Optical quantum computing system, optical quantum chip, and optical quantum computer

Abstract

The application provides an optical quantum computing system, an optical quantum chip and an optical quantum computer, and relates to the technical field of quantum computing. The system comprises a plurality of optical waveguides, a plurality of spatial processing modules and a plurality of temporal processing modules are arranged on the plurality of optical waveguides, wherein: the spatial processing module is used for performing quantum state transformation on input optical quantum information to realize spatial parallel processing; the temporal processing module is used for performing time adjustment processing on the input optical quantum information to realize time multiplexing; the spatial processing modules and the temporal processing modules are alternately arranged between the spatial processing modules and the temporal processing modules, and the adjacent spatial processing modules and the temporal processing modules are connected through the plurality of optical waveguides. The embodiment of the application can realize the dual effective expansion of the scale and the depth of quantum computing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of quantum computing, in particular to an optical quantum computing system, an optical quantum chip and an optical quantum computer. BACKGROUND

[0002] With the development of quantum computing technology, quantum computing can exhibit potential far beyond classical computers in processing certain computing problems. At present, there are various physical systems to realize quantum computing, and the mainstream technical paths include superconducting quantum computing, ion trap quantum computing, optical quantum computing, etc. Compared with other physical systems, optical quantum computing has advantages such as long coherence time, room temperature operation, high speed transmission, etc.

[0003] In order to build a general-purpose quantum computer capable of executing complex algorithms, it is necessary to be able to prepare and manipulate a large-scale quantum bit array. In the field of optical quantum computing, the existing technology mainly realizes it through the following two paths, one is a pure spatial parallel scheme, and the other is a pure time multiplexing scheme. Among them, the pure spatial parallel scheme is limited by the bottlenecks such as physical integration density, manufacturing process and control complexity, and it is difficult to realize effective expansion of the computing scale; while the pure time multiplexing scheme is limited by the computing delay caused by serial processing, cumulative photon loss and the limitations of topology construction, and it is difficult to support complex quantum algorithms with high depth. SUMMARY

[0004] Therefore, the embodiments of the present application are dedicated to providing an optical quantum computing system, an optical quantum chip and an optical quantum computer to solve the problem that the computing scale and the computing depth of quantum computing are limited in the prior art.

[0005] In a first aspect, an embodiment of the present application provides an optical quantum computing system, comprising a plurality of optical waveguides, a plurality of spatial processing modules and a plurality of time processing modules are arranged on the plurality of optical waveguides, wherein: the spatial processing module is configured to perform quantum state transformation on input optical quantum information to realize spatial parallel processing; the time processing module is configured to perform time adjustment processing on the input optical quantum information to realize time multiplexing; the spatial processing module and the time processing module are alternately arranged between them, and the adjacent spatial processing module and the time processing module are connected through the plurality of optical waveguides.

[0006] In combination with the first aspect, in some implementations of the first aspect, the spatial processing module comprises at least one spatial interference unit, wherein: the spatial interference unit is connected with at least two optical waveguides of the plurality of optical waveguides, and is configured to perform interference processing on the optical quantum information in the at least two optical waveguides.

[0007] With reference to the first aspect, in some implementations of the first aspect, at least one of the spatial interference units is connected between any two adjacent optical waveguides in the plurality of optical waveguides.

[0008] With reference to the first aspect, in some implementations of the first aspect, the spatial processing module comprises a plurality of phase adjustment units, wherein: at least one phase adjustment unit is correspondingly arranged on each optical waveguide; and the phase adjustment unit is configured to perform phase adjustment processing on the optical quantum information in the optical waveguide.

[0009] With reference to the first aspect, in some implementations of the first aspect, the temporal processing module comprises a plurality of time delay units, wherein: at least one time delay unit is correspondingly arranged on each optical waveguide; and the time delay unit is configured to perform time delay processing on the optical quantum information in the optical waveguide.

[0010] With reference to the first aspect, in some implementations of the first aspect, the time delay unit comprises an input end and an output end, the input end comprises a first input end and a second input end, and the output end comprises a first output end and a second output end, wherein: the first input end of the time delay unit is connected with the optical waveguide, configured to receive target optical quantum information input into the time delay unit from the optical waveguide; the second input end of the time delay unit is connected with the second output end through a delay line, constituting a time delay path, so as to realize time delay processing on the target optical quantum information when the target optical quantum information passes through the time delay path; and the first output end of the time delay unit is connected with the optical waveguide, configured to output the target optical quantum information after time delay processing to the optical waveguide.

[0011] With reference to the first aspect, in some implementations of the first aspect, the spatial processing module further comprises a plurality of phase initialization units, wherein: at least one phase initialization unit is correspondingly arranged on each optical waveguide and connected with a signal input end through the phase initialization unit; and the phase initialization unit is configured to perform initialization phase modulation processing on a photon signal input from the signal input end, and output optical quantum information obtained after initialization phase modulation processing to the optical waveguide.

[0012] With reference to the first aspect, in some implementations of the first aspect, the spatial processing module further comprises a plurality of phase compensation units, wherein: at least one phase compensation unit is correspondingly arranged on each optical waveguide and connected with a signal output end through the phase compensation unit; and the phase compensation unit is configured to perform phase compensation processing on the optical quantum information output from the optical waveguide, and output optical quantum information obtained after phase compensation processing to the signal output end.

[0013] In a second aspect, an embodiment of the present application provides a photonic quantum chip integrated with the photonic quantum computing system according to the first aspect.

[0014] In a third aspect, an embodiment of the present application provides a photonic quantum computer configured with the photonic quantum computing system according to the first aspect.

[0015] In the present application, by arranging a plurality of spatial processing modules and time processing modules alternately on a plurality of optical waveguides, quantum state transformation is performed on photonic quantum information by the spatial processing modules to realize spatial parallel processing, and time adjustment processing is performed on the photonic quantum information by the time processing modules to realize time multiplexing, so that spatial parallel processing and time multiplexing can be organically combined, thereby being able to combine the advantages of both and avoid the disadvantages, realizing the dual effective expansion of quantum computing scale and computing depth. BRIEF DESCRIPTION OF DRAWINGS

[0016] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description of embodiments of the present application taken in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of embodiments of the present application and are incorporated in and constitute a part of the specification, illustrate embodiments of the present application and serve to explain the present application, but do not limit the present application. In the drawings, the same reference numerals refer to the same components or steps throughout the drawings.

[0017] Figure 1 Fig. 1 shows a structural schematic diagram of a photonic quantum computing system according to an embodiment of the present application.

[0018] Figure 2 Fig. 2 shows a structural schematic diagram of a spatial processing module according to an embodiment of the present application.

[0019] Figure 3 Fig. 3 shows a structural schematic diagram of a spatial processing module according to another embodiment of the present application.

[0020] Figure 4 Fig. 4 shows a structural schematic diagram of a spatial interference unit according to an embodiment of the present application.

[0021] Figure 5 Fig. 5 shows a structural schematic diagram of a spatial processing module according to yet another embodiment of the present application.

[0022] Figure 6 Fig. 6 shows a structural schematic diagram of a time processing module according to an embodiment of the present application.

[0023] Figure 7 Fig. 7 shows a structural schematic diagram of a time delay unit according to an embodiment of the present application.

[0024] Figure 8Fig. 1 shows a structural schematic diagram of a spatial processing module according to an embodiment of the present application.

[0025] Figure 9 Fig. 2 shows a structural schematic diagram of a spatial processing module according to another embodiment of the present application.

[0026] Figure 10 Fig. 3 shows a structural schematic diagram of a photonic quantum computing system according to an embodiment of the present application.

[0027] Figure 11 Fig. 4 shows a structural schematic diagram of a photonic quantum computer according to an embodiment of the present application.

[0028] Figure 12 Fig. 5 shows a flowchart of a method for constructing a photonic quantum computing system according to an embodiment of the present application.

[0029] Figure 13 Fig. 6 shows a structural schematic diagram of a device for constructing a photonic quantum computing system according to an embodiment of the present application.

[0030] Figure 14 Fig. 7 shows a structural schematic diagram of a photonic quantum computer according to another embodiment of the present application. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0032] In addition, in order to better illustrate the present application, numerous specific details are given in the specific embodiments below. Those skilled in the art should understand that the present application can also be implemented without some specific details. In some examples, methods and means familiar to those skilled in the art are not described in detail, in order to highlight the main ideas of the present application.

[0033] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0034] In addition, if the terms "first", "second", "third" and "fourth" etc. are used only to distinguish description, and cannot be understood as indicating or implying relative importance.

[0035] Quantum computing is a new computing mode that regulates quantum information units to perform computation according to the laws of quantum mechanics. It can exhibit the potential far beyond classical computers in processing certain computing problems by using the superposition and entanglement of quantum bits. At present, there are various physical systems to realize quantum computing, including superconducting quantum computing, ion trap quantum computing, optical quantum computing, etc. Compared with other physical systems, optical quantum computing has advantages such as long coherence time, room temperature operation, and high-speed transmission. Quantum bits in optical quantum computing are usually encoded in the physical degrees of freedom of photons, such as polarization, path, time, frequency, or orbital angular momentum.

[0036] In order to build a general-purpose quantum computer capable of performing complex algorithms, it is necessary to be able to prepare and manipulate a large-scale quantum bit array. In the field of optical quantum computing, the existing technology mainly realizes it through the following two paths. One is a pure spatial parallel scheme, which mainly uses phase shifters and other optical elements to build an optical network containing a large number of parallel processing units in a spatially parallel manner on a photonic integrated chip, such as the Clements and Reck architecture. However, the number of required optical elements increases polynomially with the size of the problem, which leads to a sharp deterioration of chip area, wiring density, and electromagnetic crosstalk between elements, making it extremely difficult to expand to the level of millions of bits in a physical sense. The other is a pure time multiplexing scheme, which generates a long sequence of photon pulses in the time dimension. By using high-speed optical switches and other optical elements, interactions between photon pulses at different time points can be generated, thereby weaving a one-dimensional (1D) or quasi-two-dimensional (2D) entangled state in the time axis. The problem of this scheme is the limitation of computation delay and depth. The time multiplexing scheme is essentially a serial processing. To make the first photon interact with the first thousand photons in the sequence, the first photon must wait in a delay loop, which introduces a huge computation delay. The unavoidable loss of photons in the fiber will accumulate, which will seriously limit the depth of the executable quantum algorithm.

[0037] In summary, in the existing technology, the pure spatial parallel scheme is limited by the polynomial growth bottleneck of physical integration density, manufacturing process, and control complexity, making it difficult to effectively expand the computing scale, thereby forming a “width” bottleneck; while the pure time multiplexing scheme is limited by the computation delay caused by serial processing, the accumulation of photon loss, and the limitations of topology construction, making it difficult to support complex quantum algorithms with high depth, thereby forming a “depth” bottleneck. The inherent defects of these two schemes seriously hinder the final realization of scalable, high-performance, fault-tolerant optical quantum computers. Therefore, there is an urgent need in the art for a new computing system that can combine the strengths of both and avoid their weaknesses, to realize the dual effective expansion of quantum computing scale and computing depth.

[0038] Based on this, the present application aims to provide a brand-new scalable space-time hybrid optical quantum computing system, and introduces a software and hardware collaborative design concept to systematically break through the above-mentioned dual restrictions of "width" and "depth". Specifically, the optical quantum computing system provided by the embodiment of the present application can realize the dual reuse of quantum computing in the spatial dimension and the time dimension by alternately deploying spatial processing modules and time processing modules on multiple parallel optical waveguides. In addition, the system construction method provided by the embodiment of the present application can automatically search for the optimal system parameters in the optical quantum computing system according to the target transformation matrix capable of solving the target quantum computing problem, thereby efficiently and accurately mapping the target transformation matrix in the optical quantum computing system. This software and hardware collaborative design method not only greatly reduces the photon loss and computing delay, but also provides unprecedented flexibility and programmability for constructing high-dimensional entanglement and executing complex quantum algorithms, and finally opens up a brand-new technical path for realizing truly scalable, high-performance fault-tolerant optical quantum computers.

[0039] The following will be described in detail Figures 1 to 10 The optical quantum computing system provided by the present application is described in detail.

[0040] Figure 1 As shown in the structural schematic diagram of the optical quantum computing system provided by an embodiment of the present application. The optical quantum computing system can be integrated on a chip, and each module and circuit in the optical quantum computing system can be manufactured on a single substrate at one time by using a monolithic integrated process, wherein the material of the substrate may, for example, include silicon, lithium niobate, silicon nitride, etc.

[0041] As Figure 1 shown, the optical quantum computing system includes multiple optical waveguides 11, and multiple spatial processing modules 12 and multiple time processing modules 13 are arranged on the multiple optical waveguides 11, wherein: the spatial processing module 12 is configured to perform quantum state transformation on the input optical quantum information to realize spatial parallel processing; the time processing module 13 is configured to perform time adjustment processing on the input optical quantum information to realize time multiplexing; the spatial processing module 12 and the time processing module 13 are alternately arranged, and the spatial processing module 12 and the time processing module 13 are connected through the multiple optical waveguides 11.

[0042] In some examples, the optical waveguide 11 can be a line that can transmit optical quantum information, and the material of the line may, for example, be a low-loss optical fiber waveguide, a high-density integrated silicon-based optical waveguide, a thin film lithium niobate waveguide, etc. The optical quantum computing system provided by the embodiment of the present application can include multiple parallel optical waveguides 11, and the spatial processing module 12 and the time processing module 13 arranged alternately on the optical waveguide 11, so as to realize the purpose of multiplexing in the space and time dimensions.

[0043] In some examples, the spatial processing module 12 can be a reconfigurable linear optical interference network, the core function of which is to implement an input multi-channel parallel processing in space to achieve a feature transformation, such as a unitary transformation. The spatial processing module 12 is capable of performing quantum state transformation processing on the optical quantum information input into the module, where the quantum state transformation processing includes but is not limited to phase modulation, multi-mode interference, etc.

[0044] In some specific examples, the spatial processing module 12 can include a plurality of optical elements, which can be tunable elements, such as tunable phase modulators, tunable beam splitters, or Mach-Zehnder interferometers (MZIs), etc. These optical elements can be arranged in a specific topology, such as a single-layer arrangement or a multi-layer arrangement. The element setting network architecture in the spatial processing module 12 can theoretically achieve a feature transformation corresponding to a transformation matrix of a specific dimension, such as an N N unitary matrix, by precisely setting the parameter values inside each element. Based on this, the element setting network architecture in the spatial processing module 12 can be designed as a single-layer arrangement structure to achieve a basic transformation, or as a cascaded structure of a multi-layer arrangement to achieve a more complex and higher fidelity transformation.

[0045] It should be noted that the tunable elements in the spatial processing module 12 described above can all use active modulation. Specifically, for example, a thermo-optic phase modulator can be used, which integrates a micro-resistance heater above or on the side of the waveguide, and changes the effective refractive index of the waveguide by using the thermo-optic effect of the material to achieve phase modulation. For another example, an electro-optic phase modulator can be used, which uses the electro-optic effect of the material to change the refractive index by applying an external electric field to achieve high-speed phase modulation. Among them, the electro-optic modulation has a faster response speed than the thermo-optic modulation.

[0046] In addition, in some examples, the temporal processing module 13 can be a module capable of performing temporal adjustment processing on the optical quantum information input into the module, where the temporal adjustment processing can be, for example, time delay processing. The temporal processing module 13 can include, for example, a group of parallel tunable delay structures, so that a programmable time delay processing can be independently introduced for each optical waveguide corresponding to a line. Among them, the tunable delay structure can be an active programmable optical switch delay structure based on a tunable element, such as an MZI.

[0047] Based on the above description of the spatial processing module 12 and the temporal processing module 13, the quantum information processing system 1 can be implemented as shown in FIG. 2. As shown in FIG. 2, the quantum information processing system 1 can include a quantum information input module 21, a quantum information processing module 22, and a quantum information output module 23. Figure 1The illustrated optical quantum computing system can include a plurality of spatial processing modules 12 and a plurality of temporal processing modules 13 monolithically integrated on the same substrate, which are arranged alternately to form a processing core. The number of alternations of the spatial processing modules 12 and the temporal processing modules 13 can be determined according to the complexity of the quantum computation corresponding to the quantum computing problem to be solved. The higher the complexity, the more the number of alternations can be. In addition, it should be noted that the network architecture of the elements used in each spatial processing module 12 can be different, and the control parameters corresponding to the tunable elements can also be different. The tunable delay structure used in each temporal processing module 13 can also be different, and the delay time corresponding to the tunable delay structure can also be different, which is not limited here.

[0048] Exemplarily, the photon signal emitted from the light source can be input from the left signal input end of the optical waveguide 11 to the system, sequentially pass through the spatial processing module 12, the temporal processing module 13, the spatial processing module 12, the temporal processing module 13, etc., and finally output from the right signal output end after passing through the spatial processing module 12. The output information can be the optical quantum information obtained after the optical quantum computing system processes the input photon signal in a specified order.

[0049] In addition, the arrangement of the spatial processing module 12 and the temporal processing module 13 alternately arranged can be, in addition to the arrangement as shown in Figure 1 In some examples, the arrangement can start with the temporal processing module 13 and end with the spatial processing module 12. In addition, in some examples, the arrangement can start with the spatial processing module 12 and end with the temporal processing module 13. In addition, in some examples, the arrangement can start with the temporal processing module 13, pass through the alternation structure of the spatial processing module 12 and the temporal processing module 13, and end with the temporal processing module 13.

[0050] In this way, by arranging a plurality of spatial processing modules and temporal processing modules alternately on a plurality of optical waveguides, the spatial processing modules are used to perform quantum state transformation on the optical quantum information to realize spatial parallel processing, and the temporal processing modules are used to perform time adjustment processing on the optical quantum information to realize time multiplexing, so that the spatial parallel processing and the time multiplexing can be organically combined, thereby being able to combine the advantages of the two and avoid the disadvantages, realizing the dual effective expansion of the scale and the depth of quantum computing.

[0051] Based on this, in some embodiments, as shown in Figure 2 and Figure 3 The spatial processing module 12 includes at least one spatial interference unit 121, wherein the spatial interference unit 121 is connected with at least two optical waveguides in the plurality of optical waveguides, and is used for interference processing on the optical quantum information in the at least two optical waveguides.

[0052] In some examples, the spatial interference unit 121 can be a unit capable of multi-mode interference processing of multi-channel optical quantum information, which can be, for example, an MZI, an adjustable beam splitter, etc. For example, taking an MZI as an example, the connection mode between the MZI and the two optical waveguides can be as shown in FIG. 11A, for example, the input upper end and the output upper end of the MZI 1211 are connected with the first optical waveguide 111, and the input lower end and the output lower end of the MZI 1211 are connected with the second optical waveguide 112. After the optical quantum information in the first optical waveguide 111 and the optical quantum information in the second optical waveguide 112 are input into the MZI 1211, interference processing between the two optical quantum information can be realized, and the optical quantum information obtained after the interference processing is output to the first optical waveguide 111 and the second optical waveguide 112, respectively. Figure 4

[0053] Based on this, for the arrangement mode of the spatial interference unit 121 in the spatial processing module 12, for example, the arrangement mode as shown in FIG. 11B can be used, that is, each two adjacent optical waveguides can be connected with one spatial interference unit 121, for example, the first optical waveguide 111 and the second optical waveguide 112 can be connected with one spatial interference unit 121, the third optical waveguide 113 and the fourth optical waveguide 114 can be connected with one spatial interference unit 121, and so on. Figure 2

[0054] In other embodiments, in order to improve the efficiency of quantum computing, the spatial interference unit 121 in the above-mentioned spatial processing module 12 can also be arranged in a parity interleaving topology, that is, at least one spatial interference unit 121 is connected between any two adjacent optical waveguides in the plurality of optical waveguides.

[0055] Based on this, for the arrangement mode of the spatial interference unit 121 in the spatial processing module 12, for example, the arrangement mode as shown in FIG. 11C can also be used, that is, any two adjacent optical waveguides are connected with one spatial interference unit 121, for example, the first optical waveguide 111 and the second optical waveguide 112 can be connected with one spatial interference unit 121, the second optical waveguide 112 and the third optical waveguide 113 can be connected with one spatial interference unit 121, the third optical waveguide 113 and the fourth optical waveguide 114 can be connected with one spatial interference unit 121, and so on. Figure 3 It should be noted that in addition to the single-layer arrangement mode shown in the above

[0056] and Figure 2 , a plurality of times of repeated arrangement along the optical waveguide can also be used to form a multi-layer arrangement mode of a multi-layer cascading structure, so as to realize more complex and higher fidelity feature transformation. Figure 3

[0057] ​​​In addition, in some embodiments, as shown in Figure 5 The spatial processing module includes a plurality of phase adjustment units 122, wherein at least one phase adjustment unit 122 is arranged on each optical waveguide 11, and the phase adjustment unit 122 is configured to perform phase adjustment on the optical quantum information in the optical waveguide 11.

[0058] In some examples, the phase adjustment unit 122 can be a unit capable of adjusting the phase of the optical quantum information, which can be, for example, an adjustable phase modulator.

[0059] For example, as shown in Figure 5 Each optical waveguide 11 can be provided with one phase adjustment unit 122, and each phase adjustment unit 122 can perform phase adjustment on the optical quantum information in the optical waveguide.

[0060] It should be noted that the spatial interference unit described in the foregoing embodiments and the phase adjustment unit described in the present embodiment can exist alone in the optical quantum computing system, or can exist simultaneously in the optical quantum computing system, and the present disclosure does not limit the same.

[0061] In addition, in some embodiments, as shown in Figure 6 The time processing module 13 includes a plurality of time delay units 131, wherein at least one time delay unit 131 is arranged on each optical waveguide 11, and the time delay unit 131 is configured to perform time delay on the optical quantum information in the optical waveguide 11.

[0062] In some examples, the time delay unit can be a unit capable of performing time delay on the optical quantum information, which can be, for example, a tunable delay line, which can be a MZI-based active programmable optical switch delay unit. The delay time corresponding to the time delay unit can be zτ, wherein τ represents the time interval between adjacent photon pulses when a photon signal is input into the optical waveguide, and z can be an integer greater than or equal to 0.

[0063] For example, as shown in Figure 6 Each optical waveguide 11 can be provided with one time delay unit 131, and each time delay unit 131 can perform time delay on the optical quantum information in the optical waveguide, wherein the delay time corresponding to the time delay unit 131 arranged on different optical waveguides can be different.

[0064] Based on this, in some specific embodiments, the time delay unit includes an input terminal and an output terminal. The input terminal includes a first input terminal and a second input terminal, and the output terminal includes a first output terminal and a second output terminal. Specifically: the first input terminal of the time delay unit is connected to an optical waveguide and is used to receive target quantum information input to the time delay unit from the optical waveguide; the second input terminal and the second output terminal of the time delay unit are connected via a delay line to form a delay path, so as to realize time delay processing of the target quantum information when it passes through the delay path; the first output terminal of the time delay unit is connected to the optical waveguide and is used to output the time-delayed target quantum information to the optical waveguide.

[0065] In some examples, the time delay unit can be an active, programmable optical switch delay unit including a tunable element. This tunable element may have two inputs and two outputs, with one input connected to one output to form a delay path. Specifically, a delay line can be used, such as a low-loss fiber waveguide, a high-density integrated silicon-based optical waveguide, or a spatial optical path. Furthermore, the target optical quantum information can be any optical quantum information transmitted in the optical waveguide.

[0066] For example, taking an MZI-based active, programmable optical switch delay unit as an example, such as... Figure 7 As shown, in the time delay unit 131, the first input terminal of the MZI is the lower input terminal 1311, the first output terminal of the MZI is the lower output terminal 1312, the second input terminal of the MZI is the upper input terminal 1313, and the second output terminal of the MZI is the upper output terminal 1314. The upper input terminal 1313 and the upper output terminal 1314 of the MZI are connected by a low-loss optical fiber waveguide to form a delay path, and their optical path difference can be the time interval of one or more adjacent input photon pulses. The lower input terminal 1311 of the MZI is used to receive the photon information input from the optical waveguide 11, and the lower output terminal 1312 of the MZI is used to output the time-delayed photon information to the optical waveguide 11. Furthermore, under the above connection method, the MZI, as a time delay unit, can also perform interference processing on the quantum information input from the second input terminal and the quantum information present in the delay line. In this way, the two internal phase shifters integrated on the MZI interferometer arm can precisely control the interference ratio and phase of the input pulse signal by applying control voltage or current, thereby realizing the control of the interference result.

[0067] Additionally, in some embodiments, such as Figure 8As shown, the spatial processing module 12 further comprises a plurality of phase initialization units 123, wherein at least one phase initialization unit 123 is correspondingly arranged on each optical waveguide 11 and connected to the signal input end 115 through the phase initialization unit 123; the phase initialization unit 123 is configured to perform initialization phase modulation on the photon signal input from the signal input end 115, and output the quantum information of light obtained through the initialization phase modulation to the optical waveguide 11.

[0068] In some examples, the signal source can emit photon signals to the signal input end 115 of each optical waveguide 11, for example, emit photon pulses with a time interval of τ to each optical waveguide 11, and a total of k photon pulses can be emitted. The phase initialization unit 123 can be arranged at the starting position of the optical waveguide 11, and is configured to perform initial and independently adjustable initialization phase modulation on the photon pulse carrying quantum information to prepare a specific input quantum state. The phase initialization unit 123 can be, for example, an adjustable phase modulator.

[0069] Exemplarily, as shown in FIG. 1, Figure 8 A phase initialization unit 123 can be arranged after the signal input end 115 of each optical waveguide 11, and each phase initialization unit 123 can perform independently adjustable initialization phase modulation on the quantum information of light in the optical waveguide to prepare a specific input quantum state.

[0070] In addition, in some embodiments, as shown in FIG. 1, Figure 9 As shown, the spatial processing module 12 further comprises a plurality of phase compensation units 124, wherein at least one phase compensation unit 124 is correspondingly arranged on each optical waveguide 11 and connected to the signal output end 116 through the phase compensation unit 124; the phase compensation unit 124 is configured to perform phase compensation on the quantum information of light output from the optical waveguide 11, and output the quantum information of light obtained through the phase compensation to the signal output end 116.

[0071] In some examples, the phase compensation unit 124 can be arranged at the end position of the optical waveguide 11, and is configured to perform phase compensation on the quantum state of the quantum information of light finally output from the optical waveguide 11 to meet the requirements of subsequent single-photon detection or further processing. The phase compensation unit 124 can be, for example, an adjustable phase modulator.

[0072] Exemplarily, as shown in FIG. 1, Figure 9 A phase compensation unit 124 can be arranged before the signal output end 116 of each optical waveguide 11, and each phase compensation unit 124 can perform independently adjustable phase compensation on the quantum information of light finally output from the optical waveguide to meet the requirements of subsequent single-photon detection or further processing.

[0073] Based on the above, on the basis of the various embodiments described above, the following will be combined with Figure 10 Take a specific example to better illustrate the structure of the optical quantum computing system designed in the present application.

[0074] As Figure 10 The optical quantum computing system shown can include n optical waveguides (n is an integer greater than 2), the left side is the signal input end, and the right side is the signal output end. From the signal input end of the n optical waveguides, an input phase modulation layer 91, an initial time delay layer 92, a spatial unitary transformation layer 93, a time delay layer 94, and an output phase modulation layer 95 are respectively arranged.

[0075] Here, the input phase modulation layer 91 belongs to the spatial processing module, and each first phase modulator included in the input phase modulation layer 91 is the phase initialization unit mentioned in the various embodiments described above. The input phase modulation layer 91 can directly act on the input n-way photonic pulse signal, and independently and adjustably modulate the phase of the photonic pulse input into the n optical waveguides to prepare a specific input quantum state.

[0076] The initial time delay layer 92 belongs to the time processing module, and each tunable delay line included in the initial time delay layer 92 is the time delay unit mentioned in the various embodiments described above. The delay time corresponding to each tunable delay line in the initial time delay layer 92 can be independently and accurately set to one or more time intervals τ of adjacent input photonic pulses. The function of the initial time delay layer 92 is to preliminarily map the quantum bits purely encoded in space to the time-space mixed degree of freedom, and to prepare for subsequent cross-time slice interaction.

[0077] The spatial unitary transformation layer 93 belongs to the spatial processing module, and the MZI included in the spatial unitary transformation layer 93 is the spatial interference unit mentioned in the various embodiments described above. The spatial unitary transformation layer 93 can include a tunable MZI network. The MZI network is arranged in an odd-even interleaved topology, and can realize n unitary matrix transformation between the optical quantum information transmitted by the n parallel optical waveguides. This is the core layer for realizing quantum interference and information exchange between spatial modes.

[0078] The time delay layer 94 belongs to the time processing module, each tunable delay line included in the time delay layer 94 is the time delay unit mentioned in each of the above embodiments. The delay time corresponding to each tunable delay line in the time delay layer 94 can also be independently and accurately set to the time interval τ of one or more input adjacent photon pulses. The role of the time delay layer 94 is to perform a time "shift" on the photon pulses in each spatial mode after each spatial transformation, thereby ensuring that the photon pulses from different time slices can enter the same MZI for interference when entering the spatial unitary transformation layer in the next cycle, realizing quantum interaction across time.

[0079] The output phase modulation layer 95 belongs to the spatial processing module, each second phase modulator included in the output phase modulation layer 95 is the phase compensation unit mentioned in each of the above embodiments. The role of the output phase modulation layer 95 is to perform phase compensation on the quantum state of the final output optical quantum information to meet the needs of subsequent single-photon detection or further processing.

[0080] It should be noted that the core computing module of the optical quantum computing system can include the above-mentioned spatial unitary transformation layer 93 and the time delay layer 94, which can be cyclically arranged in the optical waveguide, that is, the initialized optical quantum information can pass through the spatial unitary transformation layer 93 and the time delay layer 94 t times (t is an integer greater than or equal to 1) to realize the dual expansion of quantum computing scale and computing depth.

[0081] The core design idea of the above-mentioned optical quantum computing system is to organically combine time multiplexing and spatial parallelism to efficiently realize a unitary transformation effect equivalent to N spatial modes with n physical spatial modes, that is, n optical waveguide lines, where N is greater than n. This design is expected to significantly reduce the number of physical elements required to implement a linear optical network from O(N²), thereby fundamentally solving the expansion problem of the pure spatial scheme, while shortening the line depth, solving the problem of computing delay and cumulative photon loss of pure time multiplexing, and realizing a good approximate unitary matrix mapping and problem solving with O(N) complexity.

[0082] As can be seen from the above, the optical quantum computing system provided by the present application realizes an N-dimensional quantum computing task by adopting a time-space hybrid architecture of n physical spatial modes combined with tunable delay lines and inputting k consecutive photon pulse signals. Compared with the pure spatial scheme which requires O(N²) optical elements, the physical resource consumption of the present application is only O(N). This greatly reduces the requirements for chip area, manufacturing process and control system complexity, and clears the most important physical obstacles for building a super-large-scale optical quantum computer.

[0083] Furthermore, in the hybrid spatiotemporal architecture of the optical quantum computing system provided in this application, each optical quantum information only needs to pass through one tunable delay line once through the core computing module. With a total of t such computing modules, this means that each input pulse only needs to pass through t tunable delay lines. In contrast, in a pure time-multiplexing scheme, to achieve N-mode interaction, photons may need to cycle through the delay loop O(N) times, resulting in huge computational delays and unacceptable photon losses. This application significantly reduces the "temporal depth" of the computation from O(N), which greatly shortens the total transit time of photons in the system and effectively controls accumulated losses, thus making it possible to execute highly complex quantum algorithms requiring numerous sequential gate operations.

[0084] The above text combined Figures 1 to 10 This application provides detailed embodiments of the optical quantum computing system. Furthermore, this application also provides an optical quantum chip that can integrate the optical quantum computing system provided in any embodiment of this application. The optical quantum chip can be an optical quantum integrated circuit, which can be configured in an optical quantum computer as the core carrier for quantum information processing. This optical quantum chip can be used to programmably manipulate the quantum state of a single photon, realizing core functions such as quantum logic operations, quantum entanglement generation, and quantum state storage.

[0085] Below, for reference Figure 11 This application describes an optical quantum computer according to embodiments thereof. Figure 11 The diagram shown is a schematic diagram of the structure of an optical quantum computer provided in an exemplary embodiment of this application.

[0086] like Figure 11 As shown, in the optical quantum computer 1100, the core components of the quantum information processing link can include a quantum light source 1101, an optical quantum chip 1102, and a quantum detector 1103. The quantum light source 1101 is the "source of quantum information," the optical quantum chip 1102 is the "processor for quantum computing," and the quantum detector 1103 is the "readout device for quantum results." These three components form a closed loop through high-precision optical connections, jointly completing the generation, manipulation, and measurement of quantum states.

[0087] Exemplarily, the core role of the quantum light source 1101 is to generate quantum states with certain quantum properties, such as single photons, squeezed light, and the like. These quantum states are the basic information carriers (qubits) of optical quantum computing. The types of the quantum light source 1101 can include: quantum dot single photon sources (based on spontaneous emission of semiconductor quantum dots), diamond color center light sources (such as NV centers), spontaneous parametric down conversion (SPDC) light sources (generate entangled photon pairs or multi-photon states through nonlinear optical processes), and the like. For example, a quantum dot single photon source can emit single photons with a purity of more than 99% under pulsed laser excitation, and on-demand switching of polarization states can be achieved through external field regulation.

[0088] Exemplarily, the optical quantum chip 1102 is the core carrier of quantum information processing, and its role is to programmably manipulate the quantum states of single photons, to realize core functions such as quantum logic operations, quantum entanglement generation, and quantum state storage.

[0089] Exemplarily, the core role of the quantum detector 1103 is to accurately measure the existence, time, polarization, frequency, or phase amplitude of the light field of the photon, to convert the “quantum information” of the optical quantum state into readable electrical signals, and finally output the quantum computing result. The types of the quantum detector 1103 can include: superconducting nanowire quantum detectors (SNSPD, time resolution < 10 ps), single photon avalanche diodes (SPAD, high integration), and balanced homodyne detection for continuous variable measurement. For example, the detection efficiency of SNSPD for 1550 nm photons can reach 95%, and the dark count rate is lower than 0.1 count / s, which is the core equipment for high-precision quantum measurement.

[0090] In addition, it should be noted that, in actual applications, the optical quantum computing system and / or the optical quantum chip provided in the present application can be implemented based on a basic material such as lithium niobate (LiNbO3), and then integrated in an optical quantum computer. The specific processing process of the optical quantum computer can be referred to the specific description of related technologies, and will not be repeated here.

[0091] The above is a detailed introduction to the optical quantum computing system based on the space-time hybrid architecture provided in the embodiments of the present application. Next, the construction method of the optical quantum computing system provided in the embodiments of the present application will be introduced in view of the existing architecture type of the optical quantum computing system and the optical quantum computing system based on the space-time hybrid architecture provided in the embodiments of the present application.

[0092] In the prior art, for a pure spatial parallel scheme, although there is a method for a fully connected general architecture (such as the Clements architecture or the Reck architecture) to decompose a target unitary matrix into physical parameters corresponding to each optical element, in actual applications, in order to reduce optical loss, reduce the number of elements or adapt to a specific algorithm, a non-full connection, sparse or specific topology architecture is often used. For these more general spatial parallel architectures, the prior art lacks a general analytical parameter decomposition method, making it difficult to quickly and accurately map a target quantum logic gate (i.e., a target transformation matrix, such as a target unitary transformation) to hardware parameters. In addition, for a pure time-multiplexing scheme, for any given target computing task (i.e., a target transformation matrix, such as a target unitary transformation), there is also a lack of an analytical parameter decomposition method to quickly determine the on-off state of the optical switch and the dynamic phase modulation parameter at each time, which limits its ability to execute complex general algorithms.

[0093] In summary, for the optical quantum computing system of the pure spatial parallel architecture or the pure time-multiplexing architecture, especially for the optical quantum computing system of the non-full connection spatial parallel architecture, the pure time-multiplexing architecture and the more complex time-space hybrid architecture, due to the lack of a general analytical parameter decomposition method, the prior art cannot quickly and accurately map the target quantum logic gate (i.e., the target transformation matrix, such as the target unitary transformation) to the physical parameters of the underlying optical elements, thus lacking generality, and further leading to an inability to accurately construct a system.

[0094] Based on this, the present application aims to provide a general analytical parameter decomposition method and introduce a software and hardware co-design concept to solve the problems in the prior art. Specifically, the construction method of the optical quantum computing system provided by the embodiments of the present application determines the equivalent transformation matrix corresponding to the initial optical quantum computing system constructed for a target quantum computing problem, adjusts the system parameters of the initial optical quantum computing system based on the difference between the equivalent transformation matrix and the target transformation matrix for solving the target quantum computing problem, and then obtains the optical quantum computing system, realizing the mapping from any target transformation matrix to system parameters, so that the optical quantum computer is no longer limited to a specific hardware topology or a dedicated algorithm. Whether it is a pure time-multiplexing, non-full connection sparse spatial parallel architecture or other complex hybrid architecture, the present application can achieve accurate parameter mapping, thereby accurately constructing a system.

[0095] The construction method of the optical quantum computing system provided by the embodiments of the present application will be described in detail below. Figures 12 to 14 The construction method of the optical quantum computing system provided by the embodiments of the present application will be described in detail below.

[0096] Figure 12Fig. 1 shows a flowchart of a method for constructing a photonic quantum computing system according to an embodiment of the present application. The method can be applied to an electronic device, which can include a photonic quantum computer or the like. As shown in Fig. 1, the method can include the following steps. Figure 12

[0097] S1210, determining a target transformation matrix for solving a target quantum computing problem.

[0098] In some examples, the target quantum computing problem can be a specific quantum computing problem to be solved. The target transformation matrix can be a matrix capable of solving the target quantum computing problem, such as an N-dimensional target unitary matrix. The dimension N of the matrix represents the total number of quantum modes required to solve the target quantum computing problem. It should be noted that for different system hardware architectures, N has different physical mapping meanings: in a pure spatial parallel architecture, N corresponds to the number of optical waveguides; in a pure time or space-time hybrid architecture, N is less than or equal to the product of the number of optical waveguides and the time step (Time-bins).

[0099] In an example, an N-dimensional target unitary matrix can be determined according to the mathematical description of the specific quantum computing problem to be solved.

[0100] S1220, constructing an initial photonic quantum computing system based on the complexity of the target quantum computing problem.

[0101] In some examples, the initial photonic quantum computing system can include a photonic quantum computing system of any architecture type, such as a photonic quantum computing system of a pure spatial parallel architecture, a photonic quantum computing system of a pure time multiplexing architecture, or a photonic quantum computing system of a space-time hybrid architecture.

[0102] In an example, the initial photonic quantum computing system can include a plurality of adjustable system parameters, which can include at least one of a system architecture parameter and an element configuration parameter. The system architecture parameter can be a parameter for defining the outer architecture of the system, and the element configuration parameter can be a parameter for configuring the tunable elements in the system.

[0103] In some embodiments, the system architecture parameter can include at least one of an element connection structure, a photon signal step, and a number of optical waveguides, such as a unit connection structure of a spatial processing module, a unit connection structure of a time processing module, a unit connection structure between the spatial processing module and the time processing module, a number of repetitions of a core computing unit, a photon signal step, and a number of optical waveguides in the initial photonic quantum computing system. The unit connection structure can be the arrangement and connection mode of the units in the module, and the photon signal step can be the number of photon pulses emitted by the signal source to the optical waveguide.

[0104] ​In some embodiments, the element configuration parameters can include control parameters of the tunable elements, such as the unit delay time length of the time processing module, the unit phase adjustment degree of the space processing module, and the like in the initial optical quantum computing system. The unit delay time length of the time processing module can be the delay time length corresponding to each unit in the time processing module, and the unit phase adjustment degree of the space processing module can be the phase adjustment degree corresponding to each unit in the space processing module, such as the control voltage or control current corresponding to a Mach-Zehnder Interferometer (MZI), and the like.

[0105] For example, according to the complexity of the target quantum computing problem, the architecture search algorithm can be used to dynamically explore and determine the optimal topology structure of the optical quantum circuit and other system architecture parameters, and then an initialized optical quantum computing architecture can be constructed as the initial optical quantum computing system.

[0106] In some specific examples, for a pure space parallel architecture, a space topology connection graph can be determined by a search algorithm. Specifically, the number of layers of physical waveguides, the specific arrangement of MZIs or beam splitters (such as a rectangular grid, a triangular grid, or a sparse random connection), and the determination of which parameters are fixed and which are adjustable can be included, so as to balance the hardware complexity and the expression ability.

[0107] In some other specific examples, for a pure time architecture, time multiplexing sequence parameters can be determined. Specifically, the required time steps, the number of fiber delay loops, and the delay time (such as 1τ, 2τ, …) corresponding to each delay loop can be included, so as to construct a time-domain interference network that meets the dimension N requirement.

[0108] In some other specific examples, for a space-time hybrid architecture, the architecture search algorithm can be used to consider the space resources and time resources cooperatively, and dynamically adjust the line structure to determine the space-time hybrid line outer layer parameters, such as the physical mode number n of the space-time hybrid optical quantum computing architecture, the topology architecture of the space processing module, the number of layers of the space processing module, the delay time nτ of each time processing module, and the like.

[0109] In addition, in addition to using the architecture search algorithm, any one or a combination of more than one of the full connection pruning algorithm, the constructive generation algorithm, and the differentiable architecture search algorithm, and the like can be used to construct the initial optical quantum computing system.

[0110] S1230, determining an equivalent transformation matrix corresponding to the initial optical quantum computing system.

[0111] In some examples, the equivalent transformation matrix can be, for example, a physical transformation matrix corresponding to the physical modes of the initial optical quantum computing system. Illustratively, the equivalent transformation matrix can be determined based on a transformation matrix corresponding to a system equivalent model of the initial optical quantum computing system, which can be constructed.

[0112] For different types of system architectures, different ways can be employed to construct the system equivalent model and determine the equivalent transformation matrix.

[0113] In some embodiments, for an initial optical quantum computing system with a pure spatial parallel architecture, the equivalent transformation matrix can be determined by directly constructing a cascaded transmission model as the system equivalent model. Specifically, since the photons only transmit in space without involving time loops, each optical component (e.g., beam splitter, phase shifter) in the circuit of the initial optical quantum computing system can be directly represented as a corresponding basic small unitary matrix. All the basic small unitary matrices can be sequentially multiplied in the order of the photon transmission path to obtain a global equivalent unitary matrix U(θ) describing the input-output relationship of the entire optical quantum network. The global equivalent unitary matrix U(θ) is the equivalent transformation matrix corresponding to the initial optical quantum computing system. Here, θ can be a set of phase parameters of all spatial processing elements (or units).

[0114] In other embodiments, for an initial optical quantum computing system with a pure time architecture or a space-time hybrid architecture, the initial optical quantum computing system can be expanded in the spatial dimension using an equivalent spatial dimension expansion technique to generate a system equivalent model in pure spatial dimension, and then the equivalent transformation matrix can be determined. Based on this, the above step S1230 can specifically include: expanding the system model corresponding to the initial optical quantum computing system in the spatial dimension to obtain a system equivalent model; and determining the equivalent transformation matrix based on a transformation matrix corresponding to the system equivalent model.

[0115] In some examples, the system equivalent model obtained after expansion in the spatial dimension can be a static, fully spatialized quantum optical network model.

[0116] Illustratively, taking an initial optical quantum computing system with a space-time hybrid architecture as an example, to implement an N-dimensional feature transformation, a time step k (i.e., the number k of photon pulses emitted by the signal source to the optical waveguide) can be set, and k can be optimized as a tunable system parameter in the parameter optimization process. The equivalent spatial dimension expansion technique can be used to expand the system model corresponding to the initial optical quantum computing system in the equivalent spatial dimension, i.e., to expand the system model of the n-mode, time step k space-time hybrid architecture in the spatial dimension. This expansion operation equivalently converts the space-time hybrid system model into a static, fully spatialized optical quantum network model, i.e., a system equivalent model. The system equivalent model has Each input and output mode may contain [number] input and output modes, which can include [number] input and output modes. Individual delay line modes and One physical model (here) The delay line mode can be the mode corresponding to the photonic quantum information temporarily stored in the delay line during quantum state detection, and the physical mode can be the mode corresponding to the photonic quantum information that can be directly detected during quantum state detection. Thus, the equivalent model of this system can correspond to one... The dimensional physical evolution matrix (i.e., the transformation matrix). Furthermore, due to the dimensionality of the expanded physical evolution matrix... Typically much larger than the dimension N required to solve the target quantum computing problem, therefore, it is possible to expand it into... From the N×N dimensional physical evolution matrix, one or more N×N dimensional submatrices are selected using the sliding window method or other random methods. This serves as the equivalent transformation matrix, preparing for subsequent comparison with the target transformation matrix. The submatrix... The selection process corresponds to physically determining the injection and measurement time windows of the photon pulse. Thus, by using these extracted N-dimensional sub-matrices as the objects of subsequent co-optimization, the interference from the invalid parameter space can be significantly reduced, optimization efficiency improved, and preparation made for achieving high-precision parameter mapping.

[0117] S1240, based on the difference between the equivalent transformation matrix and the target transformation matrix, adjust the system parameters of the initial optical quantum computing system to obtain the optical quantum computing system.

[0118] For example, the equivalent transformation matrix corresponding to the initial optical quantum computing system can be quantized using a preset difference quantization algorithm. Transformation matrix with target The system parameters of the initial optical quantum computing system are continuously adjusted to minimize the differences between the parameters, and then converted into physical drive signals recognizable by the optical quantum chip controller. Through a high-speed digital-to-analog converter and a multi-channel logic controller, the generated voltage and sequence signals are precisely applied to the corresponding tunable phase shifters, variable beam splitters, and high-speed optical switches on the chip housing the optical quantum computing system. This ultimately yields an optical quantum computing system suitable for solving the target quantum computing problem.

[0119] In addition, when the optical quantum computing system is used to complete a calculation task corresponding to a target quantum computing problem, the photon pulse sequence encoding the initial quantum information can be input to the chip according to the optimized time step k, so that the photon pulse sequence evolves in the optical quantum computing system. The quantum state of the final output optical quantum information is a high-fidelity simulation of the quantum state of the initial quantum information after the target unitary matrix is applied to the quantum state of the initial quantum information, thereby completing the calculation task corresponding to the target quantum computing problem.

[0120] In this way, the embodiment of the present application determines the equivalent transformation matrix corresponding to the initial optical quantum computing system constructed for the target quantum computing problem, adjusts the system parameters of the initial optical quantum computing system based on the difference between the equivalent transformation matrix and the target transformation matrix used to solve the target quantum computing problem, and then obtains the optical quantum computing system, thereby realizing the mapping from the arbitrary target transformation matrix to the system parameters, so that the optical quantum computer is no longer limited to a specific hardware topology or a special algorithm. Whether it is a pure time multiplexing, a non-full connection sparse space architecture, or other complex hybrid architectures, the present application can realize accurate parameter mapping, thereby accurately constructing the system.

[0121] Based on this, in some embodiments, S1240 can specifically include: constructing a cost function according to the difference between the equivalent transformation matrix and the target transformation matrix; determining a parameter optimization result corresponding to the initial optical quantum computing system by using a preset optimization algorithm, with the goal of minimizing the function value of the cost function; and adjusting the system parameters of the initial optical quantum computing system based on the parameter optimization result to obtain the optical quantum computing system.

[0122] In some examples, a cost function can be established to quantify the difference between the equivalent transformation matrix and the target transformation matrix. The cost function can be a fidelity loss function as shown in the following formula (1).

[0123]

[0124] Wherein, P represents a set of adjustable parameters (i.e., system parameters) in the initial optical quantum computing system, Tr() represents a trace operation of a matrix, represents the equivalent transformation matrix corresponding to the optical quantum computing system under the system parameter set P, represents the target unitary matrix.

[0125] For example, with the goal of minimizing the function value of the cost function by using a preset optimization algorithm, iterative search can be performed in a multi-dimensional parameter space containing the system parameter set P until an optimal parameter set is found that can most approximate ​​As a result of parameter optimization, the optimal parameter set can be further used to obtain the optimal parameter set adjusting system parameters of the initial optical quantum computing system.

[0126] In addition, in some examples, the preset optimization algorithm described above can be a gradient optimization algorithm, such as a stochastic gradient descent algorithm. The gradient optimization algorithm is suitable for the case where the parameter space is continuously differentiable. Illustratively, the gradient of the cost function with respect to each phase parameter is calculated by automatic differentiation technology, and the parameter is updated using an optimizer such as Adam, SGD (stochastic gradient descent), L-BFGS, etc. This scheme has fast convergence speed and is suitable for large-scale parameter arrays.

[0127] In other examples, the preset optimization algorithm described above can also be a non-gradient heuristic algorithm, such as a particle swarm algorithm, a simulated annealing algorithm, etc. The non-gradient heuristic algorithm is suitable for the case where the parameter space contains discrete variables (such as switch states) or the gradient is difficult to calculate. Illustratively, algorithms such as genetic algorithms, particle swarm algorithms, simulated annealing, or differential evolution are used to search for optimal parameter solutions in the global range by simulating the evolution or group behavior in nature, effectively avoiding local optima.

[0128] In other examples, optimal parameter prediction based on a machine learning model can also be used. A prediction model is constructed using a deep neural network (such as a long short-term memory network, a Transformer, or a graph neural network). A large number of "target matrix-optimal parameter" data samples are used to train the model in advance; in the inference stage, the target transformation matrix is directly input into the model, and a set of predicted parameters is output by the model. The predicted parameters can be directly used as the result of parameter optimization, or as the initial value of gradient optimization to accelerate convergence.

[0129] Therefore, in order to find the most efficient mapping scheme, the parameter optimization results under different parameter configurations can be evaluated and screened. In some embodiments, the step S1220 can specifically include: constructing a target optical quantum computing system based on the complexity of the target quantum computing problem; adjusting the system architecture parameters of the target optical quantum computing system to obtain a plurality of initial optical quantum computing systems.

[0130] Correspondingly, the above-mentioned adjustment of the system parameters of the initial optical quantum computing system based on the parameter optimization result to obtain the optical quantum computing system can specifically include: determining a parameter optimization result that meets a pre-designed calculation accuracy condition from the parameter optimization results corresponding to the plurality of initial optical quantum computing systems respectively, as a target parameter optimization result; adjusting the system architecture parameters and element configuration parameters of the initial optical quantum computing system based on the target parameter optimization result to obtain the optical quantum computing system.

[0131] In some examples, the target photonic quantum computing system can be an initial photonic quantum computing system constructed in a manner related to the aforementioned in accordance with the complexity of the target quantum computing problem. The pre-design computation accuracy condition can be that the fidelity of the photonic quantum computing system constructed based on the parameter optimization result is greater than a preset threshold, which can be 95%, for example. The fidelity can be used to represent the similarity between the transformation matrix finally implemented by the photonic quantum computing system and the target transformation matrix.

[0132] For example, an original target photonic quantum computing system can be constructed first, and a plurality of rounds of parallel or serial parameter optimization processes can be performed by adjusting the system architecture parameters of the target photonic quantum computing system multiple times (such as trying different time steps k and trying different spatial topology sparseness).

[0133] Specifically, a plurality of initial photonic quantum computing systems corresponding to a plurality of system architecture parameters can be constructed by adjusting the system architecture parameters of the target photonic quantum computing system multiple times or iteratively adjusting the system architecture parameters of the photonic quantum computing system obtained after the last adjustment. System parameter optimization is performed for each initial photonic quantum computing system, and then a parameter optimization result corresponding to each initial photonic quantum computing system is obtained. Each parameter optimization result is stored in a candidate result pool. A parameter optimization result with a fidelity greater than 95% is obtained from the candidate result pool as a target parameter optimization result, and then the system architecture parameters and element configuration parameters of the initial photonic quantum computing system are adjusted based on the target parameter optimization result to obtain a final photonic quantum computing system.

[0134] Based on this, in some embodiments, when the number of target parameter optimization results is multiple, the above-mentioned adjustment of the system architecture parameters and element configuration parameters of the initial photonic quantum computing system based on the target parameter optimization result to obtain the photonic quantum computing system can specifically include: determining evaluation index values corresponding to the plurality of target parameter optimization results, respectively; determining a parameter optimization result with the highest evaluation index value from the plurality of target parameter optimization results as an optimal parameter optimization result; and adjusting the system architecture parameters and element configuration parameters of the initial photonic quantum computing system based on the optimal parameter optimization result to obtain the photonic quantum computing system.

[0135] For example, for a plurality of target parameter optimization results that meet the condition, at least one parameter evaluation index can be introduced to evaluate and screen the plurality of target parameter optimization results. The parameter evaluation index can be an index for representing the key performance of the photonic quantum computing system constructed based on a specific parameter optimization result, such as a computation accuracy index, a hardware adaptation index, a resource consumption index, etc.

[0136] Specifically, the score evaluation of each target parameter optimization result can be performed according to the at least one parameter evaluation index, and then the evaluation index value corresponding to each target parameter optimization result can be obtained. According to the evaluation index value corresponding to each target parameter optimization result, the multiple target parameter optimization results are sorted, and the target parameter optimization result with the highest evaluation index value is determined as the optimal parameter optimization result. Based on the optimal parameter optimization result, the system parameters of the initial optical quantum computing system are adjusted, and the optical quantum computing system that is most suitable for the target quantum computing problem and has the highest computing efficiency can be obtained.

[0137] In addition, in some embodiments, in order to improve the evaluation accuracy of the parameter optimization result, the above-mentioned evaluation index value can be a score obtained based on multi-dimensional comprehensive judgment of multiple parameter evaluation indexes. Based on this, in some embodiments, the above-mentioned determination of the evaluation index value corresponding to each of the multiple target parameter optimization results can specifically include: performing the following steps for each of the multiple target parameter optimization results to obtain the evaluation index value corresponding to each of the multiple target parameter optimization results: determining the index value corresponding to each of the multiple target parameter optimization results under multiple parameter evaluation indexes; and determining the evaluation index value corresponding to each of the multiple target parameter optimization results based on the index value corresponding to each of the multiple target parameter optimization results under the multiple parameter evaluation indexes.

[0138] In some examples, the multiple parameter evaluation indexes can include at least two of a calculation accuracy index, a hardware adaptation index, and a resource consumption index. The calculation accuracy index can be used to represent the fidelity of the optical quantum computing system constructed based on the parameter optimization result, for example, the index value corresponding to the calculation accuracy index can be determined by the similarity between the transformation matrix finally implemented by the optical quantum computing system and the target transformation matrix. The hardware adaptation index can be used to represent the adaptability between the parameter optimization result and the hardware for implementing the optical quantum computing system, for example, the index value corresponding to the hardware adaptation index can be determined by evaluating whether the parameter optimization result falls within the linear working area of the hardware, whether it is sensitive to manufacturing errors (i.e., robustness), and the like. The resource consumption index can be used to represent the resource consumption of the optical quantum computing system constructed based on the parameter optimization result, for example, the index value corresponding to the resource consumption index can be determined by evaluating the total photon transmission path length required for the calculation (corresponding to photon loss), the total number of time steps required (corresponding to calculation delay), and the number of active elements required (corresponding to energy consumption), and the like.

[0139] Exemplarily, a corresponding weight can be set for each parameter evaluation index, and the target parameter optimization result is evaluated and scored for each parameter evaluation index to obtain the index value corresponding to the target parameter optimization result under each parameter evaluation index. The index values corresponding to the target parameter optimization result under the plurality of parameter evaluation indexes are weighted and summed according to the weights corresponding to the plurality of parameter evaluation indexes, and then the final evaluation index value corresponding to the target parameter optimization result can be obtained.

[0140] In this way, through the cooperative optimization mode of the above-mentioned embodiments of the present application, it is ensured that the adaptation to the hardware system is realized under the premise of meeting the calculation precision, and at the same time, the least calculation steps are used as much as possible, thereby reducing the overall calculation delay and photon loss of the optical quantum computing system.

[0141] In addition, in some embodiments, the method can further include, in the case that there is no parameter optimization result in the plurality of parameter optimization results that meets the pre-designed calculation precision condition, returning to execute the step S1220 to perform parameter adjustment based on the re-constructed initial optical quantum computing system.

[0142] Exemplarily, if none of the plurality of parameter optimization results calculated meets the pre-designed calculation precision condition, that is, there is no parameter optimization result that meets the pre-designed calculation precision condition, the step S1220 can be returned to execute, the initial optical quantum computing system is re-constructed, and then the subsequent parameter optimization step is executed based on the re-designed initial optical quantum computing system.

[0143] In this way, by converting the complex quantum circuit design problem into a clearly defined mathematical optimization problem, automation from algorithm to hardware execution is realized. Designers do not need to manually design complex system architecture, but only need to provide the target transformation matrix, and the system can automatically generate the most efficient and highest precision physical implementation scheme. This cooperative optimization ensures that the quantum computing task can run with the best performance under the given hardware resources, maximizing the utilization efficiency of the hardware.

[0144] In summary, on the one hand, the embodiments of the present application determine the equivalent transformation matrix by establishing a space-time equivalent expansion model, and incorporate the system architecture parameters and element configuration parameters into a unified cooperative optimization framework, so that the optical quantum computer is no longer limited to a specific hardware topology or a special algorithm. Whether it is a pure time multiplexing, non-full connection sparse space architecture, or a complex space-time hybrid architecture, the present application can provide an accurate parameter mapping scheme. This greatly expands the ability of optical quantum hardware to execute any quantum logic gate (i.e., any target transformation matrix), significantly improving the universality and programmability of the system.

[0145] In another aspect, the embodiment of the present application adopts a hierarchical collaborative design strategy combining architecture search and parameter fine-tuning, and dynamically adjusts system architecture parameters such as the topology connection of the space-parallel architecture and the time step k of the time architecture as variable parameters during the optimization process. Compared with the traditional blind global search, this collaborative strategy effectively reduces the invalid search space and avoids the optimization algorithm from falling into local optimum. By dynamically adjusting the architecture size (such as automatically finding the minimum necessary number of time cycles k), the embodiment of the present application can quickly converge to the optimal parameter solution under the premise of ensuring high fidelity, making it possible to realize online compilation and rapid configuration of large-scale optical quantum chips.

[0146] In still another aspect, the embodiment of the present application introduces a multi-dimensional comprehensive evaluation mechanism in the final decision-making stage. When filtering the optimal solution, not only the mathematical approximation error is considered, but also physical indicators such as photon loss, calculation delay (number of time steps), and hardware energy consumption are considered, so that the "lowest cost" execution path can be automatically filtered out. For example, under the same calculation accuracy, the architecture configuration with fewer time cycles or fewer optical elements is preferentially selected. This not only directly reduces the transmission loss of photons in the chip (which is crucial for optical quantum computing) and reduces the overall delay of the system, but also indirectly improves the success rate and robustness of the calculation task running on imperfect hardware.

[0147] The above describes the method for constructing the optical quantum computing system according to the embodiments of the present application in detail. Figure 1 The above describes the method for constructing the optical quantum computing system according to the embodiments of the present application in detail. Figure 2 The above describes the method for constructing the optical quantum computing system according to the embodiments of the present application in detail.

[0148] Figure 13 As shown in FIG. 13, the apparatus 1300 for constructing the optical quantum computing system according to an embodiment of the present application includes: Figure 13 As shown in FIG. 13, the apparatus 1300 for constructing the optical quantum computing system according to an embodiment of the present application includes:

[0149] The matrix determination module 1301 is configured to determine a target transformation matrix for solving a target quantum computing problem.

[0150] The system construction module 1302 is configured to construct an initial optical quantum computing system based on the complexity of the target quantum computing problem.

[0151] The matrix equivalence module 1303 is configured to determine an equivalent transformation matrix corresponding to the initial optical quantum computing system.

[0152] The parameter adjustment module 1304 is configured to adjust system parameters of the initial optical quantum computing system based on a difference between the equivalent transformation matrix and the target transformation matrix, to obtain the optical quantum computing system.

[0153] In some embodiments of the present application, the matrix equivalence module 1303 is further configured to: expand the system model corresponding to the initial optical quantum computing system in a spatial dimension to obtain a system equivalent model; and determine the equivalent transformation matrix based on a transformation matrix corresponding to the system equivalent model.

[0154] In some embodiments of the present application, the parameter adjustment module 1304 is further configured to: construct a cost function according to the difference between the equivalent transformation matrix and the target transformation matrix; determine a parameter optimization result corresponding to the initial optical quantum computing system by using a preset optimization algorithm, with a function value of the cost function being minimized as an objective; and adjust the system parameters of the initial optical quantum computing system based on the parameter optimization result, to obtain the optical quantum computing system.

[0155] In some embodiments of the present application, the system parameters include system architecture parameters and element configuration parameters; and the system construction module 1302 is further configured to: construct the target optical quantum computing system based on a complexity of the target quantum computing problem; and adjust the system architecture parameters of the target optical quantum computing system to obtain a plurality of initial optical quantum computing systems. Correspondingly, the parameter adjustment module 1304 is further configured to: determine, from the parameter optimization results corresponding to the plurality of initial optical quantum computing systems respectively, a parameter optimization result satisfying a pre-designed calculation accuracy condition as a target parameter optimization result; and adjust the system architecture parameters and the element configuration parameters of the initial optical quantum computing system based on the target parameter optimization result, to obtain the optical quantum computing system.

[0156] In some embodiments of the present application, in a case where the number of the target parameter optimization results is a plurality, the parameter adjustment module 1304 is further configured to: determine an evaluation index value corresponding to each of the plurality of target parameter optimization results; determine, from the plurality of target parameter optimization results, a parameter optimization result with a highest evaluation index value as an optimal parameter optimization result; and adjust the system architecture parameters and the element configuration parameters of the initial optical quantum computing system based on the optimal parameter optimization result, to obtain the optical quantum computing system.

[0157] In some embodiments of the present application, the parameter adjustment module 1304 is further configured to: for each of the plurality of target parameter optimization results, perform the following steps to obtain an evaluation index value corresponding to the target parameter optimization result: determine an index value corresponding to the target parameter optimization result under a plurality of parameter evaluation indexes, the plurality of parameter evaluation indexes including at least two of a calculation accuracy index, a hardware adaptation index, and a resource consumption index; and determine the evaluation index value corresponding to the target parameter optimization result based on the index value corresponding to the target parameter optimization result under the plurality of parameter evaluation indexes.

[0158] In some embodiments of the present application, the system architecture parameters include at least one of element connection structure, photonic signal step length, and number of optical waveguides; and the element configuration parameters include control parameters of the tunable elements.

[0159] Hereinafter, an optical quantum computer according to an embodiment of the present application will be described with reference to Figure 14 Figure 14 Fig. 1 shows a structural schematic diagram of an optical quantum computer provided by another exemplary embodiment of the present application.

[0160] As Figure 14 shown, the optical quantum computer 1400 includes one or more processors 1401 and a memory 1402.

[0161] The processor 1401 can be a central processing unit (CPU) or other form of processing unit that has data processing capability and / or instruction execution capability, and can control other components in the optical quantum computer 1400 to perform desired functions.

[0162] The memory 1402 can include one or more computer program products that can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory, for example, can include random access memory (RAM), cache memory, and / or the like. The non-volatile memory, for example, can include read-only memory (ROM), hard disk, flash memory, and / or the like. One or more computer program instructions can be stored on the computer-readable storage media, and the processor 1401 can run the program instructions to implement the data acquisition method of various embodiments of the present application described above and / or other desired functions. Various contents such as data acquisition task templates, configuration information, instance tasks, and the like can also be stored in the computer-readable storage media.

[0163] In one example, the optical quantum computer 1400 can further include an input device 1403 and an output device 1404, which are interconnected through a bus system and / or other forms of connection mechanism (not shown).

[0164] The input device 1403 can include, for example, a keyboard, a mouse, and the like.

[0165] The output device 1404 can output various information including data acquisition task templates, configuration information, instance tasks, and the like to the outside. The output device 1404 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0166] Of course, in order to simplify, Figure 14 ​Only some of the components of the optical quantum computer 1400 related to the present application are shown, and components such as a bus, input / output interface, and the like are omitted. In addition, the optical quantum computer 1400 can include any other appropriate components according to the specific application.

[0167] In addition to the method and the device described above, an embodiment of the present application can also be a computer program product, which includes computer program instructions, which, when executed by a processor, cause the processor to perform the steps of the data acquisition method according to various embodiments of the present application described above in the specification.

[0168] The computer program product can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, etc., and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server.

[0169] In addition, an embodiment of the present application can also be a computer readable storage medium, which stores computer program instructions, which, when executed by a processor, cause the processor to perform the steps of the data acquisition method according to various embodiments of the present application described above in the specification.

[0170] The computer readable storage medium can be any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the above. More specific examples (a non-exhaustive list) of the computer readable storage medium include an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0171] The above describes the basic principles of the present application in combination with specific embodiments, but it needs to be pointed out that the advantages, benefits, effects and the like mentioned in the present application are only examples and are not limiting, and these advantages, benefits, effects and the like cannot be considered as necessary for each embodiment of the present application. In addition, the above specific details disclosed are only for the purpose of example and understanding, and are not limiting, and the above details do not limit the present application to be necessarily implemented with the above specific details.

[0172] The block diagrams of the devices, apparatuses, equipment, systems involved in the present application are only illustrative examples and are not intended to require or imply the connection, arrangement, configuration shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, systems can be connected, arranged, configured in any manner. Words such as "include", "contain", "have" and the like are open-ended words, mean "including but not limited to", and can be used interchangeably. The words "or" and "and" used herein mean the word "and / or", and can be used interchangeably unless the context clearly indicates otherwise. The word "such as" used herein means the phrase "such as but not limited to", and can be used interchangeably.

[0173] It also needs to be pointed out that in the devices, equipment and methods of the present application, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions of the present application.

[0174] The above description of the disclosed aspects is provided so that any person skilled in the art can make or use the present application. Various modifications to these aspects will be apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but is intended to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0175] The above description has been given for the purpose of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A photonic quantum computing system comprising a plurality of optical waveguides, characterized in that, The multiple optical waveguides are provided with multiple spatial processing modules and multiple time processing modules, wherein: The spatial processing module is configured to perform quantum state transformation on input optical quantum information to achieve spatial parallel processing. The time processing module is configured to perform time adjustment processing on input optical quantum information to achieve time multiplexing. The spatial processing module and the time processing module are alternately arranged, and adjacent spatial processing modules and time processing modules are connected through the multiple optical waveguides. The time processing module includes multiple time delay units, wherein: At least one time delay unit is provided on one optical waveguide. The time delay unit is configured to perform time delay processing on optical quantum information in the optical waveguide.

2. The photonic quantum computing system of claim 1, wherein, The spatial processing module includes at least one spatial interference unit, wherein: The spatial interference unit is connected with at least two optical waveguides in the multiple optical waveguides and is configured to perform interference processing on optical quantum information in the at least two optical waveguides.

3. The photonic quantum computing system of claim 2, wherein, At least one spatial interference unit is connected between any two adjacent optical waveguides in the multiple optical waveguides.

4. The photonic quantum computing system of claim 1, wherein, The spatial processing module includes multiple phase adjustment units, wherein: At least one phase adjustment unit is provided on one optical waveguide. The phase adjustment unit is configured to perform phase adjustment processing on optical quantum information in the optical waveguide.

5. The photonic quantum computing system of claim 1, wherein, The time delay unit includes an input end and an output end, the input end includes a first input end and a second input end, and the output end includes a first output end and a second output end, wherein: The first input end of the time delay unit is connected with the optical waveguide and is configured to receive target optical quantum information input to the time delay unit in the optical waveguide. The second input end of the time delay unit is connected with the second output end through a delay line to form a delay path to achieve time delay processing on the target optical quantum information when the target optical quantum information passes through the delay path. The first output end of the time delay unit is connected with the optical waveguide and is configured to output the target optical quantum information after time delay processing to the optical waveguide.

6. The photonic quantum computing system of any one of claims 2 to 4, wherein, The spatial processing module further includes multiple phase initialization units, wherein: At least one phase initialization unit is provided on one optical waveguide and is connected with a signal input end through the phase initialization unit. The phase initialization unit is configured to perform initialization phase modulation processing on a photon signal input by the signal input end and output optical quantum information obtained by initialization phase modulation processing to the optical waveguide.

7. The photonic quantum computing system of any one of claims 2 to 4, wherein, The spatial processing module further includes multiple phase compensation units, wherein: At least one phase compensation unit is provided on one optical waveguide and is connected with a signal output end through the phase compensation unit. The phase compensation unit is configured to perform phase compensation processing on optical quantum information output by the optical waveguide and output optical quantum information obtained by phase compensation processing to the signal output end.

8. An optical quantum chip, characterized by, The optical quantum chip integrates the optical quantum computing system according to any one of claims 1 to 7.

9. A photonic quantum computer, characterized by The optical quantum computer is configured with the optical quantum computing system as claimed in any one of claims 1 to 7 above.

Images