of Figure 17 for example, this quantum devices is frozen in or | 0〉state or | 1〉state.

This readout device is connected to the dc-SQUID magnetometer of this quantum devices with can comprising an irritability, determines that in the case the state of this quantum devices can comprise from voltage of this dc-SQUID magnetometer measures or electric current.Can become the value of representative to this voltage or current conversion then in the magnetic field at this quantum devices place.

Classical state duplicates the number that can be used to reduce desired readout device.For example consult the U.S. Patent application 60/675,139 of preamble reference.

Read after the state of this quantum devices, the result of this measurement can use and be embodied in a data-signal on the carrier wave and transmit.This data-signal can be a digital signal, and in some cases, can use digital machine 1102 (being shown among Figure 11) to produce this carrier wave.

5.5 the list of references of quoting

All United States Patent (USP)s of mentioning in this instructions, U.S. Patent Application Publication, U.S. Patent application, foreign patent, foreign patent application and off-patent document include, but are not limited to US6,670,630, US6,784,451, US6,822,255, US6,885,325, US6,897,468, US6,960,780, US6,979,836, US2002-0121636, US2003-0107033, US2003-0121028, US2003-0169041, US2003-0173498, US2004-0000666, US2004-0016918, US2004-0119061, US2004-0140537, US2005-0224784, US2005-0250651, US2005-0256007, U.S. Patent Application Serial 60/640,420,60/675,139 and 11/247,857, all be combined in this in full by reference, and be for all purposes.

5.6 alternate embodiments

For one of ordinary skill in the art clearly, can make up the various enforcements of above elaboration Mode is to provide other embodiment. If need, can revise each side of the present invention Face provides the present invention with system, circuit and the concept of utilizing various patents, application and document Further embodiment. Can according to the elaboration of preamble make such to the present invention or The change that the person is such. In a word, in appended claims, employed term should not Be interpreted as limiting the invention to the concrete reality that discloses in the specification and claims Execute mode, and should be interpreted as comprising all possible embodiment, together with this claim Have the right all equivalents of requirement of book. Therefore, the invention is not restricted to disclosed content, phase Anti-its category should be determined by claims fully.

Claims (79)

1. computing system comprises:

A computing machine; With

Quantum processor with this compunication, this quantum processor comprises:

(i) a plurality of quantum devices, wherein each quantum devices in these a plurality of quantum devices is the node of a dot matrix, and wherein a quantum devices in these a plurality of quantum devices has one first base state and one second base state: and

(ii) a plurality of devices that are coupled, wherein be coupled in the device one first at this a plurality of this and be coupled one first quantum devices of device in will these a plurality of quantum devices and be coupled to one second quantum devices in these a plurality of quantum devices, wherein the configuration of first quantum devices of this in this dot matrix and this second quantum devices is selected from by a kind of nearest neighbor configuration and a kind of nearest neighbor and disposes the group that constitutes.

2. system according to claim 1, wherein a quantum devices in these a plurality of quantum devices comprises a superconductor loop of being interrupted by at least one Josephson knot.

3. system according to claim 2, wherein at least one the Josephson knot in this quantum devices is a kind of compound Josephson knot.

4. system according to claim 1, wherein these a plurality of quantum devices are with a kind of two-dimensional array arrangement, wherein this two-dimensional array has by the width of more than first a node n definition and the length that is defined by more than second a node m, and wherein this two-dimensional array comprises an inside and a periphery.

5. system according to claim 4, wherein each quantum devices in this inside is coupled on the quantum devices of four nearest neighbors in this dot matrix and is coupled on the quantum devices of four nearest neighbors in this dot matrix.

6. system according to claim 1, wherein, this quantum devices in these a plurality of quantum devices has a kind of configuration of gradiometer.

7. system according to claim 1, wherein

This first quantum devices and second quantum devices are the nearest-neighbors in this dot matrix; And

Be coupled device these a plurality of second of being coupled in the device this first quantum devices is coupled to one the 3rd quantum devices in these a plurality of quantum devices, wherein this first quantum devices and the 3rd quantum devices are the inferior nearest-neighbors in this dot matrix.

8. system according to claim 1, wherein, these a plurality of at least one that are coupled in the device are coupled device and comprise a superconductor loop that is interrupted by at least one Josephson knot.

9. system according to claim 1, wherein, these a plurality of at least one that are coupled in the device are coupled device and comprise a superconductor loop that is interrupted by at least one compound Josephson knot.

10. system according to claim 1, wherein, these a plurality of at least one that are coupled in the device are coupled device and are selected from the group that is made of a kind of rf-SQUID and a kind of dc-SQUID.

11. system according to claim 1, this quantum processor further comprises:

A readout device, this readout device are configured to read a kind of state of at least one quantum devices in these a plurality of quantum devices.

12. system according to claim 11, wherein this readout device is to be selected from the group that is made of a kind of dc-SQUID and a kind of magnetometer.

13. system according to claim 1, wherein this first is coupled device and is configured to be adjusted at that between this first quantum devices and this second quantum devices one is coupled intensity so that this first quantum devices and this second quantum devices are coupled in ferromagnetic mode.

14. system according to claim 13, wherein, this first is coupled device and is configured to local the biasing that this first quantum devices applies a kind of null effect.

15. system according to claim 1, wherein, this first is coupled device and is configured to adjust and is coupled intensity so that this first quantum devices and this second quantum devices are coupled in anti-magnetic mode between this first quantum devices and this second quantum devices.

16. system according to claim 1, wherein, this first quantum devices and this second quantum devices are neighbours nearest in this dot matrix.

17. system according to claim 1, wherein, this first quantum devices and this second quantum devices are time nearest neighbours in this dot matrix.

18. system according to claim 1, wherein, this quantum processor is used to find the solution a kind of computational problem, and this computational problem is selected from by a problem with complicacy P, problem with complicacy NP, one has the problem of a complicacy NP-difficult problem and one by the complicacy NP-group formed of problem completely.

19. system according to claim 18, wherein, this quantum processor and a transmitter are coupled, and wherein of being configured to send as this computational problem that is embodied in the data-signal in a kind of carrier wave of this transmitter separates.

20. system according to claim 1, wherein this computing machine is a kind of digital machine.

21. use a quantum processor to determine a kind of method of the result of a computational problem, this method comprises:

(i) this quantum processor is initialized to a kind of original state, wherein this quantum processor comprises a plurality of quantum devices and a plurality of device that is coupled, and wherein each in these a plurality of quantum devices is coupled device and is coupled a pair of quantum devices in these a plurality of quantum devices, wherein this initialization comprises a kind of state of setting at least one quantum devices in these a plurality of quantum devices and sets a kind of intensity that is coupled that these a plurality of at least one that are coupled in the device are coupled device;

(ii) allow this quantum processor can calculate a kind of whole last current state, wherein should end last current state approach a kind of natural ground state of this computational problem; With

Thereby a kind of whole last current state of (iii) reading at least one quantum devices in these a plurality of quantum devices is determined the result of this computational problem.

22. method according to claim 21, wherein, at least one quantum devices in these a plurality of quantum devices comprises the loop of a superconductor that is interrupted by at least one Josephson knot.

23. method according to claim 21, wherein, at least one quantum devices in these a plurality of quantum devices is a kind of rf-SQUID.

24. method according to claim 21, wherein, at least one quantum devices in these a plurality of quantum devices comprises a loop of the superconductor that is interrupted by at least one Josephson knot and at least one compound Josephson knot.

25. method according to claim 21 wherein, allows this quantum processor can calculate a kind of whole last current state, this state comprises the effective temperature that reduces this quantum processor and adiabatic at least one that calculate in this quantum processor.

26. method according to claim 21, wherein, this computational problem is to be selected to have the problem of a complicacy NP-difficult problem and one by a problem with complicacy P, problem with complicacy NP, one and have the complicacy NP-group formed of problem completely.

27. method according to claim 21 wherein, is initialized as a kind of whole last current state with this quantum processor, comprising:

One first quantum devices in these a plurality of quantum devices is initialized as local biasing with a null effect; And

These a plurality of of being coupled in the device are coupled device are initially one second quantum devices that first quantum devices in these a plurality of quantum devices is coupled to these a plurality of quantum devices in ferromagnetic mode.

28. method according to claim 27, wherein, this computational problem is that a kind of maximum independent set closes problem.

29. a computer system is used for determining the result of a computational problem, this computer system comprises:

A CPU (central processing unit); With

A storer that is coupled to this CPU (central processing unit), this memory stores:

A user interface module that comprises the instruction that is used to define this computational problem;

A mapper module that comprises a kind of mapping that is used to produce this computational problem;

An analog processor NIM comprises:

(i) be used for transmitting the instruction of this mapping to an analog processor, wherein this analog processor comprises a plurality of quantum devices and a plurality of device that is coupled, this mapping comprises the initialization value of at least one quantum devices that is used for these a plurality of quantum devices, with be used for the initialization value that these a plurality of at least one that are coupled device are coupled device, wherein these a plurality of of being coupled in the device relevant quantum devices of being coupled the correspondence of device in will these a plurality of quantum devices is coupled among relevant one the nearest neighbours that answer quantum devices with this of nearest neighbours of this relevant quantum devices at least one; With

(ii) receive a result's instruction from this analog processor in response to this mapping.

30. computer system according to claim 29, this storer be a kind of driver module of storage further, this driving module comprises:

Be used for transmitting the instruction of this mapping to this analog processor; With

Be used for receiving this result's instruction from this analog processor; And the instruction that wherein is used for this mapping is transferred to this analog processor in this analog processor NIM comprises instruction from this mapping to this driver module that transmit; And the instruction that is used to receive this result in this analog processor NIM comprises in response to this mapping and receives instruction from this result of this analog processor from this driver module.

31. computer system according to claim 29, wherein, the instruction that is used to define this computational problem comprises the instruction that is used to analyze the instruction group that this computational problem is encoded.

32. computer system according to claim 29, wherein, this computational problem be selected from by a problem with complicacy P, problem with complicacy NP,

Problem with complicacy NP-difficult problem and one have the complicacy NP-group formed of problem completely.

33. computer system according to claim 29, wherein, this computational problem is to be selected from by a Yi Xin spin glass problem, a maximum independent set to close problem, a Clique problem, a maximum cutting problem, mobile sales force's problem, a k-SAT problem and the group that integer linear programming problem is formed.

34. unite the computer program of use with a computer system for one kind, this computer program comprises a computer-readable medium and a computer program mechanism that is embedded in wherein, this computer program mechanism comprises:

(A) user interface module that comprises the instruction that is used to define a computational problem;

(B) mapper module that comprises the instruction of a mapping that is used to produce this computational problem; With

(C) analog processor NIM comprises:

(i) be used for transmitting the instruction of this mapping to an analog processor, wherein this analog processor comprises a plurality of quantum devices and a plurality of device that is coupled, this mapping comprises the initialization value of at least one quantum devices that is used for these a plurality of quantum devices, with be used for the initialization value that these a plurality of at least one that are coupled device are coupled device, wherein these a plurality of at least one that are coupled in the device are coupled among one the nearest neighbours that device is coupled to the relevant quantum devices of a correspondence in these a plurality of quantum devices nearest neighbours of this relevant quantum devices and this relevant quantum devices at least one; With

(ii) receive a result's instruction from this analog processor in response to this mapping.

35. computer program according to claim 34, this computer program further comprises:

(D) a kind of driver module comprises:

Be used for transmitting the instruction of this mapping to this analog processor; With

Be used for receiving this result's instruction from this analog processor; And wherein

The instruction that is used for this mapping is transferred to this analog processor in this analog processor NIM comprises the instruction that is used for transmitting to this driver module this mapping; And

The instruction that is used to receive this result in this analog processor NIM comprises in response to this mapping and receives instruction from this result of this analog processor from this driver module.

36. computer program according to claim 34, wherein, this instruction that is used to define this computational problem comprises the instruction that is used to analyze the instruction group that this computational problem is encoded.

37. computer program according to claim 34, wherein this computational problem is to be selected to have the problem of a complicacy NP-difficult problem and one by a problem with complicacy P, problem with complicacy NP, one and have the complicacy NP-group formed of problem completely.

38. computer program according to claim 34, wherein, this computational problem is to be selected from by a Yi Xin spin glass problem, a maximum independent set to close problem, a Clique problem, a maximum cutting problem, mobile sales force's problem, a k-SAT problem and the group that integer linear programming problem is formed.

39. a quantum processor comprises:

Be arranged in a plurality of quantum devices in the dot matrix;

One more than first are coupled device, wherein be coupled device one first quantum devices and one second quantum devices in will these a plurality of quantum devices these more than first of being coupled in the device and be coupled, wherein this first quantum devices and second quantum devices are neighbours nearest in this dot matrix; With

One more than second are coupled device, wherein be coupled device one the 3rd quantum devices and one the 4th quantum devices in will this a plurality of quantum devices these more than second of being coupled in the device and be coupled, wherein the 3rd quantum devices and the 4th quantum devices are inferior nearest neighbours in this dot matrix.

40. according to the described quantum processor of claim 39, further comprise a readout device, thereby this readout device is coupled on the quantum devices in these a plurality of quantum devices and makes this readout device can measure a kind of state of this quantum devices.

41. according to the described quantum processor of claim 40, wherein this readout device is arranged on the periphery of this dot matrix.

42. according to the described quantum processor of claim 39, further comprise a local bias unit, this this locality bias unit is coupled at least one quantum devices in these a plurality of quantum devices.

43. according to the described quantum processor of claim 39, wherein more than first of these a plurality of quantum devices and these are coupled device and are arranged to a kind of planar graph with this more than second individual at least one that is coupled device.

44., wherein in this dot matrix, embed a figure according to the described quantum processor of claim 39; And wherein

One group of node that quantum devices is this figure in these a plurality of quantum devices;

These more than first groups of being coupled in the device are coupled the value that device is endowed non-zero, thereby at least two quantum devices will organizing in the quantum devices according to this figure are coupled mutually; And

These more than second groups of being coupled in the device are coupled the value that device is endowed non-zero, thereby at least two quantum devices will organizing in the quantum devices according to this figure are coupled mutually.

45. according to the described quantum processor of claim 44, wherein this figure is an on-plane surface.

46. according to the described quantum processor of claim 44, wherein this figure is to be selected from by K ₅, K _3,3, a K ₅Expansion and a K _3.3The group formed of expansion.

47. according to the described quantum processor of claim 39, wherein

This first quantum devices and second quantum devices are by the determined neighbours nearest in this dot matrix of manhatton distance; And

The 3rd quantum devices and the 4th quantum devices are by in determined this dot matrix of manhatton distance nearest neighbours.

48. according to the described quantum processor of claim 39, wherein

Being coupled device these more than first at least two of being coupled in the device is arranged to and makes them non-intersect; And

Being coupled device these more than second at least two of being coupled in the device is arranged to they is intersected.

49. a quantum processor comprises:

A plurality of quantum devices;

One more than first are coupled device, wherein more than first of these a plurality of quantum devices and these are coupled device and form a kind of planar rectangular array, this planar rectangular array has a diagonal line, and wherein be coupled device these more than first at least one that are coupled in the device and be coupled with one second quantum devices in these a plurality of quantum devices with a kind of one first quantum devices of intensity in will this a plurality of quantum devices that be coupled, this is coupled intensity and has a value in a minimum negative is coupled the scope between the intensity of just being coupled of an intensity and a maximum;

One more than second are coupled device, wherein these more than second at least one that are coupled in the device be coupled device with a kind of be coupled intensity in will this a plurality of quantum devices one the 3rd quantum devices and one the 4th quantum devices in these a plurality of quantum devices be coupled, this is coupled intensity and has a value in a minimum negative is coupled intensity and one zero scope that is coupled between the intensity, and wherein the 3rd quantum devices and the 4th quantum devices are to arrange along the diagonal line of this planar rectangular array;

Readout device at least one quantum devices that is coupled in these a plurality of quantum devices; With

Local bias unit at least one quantum devices that is coupled in these a plurality of quantum devices.

50. according to the described quantum processor of claim 49, one of them figure embeds in this array; And wherein

One group of node that quantum devices is this figure in these a plurality of quantum devices;

These more than first groups of being coupled in the device are coupled the value that device is endowed non-zero

Thereby at least two quantum devices will organizing in the quantum devices according to this figure are coupled mutually; And

Thereby these more than second groups of being coupled in the device are coupled at least two quantum devices that value that device is endowed non-zero will organize in the quantum devices according to this figure and are coupled mutually.

51. according to the described quantum processor of claim 50, wherein this figure is an on-plane surface.

52. a quantum processor comprises:

(i) a plurality of qubit devices of each node of a dot matrix of formation; With

(ii) a plurality of devices that are coupled, wherein these a plurality of of being coupled in the device

One is coupled one first qubit devices of device in will these a plurality of qubit devices is coupled to one second qubit devices in these a plurality of qubit devices, and wherein this first qubit devices in this dot matrix and a kind of configuration of this second qubit devices are to be selected from by a kind of nearest neighbor configuration and a kind of nearest neighbor to dispose the group that is constituted.

53. a quantum processor comprises:

Be arranged in a plurality of qubit devices in the dot matrix;

One more than first are coupled device, and wherein these more than first are coupled in the device

One first be coupled device in will these a plurality of qubit devices one first quantum bit and one second qubit devices in this a plurality of qubit devices be coupled, wherein this first qubit devices is configured to neighbours nearest in this dot matrix with this second qubit devices; With

One more than second are coupled device, wherein these more than second first of being coupled in the device be coupled device in will this a plurality of qubit devices one the 3rd qubit devices and one the 4th qubit devices in this a plurality of qubit devices be coupled, and wherein the 3rd qubit devices is configured to inferior nearest neighbours in this dot matrix with the 4th qubit devices.

54. a method of using a quantum processor to determine the result of a computational problem, this method comprises:

(i) this quantum processor is initialised to an original state, wherein this quantum processor comprises a plurality of quantum devices and a plurality of device that is coupled, and all a pair of quantum devices in a plurality of quantum devices is coupled to this wherein to be coupled device in these a plurality of each that are coupled in the device, wherein this quantum processor is carried out initialization and comprises a state setting at least one quantum devices in these a plurality of quantum devices and set these a plurality of at least one that are coupled in the device and be coupled of device and be coupled intensity;

(ii) allow this quantum processor to calculate a kind of whole last current state, wherein should end last current state approach a kind of natural ground state of this computational problem;

Thereby a whole last current state of (iii) reading at least one quantum devices in these a plurality of quantum devices is determined the result of this computational problem; And

(iv) produce the carrier wave of the result's that a kind of embodiment comprises this computational problem a data-signal.

55. a computer system comprises:

Be used to import the device of the computational problem that will find the solution, wherein this computational problem is to be selected to have the problem of a complicacy NP-difficult problem and one by a problem with complicacy P, problem with complicacy NP, one and have the complicacy NP-group formed of problem completely;

Be used for this computational problem is mapped to a device on the quantum processor, wherein this quantum processor comprises qubit devices and is used for the device that the qubit devices with nearest neighbor and time nearest neighbor is coupled;

Use this quantum processor to be used to obtain the device of separating of this computational problem; Be used to export the device of separating of this computational problem; With

Be used for this is separated the device that transmits as a kind of data-signal that is embodied in a kind of carrier wave.

56. be embodied in a digital signal on a kind of carrier wave, comprise a corresponding value for each node in a plurality of nodes;

At least two nodes in these a plurality of nodes dot matrix that is a kind of nodes in the quantum processor wherein, and wherein each node in the dot matrix of this node all is a quantum devices; And

Wherein a value of at least one node individually or collectively represents one to a computational problem to separate in these a plurality of nodes, and the time after a figure representing this computational problem of this computational problem has been mapped at least one part of this dot matrix has been obtained understanding by calculating this quantum processor.

57. according to the described digital signal of claim 56, wherein this computational problem is to be selected to have the problem of a complicacy NP-difficult problem and one by a problem with complicacy P, problem with complicacy NP, one and have the complicacy NP-group formed of problem completely.

58. according to the described digital signal of claim 56, wherein these a plurality of nodes comprise at least 16 nodes.

59. according to the described digital signal of claim 56, wherein the analog value of a node in these a plurality of nodes is a kind of binary value.

60. according to the described digital signal of claim 56, wherein this figure is to be selected from by K ₅, K _3,3, K ₅One the expansion and K _3,3The group that constituted of expansion.

61. be embodied in a kind of digital signal of comprising on the carrier wave to the answer of a computational problem,

Wherein the answer to this computational problem is to determine by a value assessing each node in a plurality of nodes;

Wherein these a plurality of nodes are at least two nodes in the dot matrix of a node in the quantum processor; And

Wherein a value of at least one node is to determine after this quantum processor of calculation time after a figure representing this computational problem has been mapped at least one part of this dot matrix in these a plurality of nodes.

62. be embodied on the carrier wave, comprise a kind of digital signal to a figure of a computational problem will finding the solution with a quantum processor,

Wherein this quantum processor comprises the dot matrix of a quantum devices;

Wherein the figure of this computational problem that will find the solution comprise a plurality of nodes and, for each respective nodes in these a plurality of nodes, comprise between another node in an initial value of respective nodes and this respective nodes and these a plurality of nodes one corresponding be coupled constant; And

Wherein the figure of this computational problem that will find the solution is configured to make it can be mapped on this dot matrix of this quantum processor.

63. be embodied on the carrier wave, comprise a kind of digital signal to a computational problem will finding the solution with a quantum processor,

Wherein this quantum processor comprises a kind of dot matrix of quantum devices;

Wherein this computational problem that will find the solution is switched on the figure, this figure comprise a plurality of nodes and, for each respective nodes in these a plurality of nodes, comprise between another node in an initial value of this respective nodes and this respective nodes and these a plurality of nodes one corresponding be coupled constant; And

Wherein the figure of this computational problem that will find the solution is configured to make it can be mapped on this dot matrix of this quantum processor.

64. a graphical user interface is used to obtain to the separating an of computational problem, this graphical user interface comprises:

(A) one first viewing area is used for indicating the digital signal on the carrier wave of being embodied in of a corresponding value when having received each node that comprises a plurality of nodes;

Wherein these a plurality of nodes are at least two nodes in the node dot matrix in the quantum processor, and wherein each node in this node dot matrix all is a quantum devices; And

Wherein a value of at least one node in these a plurality of nodes individually or is collectively represented the separating of this computational problem, and a time after a figure representing this computational problem has been mapped at least one part of this dot matrix of this computational problem is by calculating this quantum processor with through obtaining understanding; (B) one second viewing area is used to show separating this computational problem.

65. according to the described graphical user interface of claim 64, wherein this computational problem is to be selected to have the problem of a complicacy NP-difficult problem and one by a problem with complicacy P, problem with complicacy NP, one and have the complicacy NP-group formed of problem completely.

66. according to the described graphical user interface of claim 64, wherein these a plurality of nodes comprise at least 16 nodes.

67. according to the described graphical user interface of claim 64, wherein the analog value of a node in these a plurality of nodes is a kind of binary value.

68. according to the described graphical user interface of claim 64, wherein this figure is to be selected from by K ₅, K _3,3, K ₅One the expansion and K _3,3The group that constituted of expansion.

69. a graphical user interface is used to obtain to the separating an of computational problem, this graphical user interface comprises:

(A) one first viewing area is used to indicate when received a digital signal that is embodied in a carrier wave, has comprised an answer of this computational problem;

Wherein the answer to this computational problem is to determine by a value assessing at least one node in a plurality of nodes;

Wherein these a plurality of nodes are at least two nodes in the node dot matrix in the quantum processor, and wherein each node in this node dot matrix all is a quantum devices; And

Wherein this value of at least one node in these a plurality of nodes is to be determined after this quantum processor value of calculation time after a figure representing this computational problem has been mapped at least one part of this dot matrix; (B) one second viewing area is used to show separating this computational problem.

70. according to the described graphical user interface of claim 69, wherein this carrier wave is to receive from a remote computer.

71. a graphical user interface is used to obtain to the separating an of computational problem, this graphical user interface comprises:

(A) one first viewing area is used to indicate when produced the digital signal on the carrier wave of being embodied in that comprises this computational problem that will be found the solution by a quantum processor;

Wherein this quantum processor comprises the dot matrix of a quantum devices;

Wherein this computational problem that will find the solution comprise a plurality of nodes and, for each node corresponding in these a plurality of nodes, comprise between another nodes of an initial value being used for this node corresponding and this respective nodes and these a plurality of nodes one corresponding be coupled constant; And

Wherein this computational problem that will find the solution is configured to make it can be mapped on the dot matrix of this quantum processor; With

(B) one second viewing area is used for receiving showing separating this computational problem after the separating of this computational problem.

72. according to the described graphical user interface of claim 71, wherein this digital signal is sent to a remote computer of communicating by letter with this quantum processor.

73. according to the described graphical user interface of claim 71, wherein this digital signal is sent straight to this quantum processor.

74., be to receive wherein from a remote computer of communicating by letter with this quantum processor to separating of this computational problem according to the described graphical user interface of claim 71.

75., be directly to receive wherein from this quantum processor to separating of this computational problem according to the described graphical user interface of claim 71.

76. 7 described graphical user interface according to Claim 8, wherein this computational problem that will find the solution is to be expressed as a kind of figure in this digital signal.

77. a computing system comprises:

A local computer;

A remote computer;

A long-range quantum processor of communicating by letter with this remote computer, this quantum place

The reason device comprises:

(i) a plurality of quantum devices, wherein each quantum devices in these a plurality of quantum devices is a node of a dot matrix, and wherein one first quantum devices in these a plurality of quantum devices has a kind of first base state and one second base state;

With

(ii) a plurality of devices that are coupled, wherein these a plurality of first of being coupled in the device are coupled one first quantum devices of device in will these a plurality of quantum devices and are coupled with one second quantum devices in this a plurality of quantum devices, and wherein a kind of configuration of this first quantum devices in this dot matrix and this second quantum devices is to be selected from by a nearest neighbor configuration and an inferior nearest neighbor to dispose the group that is constituted; Wherein

This local computer is configured to send the computational problem that will find the solution to this remote computer; And

This remote computer is configured to the answer of this local computer transmission to this computational problem.

78. a computer system is used for determining the result of a computational problem, this computer system comprises:

A local computer;

A remote computer; With

An analog processor; Wherein

This local computer comprises a CPU (central processing unit) and is coupled a storer of this CPU (central processing unit), this memory stores:

A user interface module comprises the instruction that is used to define this computational problem;

A mapper module comprises the instruction of a kind of mapping that is used to produce this computational problem; With

A transmission module comprises the instruction that is used for sending to this remote computer this mapping;

This remote computer comprises a central authorities calculation unit and is coupled to a storer of this central processing unit, this memory stores:

One receives module, comprises the instruction that is used for receiving from this local computer this mapping;

An analog processor NIM comprises the instruction that is used for transmitting to this analog processor this mapping; And

This analog processor comprises a plurality of quantum devices and a plurality of device that is coupled, and wherein this mapping comprises that a plurality of initialization values of at least one quantum devices in these a plurality of quantum devices and this a plurality of at least one that are coupled in the device are coupled a plurality of initialization values of device, and wherein these a plurality of of being coupled in the device relevant quantum devices of being coupled the correspondence of device in will these a plurality of quantum devices is connected among the inferior nearest neighbours of nearest neighbours of this relevant quantum devices and this relevant quantum devices one at least one.

79. a computer system is used for determining the result of a computational problem, this computer system comprises:

A local computer;

A remote computer; With

An analog processor; Wherein

This local computer comprises a CPU (central processing unit) and is coupled a storer of this CPU (central processing unit), this memory stores:

A user interface module comprises the instruction that is used to define this computational problem;

A transmission module comprises the instruction that is used for sending to this remote computer this computational problem;

This remote computer comprises a CPU (central processing unit) and is coupled to a storer of this central processing unit, this memory stores:

One receives module, comprises the instruction that is used for receiving from this local computer this computational problem;

A mapper module comprises the instruction of a kind of mapping that is used to produce this computational problem; With

An analog processor NIM comprises the instruction that is used for transmitting to this analog processor this mapping; And

This analog processor comprises a plurality of quantum devices and a plurality of device that is coupled, and wherein this mapping comprises that a plurality of initialization values of at least one quantum devices in these a plurality of quantum devices and this a plurality of at least one that are coupled in the device are coupled a plurality of initialization values of device, and wherein this a plurality of of being coupled in the device relevant quantum devices of being coupled the correspondence of device in will these a plurality of quantum devices is coupled among the inferior nearest neighbours of nearest neighbours of this corresponding quantum devices and this corresponding quantum devices one at least one.

Images