Error correction method and device of NV color center Ising machine, electronic equipment and readable medium
By mapping electron spin states to multiple nuclear spins for redundant encoding in the NV center Ising machine, and by using majority voting and dynamic adjustment of control pulse parameters, the error accumulation problem in the measurement feedback iteration process of the NV center Ising machine is solved, achieving efficient error correction and improved system robustness, thus breaking through the bottleneck of large-scale application.
Patent Information
- Application Number
- CN202610801530.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-06-04
- Publication Date
- 2026-08-25
AI Technical Summary
The NV center Ising machine suffers from error accumulation during measurement feedback iteration, leading to a decrease in fidelity. Furthermore, the resource overhead of general quantum error correction is relatively large, making it difficult to achieve efficient error correction in resource-constrained NV center systems.
The electronic spin state of the NV color center is mapped to multiple nuclear spins for redundancy encoding. The corrected spin state is determined by majority voting rules, and the control pulse parameters are dynamically adjusted according to the statistical deviation during the error correction process. The nuclear spins naturally present around the NV color center are used as redundancy carriers, without the need to introduce additional hardware resources.
It effectively corrects flip errors during the readout process, improving the single readout fidelity from 95% to 99.3%, and raising the overall fidelity of 50 logic spins to a usable level, breaking through the bottleneck of large-scale application. Furthermore, it enhances the robustness of the system through a real-time calibration mechanism.
Smart Images

Figure CN122635587A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of quantum computing technology, and in particular to an error correction method, apparatus, electronic device, and readable medium for an NV color center Ising machine. Background Technology
[0002] The NV-center Ising machine is a dedicated quantum computing device that uses the electron spins of diamond nitrogen-vacancy (NV) centers as Ising spin carriers. It employs optically probed magnetic resonance (ODMR) readout and measurement feedback iteration to solve combinatorial optimization problems of the Ising model. However, readout errors in each measurement round propagate to the next iteration via feedback signals. As the number of spins increases, the probability of a correct overall configuration decreases sharply, and once locked into an incorrect configuration, subsequent iterations cannot self-correct, severely limiting the large-scale application of the Ising machine.
[0003] In related technologies, universal quantum error correction schemes (such as surface codes) are commonly used to correct errors in Ising machines to improve their reliability. However, universal quantum error correction protects the complete information of any quantum state, including quantum correlations such as superposition and entanglement, requiring a large number of auxiliary physical bits to achieve the error correction function. For NV center systems, each NV center has only a limited number of addressable nuclear spins, making it difficult to withstand the huge resource overhead of universal quantum error correction schemes. Therefore, how to achieve efficient error suppression in resource-constrained NV center systems has become an urgent technical problem to be solved.
[0004] There are currently no effective solutions to the problems of decreased fidelity due to error accumulation during the measurement feedback iteration process of the NV color center Ising machine in related technologies, as well as the high resource overhead of general quantum error correction. Summary of the Invention
[0005] The present invention provides an error correction method, apparatus, electronic device and readable medium for an NV color center Ising machine, which at least solves the problems in the related art of error accumulation leading to decreased fidelity in the measurement feedback iteration process of the NV color center Ising machine, and the large resource overhead of general quantum error correction.
[0006] The present invention provides an error correction method for an NV color center Ising machine, comprising: The electronic spin states of the NV color center are mapped to multiple nuclear spins for redundant encoding; Obtain the readout results of the multiple nuclear spins; Based on multiple readout results, the corrected spin state is determined by a majority voting rule; The control pulse parameters are dynamically adjusted based on the statistical deviations observed during the error correction process.
[0007] In some embodiments, the mapping of the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding includes: By using a controlled NOT gate operation between the electron spin of the NV color center and multiple nuclear spins around the NV color center, the electron spin state corresponding to the electron spin is sequentially mapped to the multiple nuclear spins for redundant encoding.
[0008] In some embodiments, the controlled NOT gate operation is implemented based on hyperfine coupling between the electron spins of the NV color center and the plurality of nuclear spins, wherein the Hamiltonian of the hyperfine coupling is:
[0009] in: For the first The hyperfine coupling constant of the nth nuclear spin is used to represent the nth nuclear spin. The strength of the interaction between nuclear spin and electron spin; For electron spin Quantity; For the first Each nuclear spin Quantity.
[0010] In some embodiments, obtaining the readout results of the plurality of nuclear spins includes: Based on the difference in hyperfine coupling constants between different nuclear spins, the state of each nuclear spin can be read out independently using frequency-selective pulses.
[0011] In some embodiments, dynamically adjusting the control pulse parameters based on the statistical deviation during the error correction process includes: Select one of the plurality of nuclear spins as a reference nuclear spin; Monitor the statistical deviation between the majority voting results and the readout results of the reference nuclear spin in multiple consecutive iterations; When the statistical deviation exceeds a preset threshold, the control pulse parameters are adjusted.
[0012] In some embodiments, monitoring the statistical deviation between the majority voting result and the readout result of the reference nuclear spin in consecutive iterations includes: A fixed-width sliding window is used to statistically analyze the average deviation between the majority voting results and the readout results of the reference nuclear spin in multiple consecutive iterations.
[0013] In some embodiments, the control pulse parameters include a control pulse angle, and fine-tuning the control pulse angle includes:
[0014] in: Indicates the rotation angle of the microwave pulse; Indicates the direction and magnitude of the detected systematic drift; This indicates the step size for each correction.
[0015] In some embodiments, the number of the plurality of nuclear spins is 3 or 5.
[0016] In some embodiments, the effective fidelity of the spin state determined by the majority voting rule is:
[0017] in: Indicates effective fidelity; Indicates the number of nuclear spins; Indicates the first The fidelity of the results read out each time.
[0018] An embodiment of the present invention also provides an error correction device for an NV color center Ising machine, comprising: The state mapping module is configured to map the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding; The readout module is configured to acquire the readout results of the multiple nuclear spins; The voting determination module is configured to determine the corrected spin state based on multiple readout results using a majority voting rule. The feedback calibration module is configured to dynamically adjust the control pulse parameters based on the statistical deviation during the error correction process.
[0019] Embodiments of the present invention also provide an electronic device, including: a processor, and a memory storing a program, the program including instructions that, when executed by the processor, cause the processor to perform the methods described above.
[0020] Embodiments of this invention also provide a non-transitory machine-readable medium storing computer instructions for causing a computer to perform the methods described above.
[0021] This invention provides an error correction method, apparatus, electronic device, and readable medium for an NV-center Ising machine. By utilizing the naturally occurring nuclear spins around the NV-center as redundancy carriers, error correction can be achieved without introducing additional hardware resources. By mapping the electronic spin states of the NV-center to multiple nuclear spins for redundant encoding, and determining the corrected spin state based on a majority voting rule, flipping errors during readout are effectively corrected. When using three nuclear spins, the effective fidelity can be increased to F0 while maintaining a single readout fidelity of F=95%. eff=99.3%, the overall fidelity of the 50 logic spins has been improved from near 0 to a usable level, a significant improvement that breaks through the bottleneck of large-scale application. At the same time, by monitoring the statistical deviation during the error correction process and dynamically adjusting the control pulse parameters, real-time calibration of systematic errors has been achieved, further improving the robustness of the system. Attached Figure Description
[0022] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other embodiments based on these drawings without creative effort.
[0023] Figure 1 This is a flowchart illustrating an error correction method for an NV color center Ising machine according to an embodiment of the present invention.
[0024] Figure 2 This is a curve showing the overall fidelity of the Ising machine with / without error correction as a function of the number of spins in an embodiment of the present invention.
[0025] Figure 3 This is an embodiment of the NV color center and its surrounding area created by the present invention. A schematic diagram of nuclear spin error correction resources.
[0026] Figure 4 This is a timing diagram of an error correction method for an NV color center Ising machine according to an embodiment of the present invention.
[0027] Figure 5 This is a schematic diagram of the sliding window mechanism for real-time feedback pulse calibration according to an embodiment of the present invention.
[0028] Figure 6 This is a schematic diagram of the structure of an error correction device for an NV color center Ising machine, according to an embodiment of the present invention.
[0029] Figure 7 This is a schematic diagram of the structure of an electronic device created by the present invention. Detailed Implementation
[0030] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.
[0031] To address the issues of decreased fidelity due to error accumulation during measurement feedback iteration in NV center Ising machines and the high resource overhead of general quantum error correction, embodiments of this invention provide an error correction method, apparatus, electronic device, and readable medium for NV center Ising machines.
[0032] like Figure 1 As shown, an embodiment of the present invention provides an error correction method for an NV color center Ising machine, comprising: Step S101: Map the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding.
[0033] Specifically, NV centers are luminescent defects in a diamond lattice formed by nitrogen atoms replacing carbon atoms and interacting with neighboring vacancies. They possess multiple charge states, including negative (NV-) and neutral (NV0). In the Ising machine based on NV centers, the NV centers serve as the logic bit carriers, and their electronic spin states carry computational information. In this embodiment, the naturally occurring nuclear spin resources within the NV center lattice are utilized, for example... nuclear spin or Nuclear spin (in this embodiment, it is referred to as nuclear spin) (Taking nuclear spins as an example). These nuclear spins and electron spins have hyperfine interactions, which can serve as natural quantum storage units. In this embodiment, the current state of the electron spin (e.g., the |0> state or the |1> state, corresponding to 0 or 1 in classical information) is "copied" to multiple surrounding nuclear spins to form redundant backups, thus realizing redundant encoding based on repetition code. It should be understood that what the Ising machine needs to protect during its iterative evolution is the classical spin population. This refers to the probability distribution of whether the spin is in the ground state or an excited state, rather than quantum correlation information such as quantum superposition or entanglement. Therefore, this embodiment uses classical repetition codes for error correction, rather than universal quantum error-correcting codes (such as surface codes or phase-flip codes) that protect arbitrary quantum states. Repetition codes require only a small number of physical bits (i.e., nuclear spins) to achieve effective error correction, while universal quantum error correction resources in existing technologies include at least thousands of auxiliary bits, which is too costly and unacceptable for NV center systems. This embodiment greatly reduces resource overhead and solves the problem that universal quantum error correction schemes in existing technologies are difficult to implement in resource-constrained NV center systems. This embodiment achieves error correction resource deployment with "zero additional hardware cost" by utilizing naturally occurring nuclear spins as redundancy carriers.
[0034] In some embodiments, mapping the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding includes: By using controlled NOT gate operations between the electron spin of the NV color center and multiple nuclear spins around the NV color center, the electron spin state corresponding to the electron spin is sequentially mapped to multiple nuclear spins for redundant encoding.
[0035] like Figure 3 The image shows the NV color center and its surrounding area. A schematic diagram of nuclear spin error correction resources, whose main structure is an NV color center in a diamond lattice. Figure 3 In the middle, the blue sphere N represents a nitrogen atom, the white sphere V represents a vacancy, and the gray tetrahedral network represents the surrounding carbon lattice. The three green spheres represent... The nuclear spins are designated as C1, C2, and C3, respectively, and are located at different distances from the NV color center, i.e., the distances of the nuclear spins C1, C2, and C3 from the NV color center are n1, n2, and n3, respectively. C1, C2, and C3 correspond to the hyperfine coupling constants A1, A2, and A3, respectively, and can be set, for example, A1 = 130 kHz, A2 = 85 kHz, and A3 = 45 kHz. In this embodiment, the electron spin is used as the control bit, and the nuclear spin is used as the target bit. During the repetition code encoding process, the logic of the controlled NOT gate (CNOT) operation between the electron spin and the nuclear spin is as follows:
[0036] Specifically: when the electron spin is in the |0> state, the nuclear spin state remains unchanged; when the electron spin is in the |1> state, the nuclear spin state flips (from |0> to |1>, or from |1> to |0>). For example... Figure 3 In and Through this operation, the logical state of the electron spin is "copied" to the nuclear spin, achieving zero additional hardware cost—using naturally existing nuclear spins. It should be understood that "copied" here refers to transferring classical information (spin population) to the nuclear spin. This involves redundant backups, rather than quantum state copying that violates the no-cloning theorem. In this embodiment, the mapping process is performed sequentially, such as... Figure 4 The diagram shown is a timing diagram of the error correction method in this embodiment. In terms of timing, it utilizes... The function implements the Ising Evolution. During error correction, state mapping is first performed in Step 1, applying a CNOT gate operation (CNOT1) for the first nuclear spin to encode it. Then, a CNOT gate operation (CNOT2) is applied for the second nuclear spin to encode it, and so on, until all nuclear spins involved in redundant encoding have been mapped. Each CNOT gate operation can be implemented by combining microwave pulses (acting on electron spins) and radio frequency pulses (acting on nuclear spins). The specific parameters of the pulse sequence (such as frequency, phase, and duration) are precisely designed based on the hyperfine coupling strength between the electron spins and each nuclear spin. This CNOT-based mapping method in this embodiment utilizes the naturally occurring hyperfine interactions of NV color centers, eliminating the need for additional coupling structures and achieving extremely high control fidelity and execution efficiency.
[0037] In some embodiments, the controlled NOT gate operation is implemented based on the hyperfine coupling between the electron spin of the NV color center and multiple nuclear spins, where the hyperfine coupling Hamiltonian is:
[0038] in: For the first The hyperfine coupling constant of the nth nuclear spin is used to represent the nth nuclear spin. The strength of the interaction between nuclear spin and electron spin; For electron spin Quantity; For the first Each nuclear spin Quantity.
[0039] Specifically, in this embodiment, the Hamiltonian is obtained through ultrafine coupling. This can describe the magnetic coupling interaction, where the NV electron spin generates a local magnetic field, and the surrounding... The nuclear spin senses this magnetic field, causing its energy level to shift. The magnitude of this shift is determined by the hyperfine coupling constant. Decision, and It also depends on the spatial position of the nuclear spin relative to the NV color center (the closer the distance and the more suitable the angle). (The larger the value). In this embodiment, a precisely designed pulse sequence is used to achieve controlled flipping (state mapping) of electron spin to nuclear spin through hyperfine coupling. Different nuclear spins have different hyperfine coupling constants. This allows for different resonant frequencies, which can be controlled separately using radio frequency pulses of different frequencies, achieving frequency selective addressing; narrowband radio frequency pulses are aimed at specific frequencies, exciting only the corresponding nuclear spins without crosstalk.
[0040] In some embodiments, the mapping time for mapping the electronic spin states of the NV color center to multiple nuclear spins is: , The total time for sequential mapping of each nuclear spin is In this embodiment, the following settings are provided. Then the mapping time When the number of nuclear spins The time mapping time is approximately .
[0041] Step S102: Obtain readout results for multiple nuclear spins.
[0042] Specifically, after completing the state mapping between electron spin and nuclear spin, the system needs to measure the encoded nuclear spin states. Because nuclear spins have a relatively long coherence time, the stored information can be retained for a sufficiently long time for readout. In this embodiment, a specific manipulation pulse sequence is applied to read the state of each nuclear spin involved in redundant encoding, thereby obtaining a set of raw measurement data.
[0043] In some embodiments, obtaining readout results for multiple nuclear spins includes: Based on the difference in hyperfine coupling constants between different nuclear spins, the state of each nuclear spin can be read out independently using frequency-selective pulses.
[0044] Specifically, in the NV center system, different nuclear spins have different spatial positions in the diamond lattice (different distances and angles from the NV center). The nuclear spins naturally distribute in the range of 10–200 kHz, and their hyperfine coupling constants with electron spins exhibit natural differences, i.e. .like Figure 3 As shown, nuclear spins C1, C2, and C3 have different hyperfine coupling constants A1, A2, and A3, respectively. This embodiment employs frequency-selective pulses for readout, meaning that radio frequency pulses or microwave pulses with specific frequencies are applied to different nuclear spins. For example, when the state of nuclear spin C1 needs to be read out, the system applies a pulse with a frequency precisely aligned to the C1 resonant frequency. Due to frequency detuning, other nuclear spins (C2, C3) do not respond to this pulse, thus avoiding crosstalk during the readout process. Figure 4 As shown, in terms of timing, during the independent readout stage of Step 2, the system applies selective pulses of different frequencies sequentially or in parallel. Figure 4 The green arrows in the image indicate laser pulse markers, corresponding to frequencies of [missing information]. f 1. f 2. f3) Separately acquire the state information of each nuclear spin. This independent readout method ensures that the measurement results of each nuclear spin do not interfere with each other, providing an accurate data basis for subsequent majority voting. Compared with the global readout method, the frequency difference provides a physical basis for independent addressing. Frequency-selective readout significantly improves the measurement accuracy of multi-bit systems and enables high-fidelity error correction for NV color-center Ising machines.
[0045] In some embodiments, the selective pulse is a narrowband microwave or radio frequency pulse with a bandwidth less than [missing information]. Each nuclear spin is read out independently. Each readout takes approximately [time missing]. When the number of nuclear spins The total readout time is approximately .
[0046] Understandably, in this embodiment, high-fidelity state mapping is achieved through CNOT gate operation, and crosstalk-free independent readout is achieved using frequency-selective pulses, thus ensuring the feasibility and reliability of the error correction scheme from a physical level.
[0047] Step S103: Based on multiple readout results, determine the corrected spin state through a majority voting rule.
[0048] In some embodiments, for One readout result A majority vote will be used, and the voting rules are as follows:
[0049] in: This represents the corrected spin state estimate; Indicates the first The readout results of each nuclear spin.
[0050] Specifically, in this embodiment, since the same electronic spin state has been copied to multiple nuclear spins in step S101, the readout results of these nuclear spins should be consistent in the absence of errors. However, in actual measurement, environmental noise or manipulation errors may cause individual nuclear spin readout results to flip. In step S103, by comparing the readout results of multiple nuclear spins, the state that appears most frequently is determined to be the correct spin state. For example, if three nuclear spins ( Encode the result, and read the result including {+1,+1,+1}, {+1,+1, 1}、{+1, 1, 1} or { 1, 1, 1}, then according to the majority voting rule, {+1,+1,+1} and { 1, 1, 1) Neither of the two read results needs to be corrected, {+1,+1, The correctly determined spin state in the readout result of {1} is "+1", {+1, 1, The correctly determined spin state in the readout result of 1} is " The error correction logic, based on classical information, is simple and efficient, effectively filtering out random readout errors and significantly improving the system's effective fidelity.
[0051] In some embodiments, an FPGA (Field Programmable Gate Array) is used for... One readout result Perform majority voting. For example... Figure 4 As shown, when At that time, 3 readout results The input signal is used for majority voting in Step 3 to obtain the corrected spin state estimate. Thus, only simple sign determination is required, and the computational delay of this process is <100 ns, which is negligible compared to the total time of a single iteration of the Ising machine.
[0052] Step S104: Dynamically adjust the control pulse parameters based on the statistical deviation during the error correction process.
[0053] Specifically, in the iterative error correction process over multiple rounds, the statistical deviation between the majority voting results and the readout results of individual nuclear spins actually reflects the drift of the system's control parameters. For example, if the readout result of a certain nuclear spin continues to deviate towards an erroneous state for a long period of time, this often indicates that the control pulse (such as a microwave pulse or radio frequency pulse) applied to that nuclear spin has an angle deviation or frequency drift. This statistical deviation is monitored in real time through step S104, and when the statistical deviation exceeds a preset safety range, a fine-tuning calibration of the control pulse parameters is automatically triggered. Through this feedback calibration mechanism, the system can adaptively compensate for long-term drift caused by environmental changes or hardware aging, ensuring the reliability of the error correction mechanism itself and achieving synergistic optimization of "error correction and calibration." Figure 4 As shown, in terms of timing, the feedback pulse and angle are implemented in the feedback calibration stage of step 4. (Feedbackpulse with angle θ i ).
[0054] In some embodiments, the control pulse parameters are dynamically adjusted based on the statistical deviation during the error correction process, including: Step S1041: Select one nuclear spin from multiple nuclear spins as a reference nuclear spin.
[0055] Specifically, to assess the accuracy of the current control pulse, a benchmark needs to be selected for comparison after the majority vote. In this embodiment, a reference spin is arbitrarily selected from the multiple nuclear spins participating in redundancy coding. It should be understood that the selection of the reference spin can be fixed (e.g., always selecting the first spin) or rotated, as long as the consistency of the reference object is ensured within the statistical window. Although a single readout result of the reference spin may contain random errors, statistically, its deviation from the majority vote result should tend to zero. If the deviation exhibits a systematic shift, it indicates that the control parameters are drifting.
[0056] Step S1042: Monitor the statistical deviation between the majority voting results and the readout results of the reference nuclear spin in multiple consecutive iterations.
[0057] In some embodiments, monitoring the statistical deviation between the majority voting result and the readout result of the reference nuclear spin in consecutive iterations includes: A fixed-width sliding window is used to statistically analyze the average deviation between the majority voting result and the readout result of the reference nuclear spin in multiple consecutive iterations.
[0058] Specifically, during operation, the system continuously records the majority vote result and the readout result of the reference nuclear spin in each iteration, and calculates the difference between the two as the statistical bias. Since the iterative evolution of the Ising machine is a continuous process, a single bias may be caused by random noise and is insufficient as a calibration basis. Therefore, this embodiment uses statistical methods to smooth the bias data to filter out random fluctuations and extract systematic trends. In this embodiment, the system maintains a width of... A time window that always covers the most recent Round-repeating data. For example... Figure 4 As shown, the window width is set in step 4 of the timing sequence. (Sliding window( ), combined with, for example Figure 5 The width of the sliding window mechanism in the schematic real-time feedback pulse calibration is... Sliding window of width This indicates that the system always calculates the average deviation of the most recent 20 iterations. For example... Figure 5 As shown, in the first In each iteration, the system calculates the difference between the majority vote result and the reference nucleus spin readout result, stores this difference in the sliding window, and simultaneously moves out of the window. The system uses older data from previous cycles. In this way, the system calculates the average deviation within the window in real time. Compared to averaging all historical data, the sliding window more sensitively reflects recent changes in system status and avoids over-smoothing of historical data.
[0059] In some embodiments, continuous monitoring The statistical deviation between the majority vote result and the single result in the round of iteration is:
[0060] in: This represents the average deviation (statistical deviation) of the sliding window, such as the average difference between the voting results over 20 rounds and a single result, used to detect long-term trend drift. This represents the number of iterations of the Ising machine, i.e., the width of the sliding window; Indicates the iteration round index, for example, from the current round. Going back 20 rounds, that is ; Indicates the first The majority vote result of the round, that is, the reliable estimate after duplicate code error correction, reflects the true orientation trend of electron spin; Indicates the first The first readout of the nuclear spin is the raw, uncorrected readout, which retains the systematic bias caused by the drift of the control pulse.
[0061] Step S1043: When the statistical deviation exceeds the preset threshold, the adjustment of the control pulse parameters is triggered.
[0062] Specifically, the system has a preset safety threshold range (e.g., set to a certain value). When the detected statistical deviation Within the threshold range ( When the deviation is within a normal range, the system considers it a normal random fluctuation and does not intervene. Once the deviation is statistically analyzed... Exceeding the preset threshold ( If the system determines that the control pulse parameters have significantly drifted (e.g., Rabi frequency drift of the microwave pulse or pulse duration deviation), it automatically triggers a calibration program to fine-tune the control pulse parameters back to the normal range. Figure 5 In the sliding window mechanism for real-time feedback pulse calibration shown, the elliptical structure in the top row represents the majority vote result. The row of oval results below represents the reference results. The Compute module receives all the differences within the sliding window. Calculate the average value Then it enters a comparator, where comparisons are made. If... This triggers a fine-tuning of the control pulse parameters (Update Rule); the microwave pulse generator output at the output end carries an update... Microwave pulse with updated To apply for the next round of iterations, through methods such as... Figure 5 The sliding window mechanism for real-time feedback pulse calibration shown achieves adaptive suppression of long-term drift. For example... Figure 5 As shown, the blue elliptical structure represents the "+1" state, and the red elliptical structure represents the "+1" state. 1” state, if the average deviation (statistical deviation) The readout of the reference nuclear spin is consistently biased towards " If the majority of votes favor the "+1" state, then the angle of the manipulation pulse applied to the spin of that reference nucleus may have a negative bias. In this case, the system automatically adjusts the parameters of the microwave pulse generator, such as correcting the Rabi frequency of the pulse or fine-tuning the pulse duration, to compensate for this bias. In this embodiment, this threshold-based closed-loop feedback mechanism achieves synergistic optimization of error correction and calibration, effectively avoiding system jitter caused by frequent adjustments, suppressing long-term system errors caused by environmental factors (such as temperature drift and magnetic field fluctuations), significantly improving the robustness of the Ising machine, and ensuring operational stability.
[0063] In some embodiments, the control pulse parameters include the control pulse angle, and fine-tuning the control pulse angle includes:
[0064] in: Indicates the rotation angle of the microwave pulse; Indicates the direction and magnitude of the detected systematic drift (statistical bias); This indicates the step size for each correction.
[0065] In some embodiments, with the first Taking an NV color center as an example, assuming the current pulse angle (nominal) Pulse), detected by sliding window (This indicates that the reference nuclear spin system is systematically positive, and the actual pulse flip angle is insufficient), then the corresponding fine-tuning is: This involves a slight positive adjustment of the pulse angle to compensate for insufficient systemic flipping. This correction is minimal, preventing overshoot oscillations, but its continuous accumulation effectively suppresses long-term drift. In this implementation, the correction step size is set to [value missing]. It can balance sensitivity and stability, so that the multi-round window averaging has smoothed random noise, and the fine-tuning amplitude is matched with the drift magnitude.
[0066] In some embodiments, the number of nuclear spins is 3 or 5.
[0067] Specifically, the number of nuclear spins directly determines the error correction capability and resource overhead of the repeating code; each NV color center can address [number of spins]. The number of nuclear spins depends on isotopic abundance and distance in a diamond lattice. Its natural abundance is 1.1%, which is 1.1% of the diamond lattice. Natural ingredients account for only 1.1%, the rest are... Elements; within a range of less than 1 nm from the NV color center The nuclear spin averages 1-3, and can be increased through isotopic enrichment (artificial enhancement). The proportion can be increased to 10% (3–10), or enriched to 50% (10–30). In repetitive code logic, if the following is adopted... Encoding each nuclear spin can correct (R-1) / 2 bit flip errors. When the number of nuclear spins... At that time, the system can correct a 1-bit error; when the number of nuclear spins... At that time, the system can correct 2-bit errors. This embodiment selects 3 or 5 nuclear spins based on a balance between error correction capability and hardware resource cost. On the one hand, in the NV color center system, the natural abundance (generally...) On average, each NV color center has only a small number (2–10) of addressable nuclear spins around it. Value too large (e.g.) (or more), although the error correction capability is enhanced, it requires excessive nuclear spin resources and increases the operation time of state mapping and readout, reducing computational efficiency; on the other hand, if Value too small (e.g.) If there is no redundancy, then error correction cannot be achieved. It can already meet the error correction requirements of most Ising machine iteration scenarios, and has the lowest resource overhead; This is suitable for critical computing nodes that require extremely high fidelity. The repeating code only requires 3 nuclear spins, which is sufficient for the natural abundance. The repeating code requires 5 nuclear spins, and enrichment of 10% is sufficient. It should be understood that in other embodiments, the number of nuclear spins can also be adaptively adjusted according to the specific hardware resource availability and the fidelity requirements of the computing task, for example, choosing 7 or more, as long as it does not exceed the coherence time limit of the NV color center.
[0068] In some embodiments, the error correction operation of the NV center Ising machine does not interrupt the iteration; only the time for state mapping and multiple readouts is added during the readout phase, with the total error correction operation time being approximately 50 μs. The total time for the NV center Ising machine in a single iteration process—mapping the electronic spin state to multiple nuclear spins for redundant encoding, obtaining readout results of multiple nuclear spins, determining the spin state based on the majority voting rule, and dynamically adjusting the control pulse parameters—is less than the coherence time of the NV center Ising machine.
[0069] Combination Figure 4 The timing diagram shown quantitatively analyzes the total time consumption of error correction operations in the NV center Ising machine. Under typical NV center experimental parameters (e.g., operating temperature 4K), the state mapping process in step S101 involves R CNOT gate operations. The independent readout process in step S102 involves R frequency-selective pulse readouts, taking approximately 15 μs. The majority voting decision in step S103 is executed by the FPGA hardware, taking a very short time of approximately 0.1 μs. Step S104 involves feedback calibration calculation and parameter update, with feedback calculation taking approximately 5 μs and feedback manipulation taking approximately 5 μs. The Ising evolution time is 10 μs. Therefore, the total incremental time for error correction operations is approximately 47 μs. In comparison, the electron spin coherence time of the NV color center is... Typically, the time taken is on the order of milliseconds (e.g., 4ms). Clearly, the total time for error correction is much less than the coherence time. This ensures that error correction operations can be fully embedded into the iterative evolution of the Ising machine without quantum state decoherence due to excessive operation time, thus ensuring that the error correction process itself does not disrupt computational continuity and achieving "online error correction." The effective fidelity of the spin state determined by the majority voting rule is higher than that of the readout fidelity of a single nuclear spin, significantly improving the effective fidelity of the readout results.
[0070] In some embodiments, the effective fidelity of the spin state determined by the majority voting rule is:
[0071] in: Indicates effective fidelity; Indicates the number of nuclear spins; Indicates the first The fidelity of the results read out each time.
[0072] In some embodiments, a 50-NV center Ising machine (comprising 50 NV centers) operates on natural diamond (¹³C abundance 1.1%, operating temperature 4 K). The specific configuration for duplicate code correction is as follows: Single-shot ODMR (Optical Detector Magnetic Resonance) readout fidelity F=90% is increased to... The overall fidelity of 50 spins is Compared to the fidelity in the case of no error correction This represents a 48-fold improvement. In terms of timing, the state mapping time in the error correction timing is... Independent readout time is The voting and feedback period is as follows: The total error correction time increment is 31μs, which is much smaller than the 4ms coherence time of the Ising machine, verifying the feasibility and lightweight advantages of the scheme in this embodiment in a large-scale system.
[0073] like Figure 2 The figure shown is a schematic diagram illustrating the overall fidelity of the error correction method in this embodiment as a function of the number of spins. Figure 2 The horizontal axis represents the number of spins, N. The vertical axis represents the overall correct probability. ), which is fidelity, and its corresponding usability threshold is 0.01. Figure 2 In the diagram, the red dashed line illustrates the fidelity change without error correction. Assuming a single readout fidelity F=95%, the overall fidelity decreases exponentially with the increase of the number of spins N. When N=50, the overall fidelity approaches 0, and the system can hardly obtain correct results. The blue solid line shows the fidelity change after using the R=3 repetition code in this embodiment for error correction. The downward trend of the curve slows significantly, and the overall fidelity can still be maintained at a high level when N=50, which is about 48 times higher than the case without error correction. The green solid line shows the fidelity change after using the R=5 repetition code for error correction. The curve remains almost flat, and when N=50, the overall fidelity is close to 100%, which is about 6000 times higher than the case without error correction. This data strongly demonstrates that the present invention, through a lightweight repetition code error correction strategy, successfully overcomes the error accumulation bottleneck in the large-scale application of the Ising machine. This nonlinear fidelity improvement stems from the statistical filtering characteristics of the majority voting mechanism for random errors, enabling the system to achieve a qualitative leap in performance with extremely low resource overhead (utilizing only natural nuclear spins). The nuclear spins around the NV color center can be precisely manipulated and read out through ultrafine coupling, serving as a natural error correction aid resource. This allows the correction of the NV color center Ising machine in this embodiment to achieve zero additional hardware cost, significantly reducing the resource overhead of quantum error correction.
[0074] The error correction method for the Ising machine with NV color centers provided by the embodiments of the present invention utilizes the 2–10 addressable nuclear spins naturally present in the NV color centers, with zero additional hardware cost and low error correction cost; R=3 repetitions can improve the fidelity from 90% to 97.2%, and R=5 can improve it to 99.1%; the measurement feedback iteration of the Ising machine is not interrupted during the error correction process, and the total time of the error correction operation (state mapping + multiple readouts + voting + calibration) is <50 μs, which is much smaller than the coherence time T2≈4 ms of the Ising machine; real-time feedback pulse calibration can suppress the long-term accumulation of systematic control errors.
[0075] like Figure 6 As shown, an embodiment of the present invention provides an error correction device for an NV color center Ising machine, comprising: The state mapping module 601 is configured to map the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding; The readout module 602 is configured to acquire readout results of multiple nuclear spins; The voting determination module 603 is configured to determine the corrected spin state based on multiple readout results and through majority voting rules. The feedback calibration module 604 is configured to dynamically adjust the control pulse parameters based on the statistical deviation during the error correction process.
[0076] Embodiments of the present invention also provide a non-transitory machine-readable medium storing a computer program, wherein the computer program, when executed by a computer's processor, is used to cause the computer to perform a method according to an embodiment of the present invention.
[0077] Embodiments of the present invention also provide a computer program product, including a computer program, wherein the computer program, when executed by a computer's processor, is used to cause the computer to perform the method of an embodiment of the present invention.
[0078] An embodiment of the present invention also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores a computer program executable by the at least one processor, which, when executed by the at least one processor, causes the electronic device to perform the method of the embodiment of the present invention.
[0079] refer to Figure 7The present invention will now describe a structural block diagram of an electronic device that can serve as an embodiment of the present invention, serving as an example of a hardware device applicable to various aspects of the present invention. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.
[0080] like Figure 7 As shown, the electronic device includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 702 or a computer program loaded into a random access memory (RAM) 703 from a storage unit 708. The RAM 703 may also store various programs and data required for the operation of the electronic device. The computing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.
[0081] Multiple components in the electronic device are connected to I / O interface 705, including: input unit 706, output unit 707, storage unit 708, and communication unit 709. Input unit 706 can be any type of device capable of inputting information into the electronic device. Input unit 706 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of the electronic device. Output unit 707 can be any type of device capable of presenting information and may include, but is not limited to, a display, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 708 may include, but is not limited to, disks and optical discs. Communication unit 709 allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, and / or wireless communication transceivers, such as Bluetooth devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.
[0082] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, CPUs, graphics processing units (GPUs), various special-purpose artificial intelligence (AI) computing units, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods and processes described above. For example, in some embodiments, the method embodiments of the present invention can be implemented as computer programs tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed on an electronic device via ROM 702 and / or communication unit 709. In some embodiments, the computing unit 701 can be configured to perform the methods described above by any other suitable means (e.g., by means of firmware).
[0083] Computer programs for implementing the methods of embodiments of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0084] In the context of embodiments of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable signal medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, or infrared systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact optical disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.
[0085] It should be noted that the term "comprising" and its variations used in the embodiments of this invention are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The modifications of "one" and "a plurality" mentioned in the embodiments of this invention are illustrative and not restrictive, and those skilled in the art should understand that unless explicitly indicated otherwise in the context, they should be understood as "one or more".
[0086] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of this invention are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.
[0087] The steps described in the method embodiments provided by the present invention can be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of protection of the present invention is not limited in this respect.
[0088] The term "embodiment" in this specification refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily imply the same embodiment, nor does it imply independence or alternativeity from other embodiments. The various embodiments in this specification are described in a related manner, with reference to each other for similar or identical parts. In particular, for apparatus, device, and system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, and relevant details are referred to in the description of the method embodiments.
[0089] The above-described embodiments are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.
Claims
1. An error correction method for an NV color center Ising machine, characterized in that, include: The electronic spin states of the NV color center are mapped to multiple nuclear spins for redundant encoding; Obtain the readout results of the multiple nuclear spins; Based on multiple readout results, the corrected spin state is determined by a majority voting rule; The control pulse parameters are dynamically adjusted based on the statistical deviations observed during the error correction process.
2. The error correction method for the NV color center Ising machine according to claim 1, characterized in that, The process of mapping the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding includes: By using a controlled NOT gate operation between the electron spin of the NV color center and multiple nuclear spins around the NV color center, the electron spin state corresponding to the electron spin is sequentially mapped to the multiple nuclear spins for redundant encoding.
3. The error correction method for the NV color center Ising machine according to claim 2, characterized in that, The controlled NOT gate operation is achieved based on the hyperfine coupling between the electron spin of the NV color center and the plurality of nuclear spins, wherein the Hamiltonian of the hyperfine coupling is: in: For the first The hyperfine coupling constant of the nth nuclear spin is used to represent the nth nuclear spin. The strength of the interaction between nuclear spin and electron spin; For electron spin Quantity; For the first Each nuclear spin Quantity.
4. The error correction method for the NV color center Ising machine according to claim 1, characterized in that, The process of obtaining the readout results of the plurality of nuclear spins includes: Based on the difference in hyperfine coupling constants between different nuclear spins, the state of each nuclear spin can be read out independently using frequency-selective pulses.
5. The error correction method for the NV color center Ising machine according to claim 1, characterized in that, The step of dynamically adjusting the control pulse parameters based on the statistical deviation during the error correction process includes: Select one of the plurality of nuclear spins as a reference nuclear spin; Monitor the statistical deviation between the majority voting results and the readout results of the reference nuclear spin in multiple consecutive iterations; When the statistical deviation exceeds a preset threshold, the control pulse parameters are adjusted.
6. The error correction method for the NV color center Ising machine according to claim 5, characterized in that, The statistical deviation between the majority voting results and the readout results of the reference nuclear spin in the monitoring of multiple consecutive iterations includes: A fixed-width sliding window is used to statistically analyze the average deviation between the majority voting results and the readout results of the reference nuclear spin in multiple consecutive iterations.
7. The error correction method for the NV color center Ising machine according to claim 5, characterized in that, The control pulse parameters include the control pulse angle, and fine-tuning the control pulse angle includes: in: Indicates the rotation angle of the microwave pulse; Indicates the direction and magnitude of the detected systematic drift; This indicates the step size for each correction.
8. The error correction method for the NV color center Ising machine according to claim 1, characterized in that, The number of nuclear spins is 3 or 5.
9. The error correction method for the NV color center Ising machine according to claim 1, characterized in that, The effective fidelity of the spin state determined by the majority voting rule is: in: Indicates effective fidelity; Indicates the number of nuclear spins; Represents the first nuclear spin The fidelity of the readout results.
10. An error correction device for an NV color center Ising machine, characterized in that, include: The state mapping module is configured to map the electronic spin states of the NV color center to multiple nuclear spins for redundant encoding; The readout module is configured to acquire the readout results of the multiple nuclear spins; The voting determination module is configured to determine the corrected spin state based on multiple readout results using a majority voting rule. The feedback calibration module is configured to dynamically adjust the control pulse parameters based on the statistical deviation during the error correction process.
11. An electronic device, comprising: A processor and a memory storing a program, characterized in that the program includes instructions that, when executed by the processor, cause the processor to perform the method according to any one of claims 1 to 9.
12. A non-transitory machine-readable medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1 to 9.
