WO2023051550A1 - 一种基于集成光学的可编程高维量子计算芯片结构 - Google Patents
一种基于集成光学的可编程高维量子计算芯片结构 Download PDFInfo
- Publication number
- WO2023051550A1 WO2023051550A1 PCT/CN2022/121834 CN2022121834W WO2023051550A1 WO 2023051550 A1 WO2023051550 A1 WO 2023051550A1 CN 2022121834 W CN2022121834 W CN 2022121834W WO 2023051550 A1 WO2023051550 A1 WO 2023051550A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- optical network
- linear optical
- linear
- entangled
- configuration
- Prior art date
Links
- 230000003287 optical effect Effects 0.000 claims abstract description 243
- 238000005259 measurement Methods 0.000 claims abstract description 64
- 230000009466 transformation Effects 0.000 claims abstract description 34
- 238000009826 distribution Methods 0.000 claims description 3
- 239000002096 quantum dot Substances 0.000 abstract description 21
- 230000001105 regulatory effect Effects 0.000 abstract description 4
- 230000001427 coherent effect Effects 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 10
- 238000000034 method Methods 0.000 description 9
- 238000002360 preparation method Methods 0.000 description 7
- 230000010365 information processing Effects 0.000 description 5
- 238000000844 transformation Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 239000004065 semiconductor Substances 0.000 description 3
- 239000000969 carrier Substances 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000010409 thin film Substances 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 239000002070 nanowire Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000005610 quantum mechanics Effects 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N10/00—Quantum computing, i.e. information processing based on quantum-mechanical phenomena
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N10/00—Quantum computing, i.e. information processing based on quantum-mechanical phenomena
- G06N10/40—Physical realisations or architectures of quantum processors or components for manipulating qubits, e.g. qubit coupling or qubit control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/70—Photonic quantum communication
Definitions
- the present invention is based on the Chinese patent application CN202111143350.X filed on September 28, 2021, entitled “A Programmable High-Dimensional Quantum Computing Chip Structure Based on Integrated Optics", and claims the priority of this patent application, which is incorporated by reference The contents disclosed therein are fully incorporated into the present disclosure.
- the invention relates to the technical field of quantum computing, in particular to a programmable high-dimensional quantum computing chip based on integrated optics.
- Quantum computing is a new computing model that follows the laws of quantum mechanics and performs calculations by regulating quantum information units, namely qubits. Quantum computing uses the superposition, interference and entanglement properties of quantum to perform calculations. It has natural parallelism and super large information storage capacity, so it has great potential that classical computing cannot match. It has great application potential in many fields.
- Linear optical systems are one of the main physical approaches to quantum computing. Its main advantages include: photons have a long coherence time and are not easily decohered by external environment interference; photons are easy to achieve high-precision manipulation; photons with multiple degrees of freedom can be used to encode high-dimensional quantum.
- the integrated optical quantum chip uses integrated optical technology to integrate discrete linear optical elements into a single semiconductor integrated chip in the form of a thin film. Good stability and better scalability. Integrated optical quantum chips can realize the miniaturization and integration of discrete component optical systems on huge optical platforms, and are considered to be the most effective way to realize large-scale optical quantum computing systems.
- An embodiment of the present invention provides a programmable high-dimensional quantum computing chip based on integrated optics, which realizes the preparation, manipulation, and projection measurement of high-dimensional optical quantum states on the chip, thereby realizing high-dimensional quantum state input, high-dimensional quantum gate operation, high-dimensional A general-purpose high-dimensional quantum computing chip with programmable control for quantum state projection measurement.
- An embodiment of the present invention provides a programmable high-dimensional quantum computing chip based on integrated optics, including: a linear coefficient configuration network, N entangled multi-photon sources, and a wavelength division multiplexer, configured to generate multiple photons, and record each entanglement
- the number of entangled photons generated by the multi-photon source is P, and the photons of different wavelengths are output according to the wavelength through the wavelength division multiplexer.
- the initial configuration of the linear optical network O is connected to the configurable entangled multi-photon source , according to the wavelength of the output photon of the wavelength division multiplexer to form the corresponding O1, O 2 ... OP , which is set to prepare the initial state of the photon output from the configurable entangled multi-photon source; the unitary operator is configured linearly
- the optical network U is correspondingly connected to the initial configuration linear optical network O to form corresponding U 1 (i) , U 2(i )...
- the projection measurement linear optical network T is correspondingly connected with the unitary operator configuration linear optical network U to form a corresponding T 1 , and T 2 .
- the initial state configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network all belong to the general linear optical network .
- the configurable entangled multi-photon source the initial state configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optics
- the network implements path coding through the first phase shifter and the second phase shifter.
- the linear coefficient configuration network includes a log 2 N-level Mach-Zehnder interferometer, arranged in the form of a "binary tree", namely Each output port of the Mach-Zehnder interferometer of the upper stage is connected to an input port of a Mach-Zehnder interferometer of the next stage, and the Mach-Zehnder interferometer of the last stage Each output port is connected to a second phase shifter and an entangled multiphoton source, wherein the Mach-Zehnder interferometer includes a first phase shifter, and two multimodes connected to the first phase shifter interferometer.
- the programmable high-dimensional quantum computing chip based on integrated optics can be realized in the form of: quantum computing, where is the quantum operation to be realized, expressed as a linear combination of multiple linear optical unitary transformations U, where ⁇ i is the coefficient of the linear term, and the initial state of calculation is It is a multi-photon high-dimensional quantum state.
- the configurable entangled multi-photon source generates photons of P wavelengths, and the photons of P wavelengths are respectively routed to P
- a linear optical network is configured in the initial state of the group, where P is a natural number, and P ⁇ 2.
- the initial configuration linear optical network may include a multi-stage chain structure.
- the unitary operator configuration linear optical network may be an optical network structure in a triangular distribution.
- the projection measurement linear optical network may include a reverse tree structure.
- the first phase shifter and the second phase shifter adjust each path of light through an external classical control signal to realize path encoding.
- each group of initial state configuration linear optical networks in the P groups of initial state configuration linear optical networks has N pieces; correspondingly, The unitary operator configuration linear optical network is divided into P groups, each group is N, and the number of the projection measurement linear optical network is P; wherein, each group of unitary operator configuration linear optical network and a set of initial state Configure the linear optical network and a projection measurement linear optical network for corresponding connections.
- the programmable high-dimensional quantum computing chip structure based on integrated optics provided by the embodiment of the present invention generates multiple photons and encodes the path of the entangled multi-photon source, and performs initial state of the photons routed according to the wavelength through a general linear optical network. Preparation, unitary transformation, linear combination, beam combining and projection measurement, realize the generation, manipulation and measurement of quantum information carriers on a single integrated optical chip, thus enabling the realization of integrated, miniaturized, scalable and programmable quantum Computing devices made possible.
- the programmable high-dimensional quantum computing chip based on integrated optics realizes the integration of optical chips for quantum computing. It has better stability and better scalability.
- the large-scale programmable high-dimensional quantum computing chip technology based on integrated optics can support the scalable realization of the linear combination scheme based on unitary operators, construct a fully programmable high-dimensional quantum bit computing chip, and realize multi-qubit quantum computing based on photons. information processing.
- the embodiment of the present invention adopts the linear combination of unitary operators based on path coding to realize high-dimensional qubit calculation, and realizes by adjusting the multi-photon path entanglement state prepared by the configurable entangled multi-photon source and the unitary operator to configure the linear optical network.
- Optical unitary transformation thus realizing high-dimensional multi-bit quantum computing by combining on-chip high-dimensional multi-photon path entanglement and path-encoding-based linear combination.
- Fig. 1 is one of the structural schematic diagrams of a programmable high-dimensional quantum computing chip based on integrated optics provided by an embodiment of the present invention
- Fig. 2 is the second schematic diagram of the structure of the programmable high-dimensional quantum computing chip based on integrated optics provided by the embodiment of the present invention
- Fig. 3 is a schematic flow diagram of the quantum computing process in the programmable high-dimensional quantum computing chip based on integrated optics provided by the embodiment of the present invention
- Fig. 4 is a schematic structural diagram of an optical network used for initial state preparation provided by an embodiment of the present invention.
- FIG. 5 is a schematic diagram of a triangular-distributed optical network structure provided by an embodiment of the present invention.
- FIG. 6 is a schematic structural diagram of an optical network for photon beam combining provided by an embodiment of the present invention.
- Fig. 7 is a schematic structural diagram of an optical network used for projection measurement provided by an embodiment of the present invention.
- Fig. 8 is a schematic structural diagram of an integrated optical chip for realizing four-dimensional qubit calculation in a two-photon entangled state provided by an embodiment of the present invention.
- an embodiment of the present invention provides a programmable high-dimensional quantum computing chip based on integrated optics, including: a configurable entangled multi-photon source 102, including a linear coefficient configuration network, N entangled multi-photon sources and wavelength division
- the multiplexer is set to generate multiple photons, and the number of entangled photons produced by each entangled multi-photon source is P, and the photons of different wavelengths are respectively output according to the wavelength through the wavelength division multiplexer.
- the first phase shifter and the second phase shifter configure the pump light entering the interference configuration network according to the output path to obtain a set of linear configuration coefficients, denoted as ⁇ 1 , ⁇ 2 ... ⁇ N .
- the linear optical network 104 is configured in the initial state, connected to the configurable entangled multi-photon source, and forms corresponding O 1 , O 2 ... OP according to the wavelength of the output photon of the wavelength division multiplexer, and is set to the
- the initial state is prepared from photons output by a configurable entangled multiphoton source.
- linear combination refers to the linear combination of unitary transformations realized by different optical networks.
- the projection measurement linear optical network 108 is correspondingly connected with the unitary operator configuration linear optical network U to form corresponding T 1 , T 2 . . . .
- projection measurement is to decompose a Hermitian operator representing an observable quantity on a system state space into multiple measurement operators. This measurement operator is actually generated by the Hermitian operator towards the corresponding eigenvalue The projection of the eigensubspace of .
- the linear term coefficients of the entangled multi-photon source path can be configured, and the linear optical network can be configured by configuring the linear optical network in the initial state and the unitary operator.
- the unitary operator of the initial state and the linear combination realizes the organic combination and use of a multi-photon source that can prepare multi-photon path entanglement states and a general linear optical network that can realize unitary transformation.
- the programmable high-dimensional quantum computing chip based on integrated optics in the embodiment of the present invention has the characteristics of programmable and extensible.
- the programmable high-dimensional quantum computing chip based on integrated optics can prepare multi-photon high-dimensional entangled quantum states and realize programmable high-dimensional qubit computing.
- Quantum chip technology based on integrated optics has made great progress.
- This technology uses semiconductor micro-nano processing technology to integrate discrete optical components into a single chip. Compared with discrete optical components, it has the advantages of small size, high stability, and strong scalability. It is an effective way to realize a large-scale optical quantum computing system.
- the large-scale integrated optical quantum chip technology can support the scalable implementation based on the unitary operator linear combination scheme, construct a fully programmable high-dimensional quantum bit computing chip, and realize photon-based multi-qubit quantum information processing.
- the manufacturing process of programmable high-dimensional quantum computing chips based on integrated optics such as silicon-based optical waveguides can be compatible with CMOS, and the programmable high-dimensional quantum computing chips based on integrated optics in the embodiments of the present invention can be further compared with traditional CMOS computing chips. Fusion, for the future design of photonic quantum information processing chips that realize photoelectric fusion and hybrid architecture.
- the programmable high-dimensional quantum computing chip based on integrated optics can also be provided with an integrated light source, which is configured to generate photons and output them to the linear coefficient configuration network.
- a programmable high-dimensional quantum computing chip based on integrated optics can also be integrated with a single-photon detector, which is set to detect the photons output by the linear optical network.
- the single photon detector may be an avalanche photodiode or a superconducting nanowire detector.
- the initial configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network all belong to the general linear optical network.
- the configurable entangled multi-photon source, the initial state configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network all realize path encoding through the first phase shifter and the second phase shifter.
- a Mach-Zehnder interferometer includes a phase shifter 211 and two multimode interferometers 212 connected to the phase shifter 211 .
- the first-stage Mach-Zehnder interferometer of the log 2 N-level Mach-Zehnder interferometer receives an input beam 221 outside the configurable entangled multiphoton source 102, and the log 2 N-level Mach-Zehnder interferometer N paths of light are formed according to the input light beam 221 , and the N paths of light are output to N entangled multi-photon sources 215 .
- each entangled multiphoton source is connected to a second phase shifter 214 . All the phase shifters of the configurable entangled photon source module are set to adjust each light output from the linear coefficient configuration network, which is used to configure the coefficient of the linear term.
- the configurable entangled multiphoton source 102 also includes a wavelength division multiplexer 216 .
- general linear optical networks include:
- Configure the linear optical network 201 in the initial state connect with the configurable entangled multi-photon source, form corresponding O 1 , O 2 ... OP according to the wavelength of the output photon of the wavelength division multiplexer, and set it to the
- the initial state is prepared from photons output by a configurable entangled multiphoton source.
- the projection measurement linear optical network 203 is correspondingly connected with the unitary operator configuration linear optical network U to form corresponding T 1 , T 2 . . . .
- the programmable high-dimensional quantum computing chip structure based on integrated optics in the embodiment of the present invention includes N entangled multi-photon sources and In a Mach-Zehnder interferometer, the N entangled multiphoton sources can be denoted as S 1 , S 2 , . . . , S N , where N is a natural number and N ⁇ 2.
- the Mach-Zehnder interferometers in the first-level Mach-Zehnder interferometer network are arranged in the form of a binary tree, that is, each output port of the upper-level Mach-Zehnder interferometer is connected to a Mach-Zehnder interferometer of the next level One input port of the instrument, and the N output ports of the last-stage Mach-Zehnder interferometer are connected to a second phase shifter and an entangled multiphoton source.
- each entangled multi-photon source By setting the first phase shifter and the second phase shifter in the configurable entangled multi-photon source and correspondingly configuring each light output from the linear coefficient configuration network, different linear term coefficients can be obtained as required. Wherein, each entangled multi-photon source generates P photons with different wavelengths.
- the photons After the photons pass through the wavelength division multiplexer, they are respectively routed to the entrances of the initial state configuration linear optical network of the P group.
- the function of the initial state configuration linear optical network is to prepare the initial state.
- the number of linear optical networks configured in the initial state is set as P groups, and the initial state configured linear optical networks in the P groups can be marked as: O 1 , O 2 , ..., O P .
- P is a natural number and P ⁇ 2.
- O 1 , O 2 ,..., OP can realize the preparation of t-dimensional quantum states encoded on P paths, where the initial state O 1 , O 2 , . . . , OP can be any t-dimensional linear optical operation, and the same operation can be performed for each OM in the chip, where M is a natural number and 1 ⁇ M ⁇ P.
- the number of linear optical networks configured by unitary operators is P*N, which can be divided into P groups, and each group includes N linear optical networks configured by unitary operators.
- the first group of unitary operator configuration linear optical network U 1 is correspondingly connected with the first group of initial state configuration linear optical network O 1
- the second group of unitary operator configuration linear optical network U 2 is connected with the second group of initial state configuration
- the linear optical network O 2 is correspondingly connected
- the Pth group unitary operator configuration linear optical network U P is correspondingly connected with the Pth group initial state configuration linear optical network OP .
- unitary operator configured linear optical networks can be denoted as U 1 (1) , U 1 (2) ... U 1 (N) , U 2 (1) , U 2 (2) ... U 2 (N) ,..., UP (1) , UP (2) ... UP (N) .
- the number of projection measurement linear optical networks is P, and each projection measurement linear optical network corresponds to a group of O and U.
- the P projection measurement linear optical networks can be denoted as: T 1 , T 2 , . . . TM , . . . T P , each TM having t ports.
- the initial state configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network are all general linear optical networks that can realize t-dimensional unitary transformation.
- unitary transformation is a transformation that preserves the inner product, and the inner product of two vectors before unitary transformation is equal to the inner product after transformation.
- a unitary transformation is a transformation made by using a unitary operator. There is a transformation of the base vector and a transformation of the operator. It can be considered that the unitary transformation is the isomorphism between two Hilbert (Hilbert) spaces .
- U j is The gate acting on the d-dimensional target (T) subspace
- ⁇ j is a complex coefficient, satisfies
- the control qubit By moving partial states of the target qubit into the extended Hilbert space, the control qubit can be more simply acted on a single qubit unit.
- a technique based on extended computing Hilbert space can be used to implement a linear combination circuit.
- Any quantum unitary operation can in principle be decomposed into a linear sum of elementary operations.
- Cartan's KAK decomposition it is possible to rewrite any two-qubit unitary operation, converting it into a linear combination of four linear terms, each of which is the tensor product of two single-qubit gates.
- the Cartan decomposition method allows n-qubit unitary operations to be reformulated as a linear combination of tensor products of n single-qubit gates.
- it is necessary to add coherent control for arbitrary unknown quantum operations and the technique is based on the logical Hilbert space expansion of gates for computation.
- Each entangled multi-photon source generates P photons with different wavelengths.
- the requirements firstly, by adjusting the phase shifter in the configurable entangled multiphoton source, the paths of N entangled multiphoton sources are encoded, and the coefficients ⁇ 1 , ⁇ 2 , ... ⁇ N of the linear term are obtained.
- the photons After passing through the wavelength division multiplexer, the photons enter different initial states according to the wavelength to configure the linear optical network, and finally measure the output of the linear optical network from different projections.
- the initial configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network are multiple t-dimensional configurable linear optical networks. According to the different entangled states of multi-dimensional entangled photons, the initial state configuration linear optical network is denoted as O 1 , O 2 ,... OM ,... OP , where M is a natural number and 1 ⁇ M ⁇ P, and each group of O There are N number of M.
- the module 311 composed of configurable entangled multi-photon source and unitary operator configuration linear optical network is set to perform step 321: perform coefficient configuration and unitary operator linear combination; initial state Configuring the linear optical network 312 is set to perform step 322 : perform any initial state preparation; and the linear optical network 313 for projection measurement is set to perform step 323 : perform projection measurement.
- step 321, step 322 and step 323 may be collectively referred to as step 331: Realization of linear combination of unitary operators based on photon path state coding.
- the initial configuration linear optical network may be a multi-stage chain structure as shown in FIG. 4
- the unitary operator configuration linear optical network may include a t-dimensional simplified form An optical network structure with triangular distribution.
- FIG. 6 is a schematic diagram of an optical network structure for photon beam combining.
- the linear optical network for projection measurement can be a reverse tree structure as shown in FIG. 7 .
- the initial quantum state can generally be set to
- Quantum computing operations performed on a programmable high-dimensional quantum computing chip based on integrated optics are realized by a path-coding-based linear combination scheme of unitary operators, where the linear coefficients are provided by a configurable entangled multi-photon source, and the linear combination of unitary operators is provided by
- the unitary operator configuration linear optical network is provided, and the unitary operator configuration linear optical network includes N*P linear optical networks that can realize the required t-dimensional unitary transformation, and the M group of photons with the same wavelength
- Each photon in N is routed to the initial state. Configure the linear optical network O M to generate the initial state. Then route to the unitary operator to configure the linear optical network U M (1) , U M (2) , ..., U M (N) to complete the unitary transformation and linear combination. Finally routed to the projection measurement linear optical network to complete the projection measurement.
- the programmable high-dimensional quantum computing chip based on integrated optics in the embodiment of the present invention uses integrated optics technology to integrate discrete linear optical elements into a single semiconductor integrated chip in the form of a thin film. Reduced, and the whole system has better stability and better scalability due to high integration.
- the important components required for programmable high-dimensional quantum computing chips based on integrated optics have been experimentally realized, such as on-chip single-photon sources and entangled photon sources, on-chip wavelength division multiplexers, and on-chip general linear optical network implementations.
- an on-chip photon source is used to generate entangled photons
- a linear optical network composed of an on-chip Mach-Zehnder interferometer and a phase controller is used to control the behavior of the photons, and then through the on-chip integrated single photon detection.
- photon detectors to detect photons, a large-scale programmable high-dimensional quantum computing chip based on integrated optics can be designed to realize complex quantum information processing applications.
- an integrated optical chip structure is established by using a programmable high-dimensional quantum computing chip approach based on integrated optics, using an on-chip path entangled multi-photon source and a general-purpose linear optical network.
- different multi-photon and multi-path entangled states are generated through the on-chip path entanglement multi-photon source, and different optical unitary transformations are configured through the on-chip general linear optical network to realize different computing tasks according to requirements;
- the output measurement can obtain the result of quantum information processing and realize general quantum information calculation.
- the first phase shifter and the second phase shifter in the entangled multi-photon source module adjust each path of light through an external classical control signal, and encode the paths of N entangled multi-photon sources , to get the linear term coefficients.
- each group of initial state configuration linear optical networks in the P group of initial state configuration linear optical networks includes N; correspondingly, unitary operator configuration linear optical networks are divided into P groups, and each group has N , the number of projection measurement linear optical networks is P; wherein, each set of unitary operator configuration linear optical networks is correspondingly connected with a set of initial state configuration linear optical networks and a projection measurement linear optical network.
- the initial state configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network are all general linear optical networks with multiple t-dimensional configurations.
- a multi-photon source that can prepare multi-photon path entangled states is organically combined with a linear optical network that can realize unitary transformation.
- the multi-photon path entangled state is prepared by regulating the multi-photon source, and different optical unitary transformations are configured through the on-chip general linear optical network.
- the programmable high-dimensional quantum computing chip based on integrated optics can prepare multi-particle high-dimensional entangled optical quantum states, using The linear combination of unitary operators realizes programmable high-dimensional qubit computing.
- FIG. 8 a schematic diagram of the structure of a programmable high-dimensional quantum computing chip based on integrated optics for realizing four-dimensional qubit computing.
- the initial state of the two-photon source is configured to obtain the coefficients ⁇ 1 and ⁇ 2 of the linear term.
- Each entangled multi-photon source generates frequency-entangled photon pairs, and a wavelength division multiplexer is used to separate signal photons and idler photons generated by the entangled photon source through four-wave mixing.
- the configurable entangled multi-photon source 801 generates photons of two wavelengths, and routes them to different networks of the initially configured linear optical network 802 according to the wavelengths.
- the initial state configuration linear optical network 802 prepares the entangled photons obtained by the entangled multi-photon source into any initial state according to the requirements, and then routes to the unitary operator configuration linear optical network 803, and the unitary operator configuration linear optical network 803 includes 4 A linear optical network, namely U 1 (1) , U 1 (2) , U 2 (1) and U 2 (2) , which are all 4-dimensional linear optical network structures, can realize unitary operators based on path coding Linear combination scheme.
- the unitary operator configuration linear optical network 803 routes the linearly combined photons to the projection measurement linear optical network 804 for projection measurement.
- the paths of the two entangled multiphoton sources are encoded, and the coefficients ⁇ 1 and ⁇ 2 of the linear term are obtained.
- photons After passing through the wavelength division multiplexer, photons enter different linear optical networks according to different wavelengths, and finally output from different projection measurement optical networks.
- Combining post-selection techniques to perform coincidence measurements of 2 photons of different wavelengths at the projection measurement optical network The corresponding multi-photon multi-path entangled state is generated at the entrance of each optical network corresponding to the configuration of the linear optical network in the initial state.
- the initial configuration linear optical network, the unitary operator configuration linear optical network and the projection measurement linear optical network are all 4-dimensional configurable linear optical networks.
- the programmable high-dimensional quantum computing chip based on integrated optics can be realized in the form of: quantum computing, where is the quantum operation to be realized, expressed as a linear combination of multiple linear optical unitary transformations U, where ⁇ i is the coefficient of the linear term, and the initial state of calculation is The result of quantum computing is a multi-photon high-dimensional quantum state.
- phase shifters adjust each path of light through external classical control signals, and by applying different external classical control signals to each phase shifter, the programmability of the optical quantum computing chip is realized.
- the final result
- the configurable entangled multi-photon source includes three entangled multi-photon sources, and each entangled multi-photon source generates photons of three different wavelengths.
- the requirements firstly, by adjusting the phase shifter in the configurable entangled multiphoton source, the paths of the three entangled multiphoton sources are encoded, and the coefficients ⁇ 1 , ⁇ 2 , ⁇ 3 of the linear term are obtained.
- the photons After passing through the wavelength division multiplexer, the photons enter different initial states according to the wavelength to configure the linear optical network, and finally measure the output of the linear optical network from different projections.
- the configurable entangled multi-photon source includes two entangled multi-photon sources, and each entangled multi-photon source generates photons of three different wavelengths.
- the requirements firstly, by adjusting the phase shifter in the configurable entangled multiphoton source, the paths of the two entangled multiphoton sources are encoded, and the coefficients ⁇ 1 and ⁇ 2 of the linear term are obtained.
- the photons After passing through the wavelength division multiplexer, the photons enter different initial states according to the wavelength to configure the linear optical network, and finally measure the output of the linear optical network from different projections.
- the embodiment of the present invention adopts the linear combination of unitary operators based on path coding to realize high-dimensional qubit calculation, and realizes the optical network realized by adjusting the multi-photon path entanglement state prepared by the entangled multi-photon source and the unitary operator configuration linear optical network.
- Unitary transformation thus realizing high-dimensional multi-bit quantum computing by combining on-chip high-dimensional multi-photon path entanglement and linear combination based on path coding.
- the coefficients of the optical unitary operation items of the linear combination are configured by configuring the phase shifter in the entangled multi-photon source, the preparation of each initial state is realized by configuring the linear optical network through the initial state, and the linear optical network is realized by configuring the linear optical network through the unitary operator.
- Each term and linear combination of unitary transformation, and finally the calculation result is obtained by projective measurement through the projective measurement linear optical network.
- the programmable high-dimensional quantum computing chip based on integrated optics can complete multi-particle high-dimensional entangled optical quantum states to realize high-dimensional qubit computing.
- the programmable high-dimensional quantum computing chip based on integrated optics provided by the embodiment of the present invention generates multiple photons and encodes the path of the entangled multi-photon source, and prepares the initial state of the photons respectively routed according to the wavelength through the linear optical network.
- Unitary transformation, linear combination, and projection measurement realize the functions of generating, manipulating, and measuring quantum information carriers on a single integrated optical chip, making it possible to realize integrated, miniaturized, scalable, and programmable quantum computing devices. possible.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Optimization (AREA)
- Artificial Intelligence (AREA)
- Computational Mathematics (AREA)
- Condensed Matter Physics & Semiconductors (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Mathematical Analysis (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Optics & Photonics (AREA)
- Optical Modulation, Optical Deflection, Nonlinear Optics, Optical Demodulation, Optical Logic Elements (AREA)
Abstract
本发明实施例提供一种基于集成光学的可编程高维量子计算芯片，通过调控纠缠多光子源制备的多光子路径纠缠态和通用线性光学网络实现的光学幺正变换，以对幺正变换进行不同线性组合实现可编程高维量子比特计算。芯片包括：可配置纠缠多光子源，包括线性系数配置网络、N个纠缠多光子源及波分复用器，设置为产生多光子，对N个纠缠多光子源的相干泵浦光进行配置，得到线性项系数，并将多个光子根据波长分别输出；初态配置线性光学网络，设置为对可配置纠缠多光子源输出的光子制备量子初态；幺正算符配置线性光学网络，设置为实现光学幺正变换，以及合束实现幺正变换线性组合；投影测量线性光学网络，设置为对合束后的光量子态进行投影测量。
Description
相关申请的交叉引用
本发明基于2021年9月28日提交的发明名称为“一种基于集成光学的可编程高维量子计算芯片结构”的中国专利申请CN202111143350.X，并且要求该专利申请的优先权，通过引用将其所公开的内容全部并入本公开。
本发明涉及量子计算技术领域，尤其涉及一种基于集成光学的可编程高维量子计算芯片。
量子计算是遵循量子力学规律通过调控量子信息单元即量子比特来进行计算的新型计算模式。量子计算利用量子的叠加、干涉以及纠缠特性进行计算，具有天然的并行性及超大的信息存储能力，从而具有经典计算不可比拟的巨大潜力，在诸如大数质因子分解、数据库搜索、化学分子模拟等众多领域具有巨大应用潜力。
线性光学系统是实现量子计算的主要物理途径之一。其主要优势包括：光子具有很长的相干时间，不容易受到外界环境干扰而退相干；光子容易实现高精度操控；光子多自由度可以用于编码高维量子。
许多量子计算应用可以在线性光学系统中实验实现，这些线性光学系统大多为分立元件光学系统。采用分立元件光学系统实现可编程量子计算时，需要对分立光学元件进行调制，实验难度较大，且对实验技术具有较高的要求。此外，自由空间线性光学系统需要大量分立 的光学元件，整个系统体积巨大，且各个分立元件容易受到外界环境温度、振动等因素影响，从而系统的稳定性和可扩展性受到限制。
集成光学量子芯片是采用集成光学技术将分立的线性光学元件以薄膜形式集成到单个半导体集成芯片上，与分立元件光学系统相比，不仅体积显著减小，而且整个系统由于高集成度而具有更好的稳定性及更好的可扩展性。集成光学量子芯片能够实现庞大光学平台上分立元件光学系统的小型化、集成化，被认为是实现大规模光学量子计算系统的最有效途径。
目前线性光学量子系统大多采用参量下变换或四波混频效应来概率性地产生单光子对，理论上可以实现确定性的单光子源，仍有相当多的技术难题需要克服，因而光子资源整体较为昂贵。另一方面，单光子具有多种自由度，对单光子进行高维编码和操控，能够显著光量子计算中降低光子资源的需求。根据光子高维量子态的制备和操控特性，结合集成光学量子芯片技术的优势，能够为线性光学量子计算系统计算能力的提升提供新的途径。
发明内容
本发明实施例提供一种基于集成光学的可编程高维量子计算芯片，通过片上实现高维光量子态的制备、操控以及投影测量，进而实现高维量子态输入、高维量子门操作、高维量子态投影测量均可编程控制的通用高维量子计算芯片。
本发明实施例提供一种基于集成光学的可编程高维量子计算芯片，包括：线性系数配置网络、N个纠缠多光子源以及波分复用器，设置为产生多个光子，记每个纠缠多光子源产生的纠缠光子的数量为P，不同波长的光子经过波分复用器根据波长分别输出，通过调节线性系数配置网络中的第一移相器与第二移相器，对进入干涉配置网络的泵浦光按输出路径进行配置，得到一组线性配置系数，记为 α
1，α
2...α
N；初态配置线性光学网络O，与所述可配置纠缠多光子源连接，根据所述波分复用器输出光子的波长形成对应的O1，O
2...O
P，设置为对所述可配置纠缠多光子源输出的光子制备初态；幺正算符配置线性光学网络U，与所述初态配置线性光学网络O对应连接，形成对应的U
1
(i)，U
2(i)...U
P
(i)(i＝1，2，...N)，设置为进行幺正变换，经过合束后，实现幺正算符的线性组合，从而得到最终量子态结果：
投影测量线性光学网络T，与所述幺正算符配置线性光学网络U对应连接，形成对应的T
1，T
2...T
P设置为对合束后的所述光量子态进行投影测量。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片，所述初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络都属于通用线性光学网络。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片，所述可配置纠缠多光子源、初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均通过所述第一移相器和所述第二移相器实现路径编码。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片，所述线性系数配置网络包括log
2N级马赫-曾德尔干涉仪，以“二叉树”的形式进行排布，即上一级马赫-曾德尔干涉仪的每一个输出端口连接下一级的一个马赫-曾德尔干涉仪的一个输入端口，最后一级马赫-曾德尔干涉仪的
个输出端口连接一个第二移相器和一个纠缠多光子源，其中，所述马赫-曾德尔干涉仪包括一个第一移相器，以及与所述第一移相器连接的两个多模干涉仪。
在本发明实施例中，基于集成光学的可编程高维量子计算芯片，可以实现形式为：
的量子计算，其中
是待实现量子操作，表示为多个线性光学幺正变换U的线性组合，其中α
i为线性项系数， 计算初态为
其是多光子高维量子态。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片结构，所述可配置纠缠多光子源生成P种波长的光子，所述P种波长的光子被分别对应路由到P组初态配置线性光学网络，其中，P为自然数，且P≥2。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片结构，所述初态配置线性光学网络可以包含一种多级链式结构。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片结构，所述幺正算符配置线性光学网络可以为呈三角形分布的光网络结构。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片结构，所述投影测量线性光学网络可以包含一种反向树状结构。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片结构，所述第一移相器和第二移相器通过外置经典控制信号来对各路光进行调节，实现路径编码。
根据本发明实施例提供的一种基于集成光学的可编程高维量子计算芯片结构，所述P组初态配置线性光学网络中的每组初态配置线性光学网络具有N个；对应的，所述幺正算符配置线性光学网络被分为P组，每组N个，所述投影测量线性光学网络的数量为P个；其中，每组幺正算符配置线性光学网络与一组初态配置线性光学网络和一个投影测量线性光学网络对应连接。
本发明实施例提供的基于集成光学的可编程高维量子计算芯片结构，通过产生多个光子并对纠缠多光子源的路径进行编码，通过通用线性光学网络对根据波长分别路由的光子进行初态制备、幺正变换、线性组合、合束和投影测量，在单个集成光学芯片上实现了对量子信 息载体的产生、操控以及测量，从而使得实现集成化、小型化、可扩展和可编程的量子计算装置成为可能。
本发明实施例提供的基于集成光学的可编程高维量子计算芯片实现了用于量子计算的光学芯片的集成化，与分立元件光学系统相比，不仅体积显著减小，而且整个系统由于高集成度而具有更好的稳定性及更好的可扩展性。规模化基于集成光学的可编程高维量子计算芯片技术能够支撑基于幺正算符线性组合方案的可扩展实现，构造出完全可编程的高维量子比特计算芯片，实现基于光子的多量子比特量子信息处理。
本发明实施例采用基于路径编码的幺正算符线性组合的方式实现高维量子比特计算，并通过调控可配置纠缠多光子源制备的多光子路径纠缠态和幺正算符配置线性光学网络实现光学幺正变换，从而通过结合片上高维多光子路径纠缠和基于路径编码的线性组合方式实现了高维多比特量子计算。
为了更清楚地说明本发明或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作一简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的基于集成光学的可编程高维量子计算芯片结构示意图之一；
图2是本发明实施例提供的基于集成光学的可编程高维量子计算芯片结构示意图之二；
图3是本发明实施例提供的基于集成光学的可编程高维量子计算芯片中的量子计算过程流程示意图；
图4是本发明实施例提供的用于初态制备的光网络结构示意图；
图5是本发明实施例提供的呈三角形分布的光网络结构示意图；
图6是本发明实施例提供的用于光子合束的光网络结构示意图；
图7是本发明实施例提供的用于投影测量的光网络结构示意图；
图8是本发明实施例提供的双光子纠缠态实现四维量子比特计算的集成光学芯片的结构示意图。
为使本发明的目的、技术方案和优点更加清楚，下面将结合本发明具体实施例及相应的附图对本发明技术方案进行清楚、完整地描述。显然，所描述的实施例仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
以下结合附图，详细说明本发明各实施例提供的技术方案。
如图1所示，本发明实施例提供一种基于集成光学的可编程高维量子计算芯片，包括：可配置纠缠多光子源102，包括线性系数配置网络、N个纠缠多光子源以及波分复用器，设置为产生多个光子，记每个纠缠多光子源产生的纠缠光子的数量为P，不同波长的光子经过波分复用器根据波长分别输出，通过调节线性系数配置网络中的第一移相器与第二移相器，对进入干涉配置网络的泵浦光按输出路径进行配置，得到一组线性配置系数，记为α
1，α
2...α
N。
初态配置线性光学网络104，与所述可配置纠缠多光子源连接，根据所述波分复用器输出光子的波长形成对应的O
1，O
2...O
P，设置为对所述可配置纠缠多光子源输出的光子制备初态。
幺正算符配置线性光学网络106，与所述初态配置线性光学网络O对应连接，形成对应的U
1
(i)，U
2
(i)...U
P
(i)(i＝1，2，...N)，设置为进行幺正变换，经过合束后，实现幺正算符的线性组合，从而得到最终量子态结果：
具体地，线性组合指的是将不同光学网络实现的幺正变换进行线性组合。
投影测量线性光学网络108，与所述幺正算符配置线性光学网络U对应连接，形成对应的T
1，T
2...T
P，设置为对合束后的所述光量子态进行投影测量。
具体地，投影测量是将一个系统状态空间上的一个代表可观测量的厄米算符进行谱分解出多个测量算符，这个测量算符实际上是厄米算符朝着对应本征值生成的本征子空间的投影。
在本发明实施例中，通过对可配置纠缠多光子源模块进行编码，可以配置纠缠多光子源路径的线性项系数，通过初态配置线性光学网络和幺正算符配置线性光学网络，可以配置初态和线性组合的幺正算符，实现了将可制备多光子路径纠缠态的多光子源与可实现幺正变换的通用线性光学网络有机结合、搭配使用。本发明实施例的基于集成光学的可编程高维量子计算芯片具有可编程和可扩展的特点，通过调控纠缠多光子源制备的多光子路径纠缠态和通用线性光学网络实现的光学幺正变换，该基于集成光学的可编程高维量子计算芯片能够制备多光子高维纠缠量子态，实现可编程高维量子比特计算。
基于集成光学的量子芯片技术，目前已经取得了较大的发展。该技术采用半导体微纳加工工艺将分立光学元件集成到单个芯片上，相较于分立光学元件具有体积小、稳定性高、可扩展性强等优势，是实现规模化光量子计算系统的有效途径。
集成光量子芯片领域近年来发展迅速，实现集成光学量子计算所需要的重要组件都已经得到了实验验证，包括片上单光子源及纠缠光子源、片上高精度量子态操控、片上线性光学网络等。以这些基本单元或模块为基础，通过精心设计光学芯片结构，能够在单个芯片上实现对量子信息载体-光子的产生、操控以及测量，从而使得实现集成化、小型化、可扩展和可编程的量子计算装置成为可能。
规模化集成光量子芯片技术能够支撑基于幺正算符线性组合方案的可扩展实现，构造出完全可编程的高维量子比特计算芯片，实现基于光子的多量子比特量子信息处理。同时，基于硅基光波导等集成光学的可编程高维量子计算芯片制造工艺可以与CMOS兼容，本发明实施例中的基于集成光学的可编程高维量子计算芯片能够进一步与传统的CMOS计算芯片融合，用于未来设计实现光电融合、混合架构的光量子信息处理芯片。
在本发明实施例中，基于集成光学的可编程高维量子计算芯片上还可以设置有集成光源，设置为产生光子并输出到线性系数配置网络。基于集成光学的可编程高维量子计算芯片上还可以集成有单光子探测器，设置为对线性光学网络输出的光子进行探测。其中，单光子探测器可以为雪崩光电二极管或者超导纳米线探测器。
其中，初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络都属于通用线性光学网络。
可配置纠缠多光子源、初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均通过所述第一移相器和所述第二移相器实现路径编码。
如图2所示，一个马赫-曾德尔干涉仪包括一个移相器211，与移相器211连接的两个多模干涉仪212。
如图2所示，log
2N级马赫-曾德尔干涉仪的第一级马赫-曾德尔干涉仪接收可配置纠缠多光子源102外部的输入光束221，log
2N级马赫-曾德尔干涉仪根据该输入光束221形成N路光，并将N路光输出到N个纠缠多光子源215。这里，每个纠缠多光子源与一个第二移相器214连接。可配置纠缠光子源模块的所有移相器设置为对线性系数配置网络输出的各路光进行调节，用来配置线性项的系数。具体地，第一移相器和第二移相器通过外置经典控制信号来对各路光进行调节，得到线性项系数。如图2所示，可配置纠缠多光子源102还包括 波分复用器216。
如图2所示，通用线性光学网络包括：
初态配置线性光学网络201，与所述可配置纠缠多光子源连接，根据所述波分复用器输出光子的波长形成对应的O
1，O
2...O
P，设置为对所述可配置纠缠多光子源输出的光子制备初态。
幺正算符配置线性光学网络202(包含合束器231)，与所述初态配置线性光学网络O对应连接，形成对应的U
1
(i)，U
2
(i)...U
P
(i)(i＝1，2，...N)，用于进行幺正变换，经过合束后，实现幺正算符的线性组合，从而得到最终量子态结果：
投影测量线性光学网络203，与所述幺正算符配置线性光学网络U对应连接，形成对应的T
1，T
2...T
P，用于对合束后的所述光量子态进行投影测量。
如图2所示，本发明实施例的基于集成光学的可编程高维量子计算芯片结构包括N个纠缠多光子源和
级马赫-曾德尔干涉仪，N个纠缠多光子源可以标记为S
1，S
2，…，S
N，其中，N为自然数且N≥2。
级马赫-曾德尔干涉仪网络中的马赫-曾德尔干涉仪以二叉树的形式进行排布，即上一级马赫-曾德尔干涉仪的每一个输出端口连接下一级的一个马赫-曾德尔干涉仪的一个输入端口，最后一级马赫-曾德尔干涉仪的N个输出端口连接一个第二移相器和一个纠缠多光子源。
通过设置可配置纠缠多光子源中的第一移相器和第二移相器对线性系数配置网络输出的各路光进行相应配置，可以根据需要获得不同的线性项系数。其中，每个纠缠多光子源产生P个波长不同的光子。
光子经过波分复用器之后分别路由到P组初态配置线性光学网络的入口处，初态配置线性光学网络的作用是制备初态。
由于每个多光子源产生P个波长不同的光子，波长相同的光子被路由到同一组初态配置线性光学网络。因此，如图2所示，本发明实施例中设置初态配置线性光学网络的数量为P组，P组初态配置线性光学网络可以标记为：O
1，O
2，...，O
P。其中，P为自然数且P≥2。每组初态配置线性光学网络均为N个。
通过O
1，O
2，...，O
P可以实现对P个路径编码的t维量子态的制备，其中，初态
O
1，O
2，...，O
P可以为任意的t维线性光学操作，对于芯片中的每一个O
M，均可以执行相同的操作，其中，M为自然数且1≤M≤P。
幺正算符配置线性光学网络的数量为P*N个，其可以分为P组，每组包括N个幺正算符配置线性光学网络。其中，第一组幺正算符配置线性光学网络U
1与第一组初态配置线性光学网络O
1对应连接，第二组幺正算符配置线性光学网络U
2与第二组初态配置线性光学网络O
2对应连接，第P组幺正算符配置线性光学网络U
P与第P组初态配置线性光学网络O
P对应连接。
具体地，幺正算符配置线性光学网络可以被标记为U
1
(1)，U
1
(2)...U
1
(N)，U
2
(1)，U
2
(2)...U
2
(N)，...，U
P
(1)，U
P
(2)...U
P
(N)。
投影测量线性光学网络的数量为P个，每个投影测量线性光学网络对应一组O和U。P个投影测量线性光学网络可以标记为：T
1，T
2，...T
M，...T
P，每个T
M均有t个端口。
初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均是多个可实现t维幺正变换的通用线性光学网络。
在相关技术中，在数学中，幺正变换是保留内积的变换，幺正变换之前的两个向量的内积等于其转换后的内积。幺正变换是使用幺正算符所做的变换，有对基矢的变换，有对算符的变换，可以认为，幺 正变换是两个希尔伯特(Hilbert)空间之间的同构。
具体地，若要实现某种幺正矩阵V
T——这里V
T可以表示为V
T＝α
jU
j，(j＝0，1，2...n-1)，其中，U
j是作用于d维目标(T)子空间的门，α
j是复系数，满足
当受控U
j门可用时，我们可以概率方式实现V
T。α
j被编码为k量子比特控制(C)的初始态
其中n＝2
k，j标记计算基，当最后在计算基中测量所有控制量子位为0时，线路成功。通过将目标量子比特的部分态移动到扩展的Hilbert空间中，控制量子比特可以更加简单地作用到单个量子比特酉中。在本发明实施例中，可以使用基于扩展计算Hilbert空间的技术来实现线性组合线路。
任何量子幺正运算在原理上都可以分解为基本运算的线性和。例如，利用Cartan的KAK分解，可以重写任意两个量子比特幺正运算，将其转换为四个线性项的线性组合，每个线性项为两个单量子比特门的张量积。此外，Cartan分解方法允许n个量子比特幺正运算被重新构造为n个单量子比特门的张量积的线性组合。为了实现量子操作的线性组合，需要为任意未知量子操作添加相干控制，该技术基于用于计算的逻辑希尔伯特空间扩展的门。
每个纠缠多光子源产生P个波长不同的光子。根据所需，首先通过调节可配置纠缠多光子源内的移相器，对N个纠缠多光子源的路径进行编码，获得线性项的系数α
1，α
2，…α
N，得到
光子经过波分复用器之后根据波长不同进入不同的初态配置线性光学网络，最终从不同的投影测量线性光学网络输出。结合后选择技术，在投影测量线性光学网络处进行P个不同波长光子的符合测量
对应在初态配置线性光学网络各个光学网络的入口处生成了相应的多光子多路径纠缠态。在初态配置线性光学网络，波长相同的第M组光子|M>
1|M>
2...|M>
N中每个光子路由到初态配置线性光学网络O
M产生初态，之后路由到幺正算符配置线性光学 网络U
M
(1)，U
M
(2)，...，U
M
(N)进行幺正变换和线性组合，对光路进行合束，经过合束后，实现幺正算符的线性组合，从而得到最终量子态结果：
最终进入投影测量光学网络T
M，对光量子态进行投影测量。其中初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均是多个t维可配置的线性光学网络。根据多维纠缠光子纠缠态的不同，初态配置线性光学网络记为O
1，O
2，...O
M，...O
P，其中，M为自然数且1≤M≤P，每组O
M均有N个。
如3所示，在本发明实施例中，可配置纠缠多光子源和幺正算符配置线性光学网络组成的模块311设置为执行步骤321：进行系数配置和幺正算符线性组合；初态配置线性光学网络312设置为执行步骤322：进行任意初态制备；投影测量线性光学网络313设置为执行步骤323：进行投影测量。这里，步骤321、步骤322和步骤323，可以总称为步骤331：基于光子路径态编码的幺正算符线性组合实现。
在本发明实施例中，初态配置线性光学网络可以是如图4所示的多级链式结构，幺正算符配置线性光学网络可以包含一种如图5所示的t维简化的呈三角形分布的光网络结构，图6是用于光子合束的光网络结构示意图，投影测量线性光学网络可以是如图7所示的反向树状结构。
对于初态配置线性光学网路，初始量子态一般可以设为|0>，即从第一个路径模输入。
基于集成光学的可编程高维量子计算芯片上执行的量子计算操作通过基于路径编码的幺正算符线性组合方案实现，其中线性系数由可配置纠缠多光子源提供，幺正算符线性组合由幺正算符配置线性光学网络提供，幺正算符配置线性光学网络包括N*P个可实现所需t维幺正变换的线性光学网络，波长相同的第M组光子|M>
1|M>
2...|M>
N中每个光子路由到初态配置线性光学网络O
M产生 初态。之后路由到幺正算符配置线性光学网络U
M
(1)，U
M
(2)，...，U
M
(N)完成幺正变换和线性组合。最终路由到投影测量线性光学网络完成投影测量。
本发明实施例中的基于集成光学的可编程高维量子计算芯片是采用集成光学技术将分立的线性光学元件以薄膜形式集成到单个半导体集成芯片上，与分立元件光学系统相比，不仅体积显著减小，而且整个系统由于高集成度而具有更好的稳定性及更好的可扩展性。
基于集成光学的可编程高维量子计算芯片所需要的重要组件都已经分别得到了实验实现，如片上单光子源及纠缠光子源、片上波分复用器、片上通用线性光学网络实现等。基于这些集成芯片组件，用片上集成的光子源来产生纠缠光子，用片上集成的马赫-曾德尔干涉仪及相位控制器组成的线性光学网络来控制光子的行为，再通过片上集成的单光子探测器来进行光子的探测，就可以设计出规模化的基于集成光学的可编程高维量子计算芯片，用于实现复杂的量子信息处理应用。
在本发明实施例中，通过基于集成光学的可编程高维量子计算芯片途径，将片上路径纠缠多光子源和通用的线性光学网络配合使用，建立集成光学芯片结构。具体地，在本发明实施例中，通过片上路径纠缠多光子源产生不同的多光子多路径纠缠态，通过片上通用线性光学网络配置不同光学幺正变换，根据需求实现不同的计算任务；通过进行输出测量可以得到量子信息处理结果，实现通用量子信息计算。
在本发明实施例中，纠缠多光子源模块内的第一移相器和第二移相器通过外置经典控制信号来对各路光进行调节，对N个纠缠多光子源的路径进行编码，得到线性项系数。
在本发明实施例中，P组初态配置线性光学网络中的每组初态配置线性光学网络包含N个；对应的，幺正算符配置线性光学网络被分为P组，每组N个，投影测量线性光学网络的数量为P个；其中， 每组幺正算符配置线性光学网络与一组初态配置线性光学网络和一个投影测量线性光学网络对应连接。
其中初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均是多个t维可配置的通用线性光学网络。
通过基于集成光学的可编程高维量子计算芯片途径，将可制备多光子路径纠缠态的多光子源与可实现幺正变换的线性光学网络有机结合。通过调控多光子源制备多光子路径纠缠态，并通过片上通用线性光学网络配置不同光学幺正变换，该基于集成光学的可编程高维量子计算芯片可以通过制备多粒子高维纠缠光量子态，采用幺正算符线性组合的方式实现可编程高维量子比特计算。
以上描述的是采用P个粒子N维纠缠光量子态实现t维量子比特计算的技术方案。
如图8所示，用于实现四维量子比特计算的基于集成光学的可编程高维量子计算芯片结构示意图。
在可配置纠缠多光子源801中，通过调节可配置纠缠多光子源801内的移相器，对双光子源的初始状态进行配置，获得线性项的系数α
1和α
2。每个纠缠多光子源均产生频率纠缠的光子对，使用波分复用器分开纠缠光子源经四波混频产生的信号光子和闲频光子。可配置纠缠多光子源801产生2种波长的光子，并按照波长路由到初态配置线性光学网络802的不同网络中。
初态配置线性光学网络802根据要求将通过纠缠多光子源得到的纠缠光子制备成任意的初始状态，然后路由到幺正算符配置线性光学网络803，幺正算符配置线性光学网络803包括4个线性光学网络，即U
1
(1)、U
1
(2)、U
2
(1)和U
2
(2)，其均是4维线性光网络结构，可以实现基于路径编码的幺正算符线性组合方案。
幺正算符配置线性光学网络803将线性组合后的光子路由到投影测量线性光学网络804进行投影测量。
在本发明实施例中，首先通过调节可配置纠缠多光子源801内的移相器，对2个纠缠多光子源的路径进行编码，获得线性项的系数α
1，α
2，得到
光子经过波分复用器之后根据波长不同进入不同的线性光学网络，最终从不同的投影测量光学网络输出。结合后选择技术，在投影测量光学网络处进行2个不同波长光子的符合测量
对应在初态配置线性光学网络的各个光学网络的入口处生成了相应的多光子多路径纠缠态。在初态配置线性光学网络，波长相同的第M组光子|M>
1|M>
2中每个光子路由到初态配置线性光学网络OM产生初态，之后路由到幺正算符配置线性光学网络U
M
(1)，U
M
(2)形成线性组合，最后对光路进行合束进入投影测量线性光学网络T
M，进行分析测量。这里，M＝1或2。其中初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均是4维可配置的线性光学网络。
本发明中，基于集成光学的可编程高维量子计算芯片，可以实现形式为：
的量子计算，其中
是待实现量子操作，表示为多个线性光学幺正变换U的线性组合，其中α
i为线性项系数，计算初态为
量子计算的结果是多光子高维量子态。
如图8所示的整个基于集成光学的可编程高维量子计算芯片结构可以通过97个移相器控制。这些移相器通过外置经典控制信号来对各路光进行调节，通过对每个移相器施加不同的外置经典控制信号，从而实现对光量子计算芯片的可编程。最后得到的结果＝
在本发明另一种实施例中，可配置纠缠多光子源包括3个纠缠多光子源，每个纠缠多光子源产生3种波长不同的光子。根据所需，首 先通过调节可配置纠缠多光子源内的移相器，对3个纠缠多光子源的路径进行编码，获得线性项的系数α
1，α
2，α
3，得到
光子经过波分复用器之后根据波长不同进入不同的初态配置线性光学网络，最终从不同的投影测量线性光学网络输出。结合后选择技术，在投影测量光学网络处进行3个不同波长光子的符合测量
对应在初态配置线性光学网络各个光学网络的入口处生成了相应的多光子多路径纠缠态。在初态配置线性光学网络，波长相同的第M组光子|M>
1|M>
2|M>
3中每个光子路由到初态配置线性光学网络O
M产生初态，之后路由到幺正算符配置线性光学网络U
M
(1)，U
M
(2)，U
M
(3)形成线性组合，对光路进行合束最后进入投影测量光学网络T
M，进行分析测量。这里，M为1、2或3。最终获得的结果＝
在本发明又一种实施例中，可配置纠缠多光子源包括2个纠缠多光子源，每个纠缠多光子源产生3种波长不同的光子。根据所需，首先通过调节可配置纠缠多光子源内的移相器，对2个纠缠多光子源的路径进行编码，获得线性项的系数α
1，α
2，得到
光子经过波分复用器之后根据波长不同进入不同的初态配置线性光学网络，最终从不同的投影测量线性光学网络输出。结合后选择技术，在投影测量光学网络处进行3个不同波长光子的符合测量
对应在初态配置线性光学网络各个光学网络的入口处生成了相应的多光子多路径纠缠态。在初态配置线性光学网络，波长相同的第M组光子|M>
1|M>
2中每个光子路由到初态配置线性光学网络O
M产生初态，之后路由到幺正算符配置线性光学网络U
M
(1)，U
M
(2)形成线性组合，对光路进行合束最后进入投影测量光学网络T
M，进行分析测量。这里，M为1、2或3。最终获得的结果＝
本发明实施例采用基于路径编码的幺正算符线性组合的方式实 现高维量子比特计算，并通过调控纠缠多光子源制备的多光子路径纠缠态和幺正算符配置线性光学网络实现的光学幺正变换，从而结合片上高维多光子路径纠缠和基于路径编码的线性组合方式实现了高维多比特量子计算。
其中，通过可配置纠缠多光子源内的移相器配置线性组合的光学幺正运算项的系数，通过初态配置线性光学网络实现每个初始态的制备，通过幺正算符配置线性光学网络实现幺正变换的每一项和线性组合，最后通过投影测量线性光学网络进行投影测量获得计算结果。基于此，该基于集成光学的可编程高维量子计算芯片可以完成多粒子高维纠缠光量子态实现高维量子比特计算。
本发明实施例提供的基于集成光学的可编程高维量子计算芯片，通过产生多个光子并对纠缠多光子源的路径进行编码，通过线性光学网络对根据波长分别路由的光子进行初态制备、幺正变换、线性组合及投影测量，在单个集成光学芯片上实现了对量子信息载体的产生、操控以及测量等功能，从而使得实现集成化、小型化、可扩展和可编程的量子计算装置成为可能。
最后应说明的是：以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。
Claims (10)
- 一种基于集成光学的可编程高维量子计算芯片，包括：可配置纠缠多光子源，包括线性系数配置网络、N个纠缠多光子源以及波分复用器，设置为产生多个光子，记每个纠缠多光子源产生的纠缠光子的数量为P，不同波长的光子经过波分复用器根据波长分别输出，通过调节线性系数配置网络中的第一移相器与第二移相器，对进入干涉配置网络的泵浦光按输出路径进行配置，得到一组线性配置系数，记为α 1，α 2…α N；初态配置线性光学网络O，与所述可配置纠缠多光子源连接，根据所述波分复用器输出光子的波长形成对应的O 1，O 2…O P，设置为对所述可配置纠缠多光子源输出的光子制备初态；幺正算符配置线性光学网络U，与所述初态配置线性光学网络O对应连接，形成对应的U 1 (i)，U 2 (i)…U P (i)(i＝1，2，…N)，设置为进行幺正变换，经过合束后，实现幺正算符的线性组合，从而得到最终量子态结果：
投影测量线性光学网络T，与所述幺正算符配置线性光学网络U对应连接，形成对应的T 1，T 2…T P，设置为对合束后的所述光量子态进行投影测量。 - 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络都属于通用线性光学网络。
- 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述可配置纠缠多光子源、初态配置线性光学网络、幺正算符配置线性光学网络和投影测量线性光学网络均通过所述第一移相器和所述第二移相器实现路径编码。
- 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述线性系数配置网络包括log 2N级马赫-曾德尔干涉仪， 以二叉树的形式进行排布，即上一级马赫-曾德尔干涉仪的每一个输出端口连接下一级的一个马赫-曾德尔干涉仪的一个输入端口，最后一级马赫-曾德尔干涉仪的个输出端口连接一个第二移相器和一个纠缠多光子源，其中，所述马赫-曾德尔干涉仪包括一个第一移相器，以及与所述第一移相器连接的两个多模干涉仪。
- 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述可配置纠缠多光子源生成P种波长的光子，所述P种波长的光子被分别对应路由到P组初态配置线性光学网络，其中，P为自然数，且P≥2。
- 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述初态配置线性光学网络包含一种多级链式结构。
- 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述幺正算符配置线性光学网络为呈三角形分布的光网络结构。
- 根据权利要求1所述的基于集成光学的可编程高维量子计算芯片，其中，所述投影测量线性光学网络包含一种反向树状结构。
- 根据权利要求3所述的基于集成光学的可编程高维量子计算芯片，其中，所述第一移相器和第二移相器通过外置经典控制信号来对各路光进行调节，实现路径编码。
- 根据权利要求5所述的基于集成光学的可编程高维量子计算芯片，其中，所述P组初态配置线性光学网络中的每组初态配置线性光学网络具有N个；对应的，所述幺正算符配置线性光学网络被分为P组，每组N个，所述投影测量线性光学网络的数量为P个；其中，每组幺正算符配置线性光学网络与一组初态配置线性光学网络和一个投影测量线性光学网络对应连接。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US18/289,747 US20240265287A1 (en) | 2021-09-28 | 2022-09-27 | Integrated Photonic-Based Programmable High-Dimensional Quantum Computation Chip Structure |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111143350.X | 2021-09-28 | ||
| CN202111143350.XA CN113935493A (zh) | 2021-09-28 | 2021-09-28 | 一种基于集成光学的可编程高维量子计算芯片结构 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023051550A1 true WO2023051550A1 (zh) | 2023-04-06 |
Family
ID=79277198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/121834 WO2023051550A1 (zh) | 2021-09-28 | 2022-09-27 | 一种基于集成光学的可编程高维量子计算芯片结构 |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US20240265287A1 (zh) |
| CN (1) | CN113935493A (zh) |
| WO (1) | WO2023051550A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116362342A (zh) * | 2023-04-19 | 2023-06-30 | 中国人民解放军军事科学院国防科技创新研究院 | 面向哈密顿量含时演化模拟的集成光学量子计算芯片结构 |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113935493A (zh) * | 2021-09-28 | 2022-01-14 | 中国人民解放军军事科学院国防科技创新研究院 | 一种基于集成光学的可编程高维量子计算芯片结构 |
| CN114640401B (zh) * | 2022-03-01 | 2024-05-17 | 南京理工大学 | 一种量子网络中非局域的多体纠缠态的并行制备方法 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170286858A1 (en) * | 2016-03-31 | 2017-10-05 | Board Of Regents, The University Of Texas System | System and method for emulation of a quantum computer |
| CN111478735A (zh) * | 2020-04-13 | 2020-07-31 | 中国人民解放军国防科技大学 | 一种基于集成光学的通用量子漫步模拟芯片结构 |
| CN112946968A (zh) * | 2021-02-01 | 2021-06-11 | 电子科技大学 | 一种混合集成的光通信波段片上量子纠缠源 |
| CN113935494A (zh) * | 2021-09-28 | 2022-01-14 | 中国人民解放军军事科学院国防科技创新研究院 | 面向分布式安全量子信息处理的集成光学芯片系统 |
| CN113935493A (zh) * | 2021-09-28 | 2022-01-14 | 中国人民解放军军事科学院国防科技创新研究院 | 一种基于集成光学的可编程高维量子计算芯片结构 |
-
2021
- 2021-09-28 CN CN202111143350.XA patent/CN113935493A/zh active Pending
-
2022
- 2022-09-27 WO PCT/CN2022/121834 patent/WO2023051550A1/zh active Application Filing
- 2022-09-27 US US18/289,747 patent/US20240265287A1/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170286858A1 (en) * | 2016-03-31 | 2017-10-05 | Board Of Regents, The University Of Texas System | System and method for emulation of a quantum computer |
| CN111478735A (zh) * | 2020-04-13 | 2020-07-31 | 中国人民解放军国防科技大学 | 一种基于集成光学的通用量子漫步模拟芯片结构 |
| CN112946968A (zh) * | 2021-02-01 | 2021-06-11 | 电子科技大学 | 一种混合集成的光通信波段片上量子纠缠源 |
| CN113935494A (zh) * | 2021-09-28 | 2022-01-14 | 中国人民解放军军事科学院国防科技创新研究院 | 面向分布式安全量子信息处理的集成光学芯片系统 |
| CN113935493A (zh) * | 2021-09-28 | 2022-01-14 | 中国人民解放军军事科学院国防科技创新研究院 | 一种基于集成光学的可编程高维量子计算芯片结构 |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116362342A (zh) * | 2023-04-19 | 2023-06-30 | 中国人民解放军军事科学院国防科技创新研究院 | 面向哈密顿量含时演化模拟的集成光学量子计算芯片结构 |
| CN116362342B (zh) * | 2023-04-19 | 2024-02-23 | 中国人民解放军军事科学院国防科技创新研究院 | 面向哈密顿量含时演化模拟的集成光学量子计算芯片结构 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113935493A (zh) | 2022-01-14 |
| US20240265287A1 (en) | 2024-08-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023051550A1 (zh) | 一种基于集成光学的可编程高维量子计算芯片结构 | |
| TWI831932B (zh) | 混合類比-數位處理器及該處理器中執行的方法 | |
| Qiang et al. | Large-scale silicon quantum photonics implementing arbitrary two-qubit processing | |
| TWI825452B (zh) | 光電計算系統 | |
| US11984934B2 (en) | Integrated photonic chip structure for universal quantum walk simulation | |
| Qu et al. | Inverse design of an integrated-nanophotonics optical neural network | |
| CN113935494B (zh) | 面向分布式安全量子信息处理的集成光学芯片系统 | |
| Katumba et al. | A multiple-input strategy to efficient integrated photonic reservoir computing | |
| US11681201B2 (en) | Training of photonic reservoir computing systems | |
| Xiao et al. | Large-scale and energy-efficient tensorized optical neural networks on III–V-on-silicon MOSCAP platform | |
| Peserico et al. | Integrated photonic tensor processing unit for a matrix multiply: a review | |
| CN113496281A (zh) | 光电计算系统 | |
| Simon et al. | Quantum simulation of discrete-time Hamiltonians using directionally unbiased linear optical multiports | |
| Saharia et al. | An elementary optical logic circuit for quantum computing: a review | |
| US11709520B2 (en) | Path-number-balanced universal photonic network | |
| US20230351167A1 (en) | Frequency multiplexed photonic neural networks | |
| Filipovich et al. | Monolithic silicon photonic architecture for training deep neural networks with direct feedback alignment | |
| Wu et al. | Programmable integrated photonic coherent matrix: Principle, configuring, and applications | |
| Shokraneh et al. | High-performance programmable MZI-based optical processors | |
| CN114970868A (zh) | 空间并行的相干伊辛机及包含其的求解器 | |
| CN115526325B (zh) | 一种高维光量子计算方法及芯片 | |
| CN116362342B (zh) | 面向哈密顿量含时演化模拟的集成光学量子计算芯片结构 | |
| Chen et al. | Hypermultiplexed integrated-photonics-based tensor optical processor | |
| Zhang et al. | An optical computing chip for executing complex-valued neural network and its on-chip training | |
| Ou et al. | Hypermultiplexed Integrated Tensor Optical Processor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22874935 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 18289747 Country of ref document: US |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |