A quantum control waveform overshoot correction method, system, storage medium and electronic device based on frequency response measurement

Abstract

The present invention discloses a quantum control waveform overshoot correction method, system, storage medium and electronic device based on frequency response measurement, including: setting pulse parameters and output mode, loading pulse signal and outputting and playing, collecting output waveform and measuring system pulse response; obtaining the system's frequency response function, frequency domain compensation frequency response function and time domain waveform function through data transformation processing; using two convolution processes to obtain the final correction data; loading the correction data to correct the waveform, and cooperating with a programmable gate array for parallel calculation. The method uses two convolution processes to convert edge data into calibration data and then use the calibration data to correct the waveform, and cooperates with a programmable gate array for parallel calculation, so that the corrected waveform is smoother, avoiding the problem of missing edges or over-correction in manual calibration, and effectively improving the calculation efficiency and calibration accuracy.

Description

Technical Field

[0001] The present invention relates to the field of quantum computer technology, and in particular to a quantum control waveform overshoot correction method and system based on frequency response measurement. Background Art

[0002] The control of the quantum bits of a superconducting quantum computer needs to be carried out synchronously through two physical ports, the XY control line and the Z control line: microwave pulse signals are input through the XY line, and DC pulse signals are input through the Z line for bit control and testing. Quantum computing requires that the DC pulse signal in the Z line not only has very fast edge rise and edge fall times to ensure that more bit gates can be configured within a certain decoherence time, but also requires that the pulse edge has extremely low overshoot to ensure the stability of the quantum bit resonant cavity frequency.

[0003] Therefore, it is necessary to generate a fast and highly flat DC pulse signal to ensure the quality of quantum bit control and testing. The existing correction method is to perform pre-distortion processing in the time domain, that is, by adjusting the value of the edge data point and observing the overshoot change of the adjusted signal, a DC pulse with a smaller overshoot is obtained after multiple adjustments, and then the calibration data is calculated based on the obtained pulse rising edge data to perform digital pre-distortion processing on the edge of the pulse waveform that needs to be played later. Since the overshoot size after pre-distortion processing depends on the calibration data, and the calibration data is manually adjusted according to the size of the waveform displayed by the oscilloscope, when adjusting multiple parameters, multiple parameter adjustments need to be made. Multiple parameters have a linkage effect when adjusted, so calibration needs to be repeated multiple times, and the optimization of the calibration results cannot be guaranteed. The calculation efficiency is low and the amount of calculation is large. When some pulses cannot be identified or are not pulses but are identified as pulses, there are problems of missed correction or over-correction. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a quantum control waveform overshoot correction method and system based on frequency response measurement.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a quantum control waveform overshoot correction method based on frequency response measurement, comprising:

[0006] S1, setting the pulse parameters and output mode of the quantum control signal generation system, loading the pulse signal and outputting it for playback, collecting the output waveform to obtain the data of the pulse response of the quantum control signal generation system;

[0007] S2, performing multi-step data transformation processing on the data points of the impulse response of the system to obtain the time domain waveform function of the system;

[0008] S3, setting a pulse waveform function, performing a first convolution process on the time domain waveform function and the pulse waveform function, converting them into calibration data, performing a second convolution process on the pulse waveform function and the calibration data, and obtaining a corrected data;

[0009] S4, loading correction data to correct the waveform, and cooperating with the programmable gate array to perform parallel calculation.

[0010] As a further description of the above technical solution, in step S2, it includes:

[0011] S21, perform Fourier transform processing on the pulse response data of the quantum control signal generation system to calculate the frequency response function of the system ;

[0012] S22, frequency response function of the system Take the inverse and calculate the frequency response function of frequency domain compensation ;

[0013] S23, frequency response function for frequency domain compensation Perform inverse Fourier transform processing to calculate the system's time domain waveform function .

[0014] As a further description of the above technical solution, the frequency response function of the system is , the algorithm formula is as follows:

[0015]

[0016] in, is the frequency response function The discrete form of is the equally spaced discrete sample data of the collected input signal, N is the total number of data samples, k and n are positive integers from 0 to N-1, is the frequency response angular frequency parameter, are equally spaced discrete values ​​of the angular frequency parameter, is a natural constant, Is an imaginary unit.

[0017] As a further description of the above technical solution, the frequency response function of the frequency domain compensation is , the algorithm formula is as follows:

[0018]

[0019] Where N is the total number of data samples, n is a positive integer from 0 to N-1, are equally spaced discrete values ​​of the angular frequency parameter, Is an imaginary unit.

[0020] As a further description of the above technical solution, the time domain waveform function of the system , the algorithm formula is as follows:

[0021]

[0022] in, are the discrete data points of the time domain waveform obtained after conversion, represents a time series, represents the frequency sequence, is the frequency response function of the complementary filter frequency domain compensation, N is the total number of data samples, and k and n are positive integers from 0 to N-1.

[0023] As a further description of the above technical solution, in step S3, it includes:

[0024] S31, pulse waveform function and time domain waveform function Perform the first convolution process and calculate the rising edge function ;

[0025] S32, take the rising edge function Edge data points are converted into calibration data ;

[0026] S33, pulse waveform function for output and calibration data Perform the second convolution process to calculate the final corrected data .

[0027] As a further description of the above technical solution, the rising edge function , the algorithm formula is as follows:

[0028]

[0029] in, for Time waveform amplitude, pulse waveform function middle is 0, is 1, , N is the total number of data samples, k and n are positive integers from 0 to N-1.

[0030] As a further description of the above technical solution, the calibration data , the algorithm formula is as follows:

[0031]

[0032]

[0033]

[0034]

[0035] Among them, the edge data is … There are M data in total, and , , , , , Indicates the edge data The range of the data.

[0036] As a further description of the above technical solution, the correction data , the algorithm formula is as follows:

[0037]

[0038] in, is the pulse waveform function to be output, is the calibration data, N is the total number of data samples, and k and n are positive integers from 0 to N-1.

[0039] It also includes a quantum control waveform overshoot correction system based on frequency response measurement, and the correction system is used to perform the correction method as described in any one of the above technical solutions, including:

[0040] The output module generates quantum control waveforms and plays and outputs them after loading extremely narrow pulses;

[0041] The measurement and display module collects the pulse waveform and the pulse response of the measurement system, saves them as data points and sends them to the calculation and processing module to display the pulse waveform before and after correction;

[0042] A processing and calculation module obtains a system frequency response function by calculation, and obtains correction data by convolution processing of the frequency response function;

[0043] The transceiver control module receives the set waveform and sampling parameters, sends them to the generating output module, receives the correction data from the processing and calculation module, and cooperates with the programmable gate array to perform parallel calculation and correct the waveform.

[0044] The above technical solution has the following advantages or beneficial effects:

[0045] The present invention collects the output waveform of the quantum control signal generating system and measures the pulse response of the quantum control signal generating system, obtains the frequency response function, the frequency response function of the frequency domain compensation and the time domain waveform function through data transformation processing, and then uses two convolution processes to convert the edge data into calibration data, and then uses the calibration data to correct the waveform, and cooperates with the programmable gate array to perform parallel calculation, so that the corrected waveform is smoother, avoids the problem of missing edges or over-correction in manual calibration, and effectively improves the calculation efficiency and calibration accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Calibration process for existing correction methods Figure 1 ;

[0047] Figure 2 Calibration process for existing correction methods Figure 2 ;

[0048] Figure 3 The pulse edge diagram after calibration by the existing correction method;

[0049] Figure 4 The process of the correction method proposed by the present invention Figure 1 ;

[0050] Figure 5 The process of the correction method proposed by the present invention Figure 2 ;

[0051] Figure 6 A schematic diagram of parallel convolution calculation in a programmable gate array in the correction method proposed by the present invention;

[0052] Figure 7 A schematic diagram of the waveform amplitude in the correction method proposed by the present invention;

[0053] Figure 8 It is a structural schematic diagram of the correction system proposed by the present invention;

[0054] Fig. 9 It is a comparative schematic diagram before and after overshoot correction in the correction method proposed by the present invention.

[0055] Legend:

[0056] 1. Output module; 2. Measurement and display module; 3. Processing and calculation module; 4. Transceiver control module. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The existing correction method is to suppress and correct overshoot by adjusting the amplitude of the pulse edge waveform at each point. The specific technical solution process is as follows: Figure 1 As shown: First, an uncorrected pulse signal is output through the waveform generator. At this time, the output pulse signal has a large overshoot. The specific value of the overshoot is measured by a broadband oscilloscope, and then the calibration data is modified to make the pulse edge smoother. The calibration data is then loaded into the waveform generator; according to the overshoot change trend, the amplitudes of multiple calibration data are traversed from large to small, and the best calibration data can be finally obtained after multiple corrections.

[0059] The specific calibration process is as follows: Figure 2 As shown: Assume that there are 6 points that can be adjusted on the edge of the pulse waveform, namely [A0, A1, A2, A3, A4, A5]. In the absence of correction, A0 is equal to 0, and A1, A2, A3, A4, and A5 are equal to 1; first reduce A5, and continue to reduce A5 after observing that the overshoot becomes smaller on the oscilloscope, so that the overshoot becomes larger. The value obtained at the lowest point of the overshoot at this time is the undetermined value of A5; reduce A4 so that A4 < A5, observe the overshoot changes displayed on the oscilloscope during the adjustment process, and record the value at the lowest point of the overshoot. Similarly, calibrate A3, A2, A1, and A0 respectively according to the above method. However, the adjustment of each calibration point will affect the overall overshoot, and the data suppression effect of other points will also be affected when each point is adjusted separately. Therefore, the above process needs to be repeated 3-5 times to adjust the correction effect of each point as a whole.

[0060] like Figure 3 As shown, the final rising edge function is: ;

[0061] Among them, y is the waveform data amplitude, t is time, and b is the rising edge control coefficient, which is used to control the speed of the rising edge. Since the frequency response of each waveform output device is different, the overshoot size obtained by strictly generating the pulse edge according to the curve of the formula will also be different, and the rising edge is slower. Therefore, when manually calibrating the edge according to the rising trend of the waveform, the calibration efficiency is low, and the calibration accuracy is difficult to guarantee. The waveform overshoot correction result is not the best result.

[0062] The specific steps of applying the calibration data to the pulse signal include: first, calculating the amplitude difference between adjacent points of the data point by point, and when the amplitude difference is greater than or equal to a certain threshold, it is considered that a pulse edge appears at the value position, and then the amplitude of the edge position is replaced with the previous calibration data. This process requires point-by-point calculations, which has low calculation efficiency and a large amount of calculations. When the set threshold is too large, edges with lower pulse amplitude differences may be skipped, and the correction result may be missed. When the set threshold is too small, normal signals with lower pulse amplitude differences may be identified as edges, resulting in noise over-correction.

[0063] The embodiments of the present application provide a method and system for correcting overshoot of a quantum-controlled waveform based on frequency response measurement, which solves the problem that the calibration effect is affected by human factors and the optimization of the calibration result cannot be guaranteed. At the same time, the process of applying the calibration data to the pulse data for pre-distortion after calibration has low computational efficiency and large amount of calculation, and some pulses cannot be recognized or are recognized as pulses instead of pulses, thereby performing over-correction.

[0064] Reference Figure 4 , an embodiment provided by the present invention: a quantum control waveform overshoot correction method based on frequency response measurement, comprising:

[0065] S1, setting the pulse parameters and output mode of the quantum control signal generation system, loading the pulse signal and outputting it for playback, collecting the output waveform to obtain the data of the pulse response of the quantum control signal generation system;

[0066] S2, performing multi-step data transformation processing on the data points of the impulse response of the system to obtain the time domain waveform function of the system;

[0067] S3, setting a pulse waveform function, performing a first convolution process on the time domain waveform function and the pulse waveform function, converting them into calibration data, performing a second convolution process on the pulse waveform function and the calibration data, and obtaining a corrected data;

[0068] S4, loading correction data to correct the waveform, and cooperating with the programmable gate array to perform parallel calculation.

[0069] Reference Figure 5 , in step S1, including:

[0070] S11, use a quantum control waveform generation device (such as a waveform generator) to load an extremely narrow pulse with only one sampling point as 1 and other points as 0, and then play and output it after passing through channel conditioning, filtering, amplification and other circuits, and the output mode is equal interval trigger playback;

[0071] S12, use a measuring device (such as a broadband oscilloscope) to measure the output waveform, adjust the trigger and sampling rate of the oscilloscope so that it can completely collect a pulse waveform, which is the pulse response of the quantum control signal generation system, and save the waveform as a data point ;

[0072] Reference Figure 5 , in step S2, including:

[0073] S21, perform Fourier transform processing on the pulse response data of the quantum control signal generation system to calculate the frequency response function of the system ;

[0074] S22, frequency response function of the system Take the inverse and calculate the frequency response function of frequency domain compensation ;

[0075] S23, frequency response function for frequency domain compensation Perform inverse Fourier transform processing to calculate the system's time domain waveform function .

[0076] In this embodiment, the frequency response function of the system is The discrete form of , the algorithm formula is as follows:

[0077]

[0078] in, is the equally spaced discrete sample data of the collected input signal, N is the total number of data samples, k and n are positive integers from 0 to N-1, is the frequency response angular frequency parameter, are equally spaced discrete values ​​of the angular frequency parameter, is a natural constant, Is an imaginary unit.

[0079] In this embodiment, the frequency response function of frequency domain compensation is , the algorithm formula is as follows:

[0080]

[0081] Where N is the total number of data samples, n is the index of the waveform sequence, and n is a positive integer from 0 to N-1. are equally spaced discrete values ​​of the angular frequency parameter, is an imaginary unit;

[0082] It should be noted that in order to ensure that the system can effectively compensate and process the frequency response, it is necessary to ensure that the frequency range of the input signal does not exceed the maximum bandwidth of the system to avoid adverse effects on the frequency response compensation effect caused by the frequency exceeding the range. , this range corresponds to the maximum bandwidth of the frequency response of the control signal generation system. After exceeding this bandwidth, the frequency response drops rapidly, resulting in a too large value after taking the inverse, thus affecting the supplementation effect.

[0083] In this embodiment, the system's time domain waveform function , the algorithm formula is as follows:

[0084]

[0085] in, are the discrete data points of the time domain waveform obtained after conversion, represents a time series, represents the frequency sequence, is the frequency response function of the complementary filter frequency domain compensation, N is the total number of data samples, and k and n are positive integers from 0 to N-1.

[0086] Reference Figure 5 , in step S3, including:

[0087] S31, pulse waveform function and time domain waveform function Perform the first convolution process and calculate the rising edge function ;

[0088] S32, take the rising edge function Edge data points are converted into calibration data ;

[0089] S33, pulse waveform function for output and calibration data Perform the second convolution process to calculate the final corrected data .

[0090] In this embodiment, the rising edge function , the algorithm formula is as follows:

[0091]

[0092] in, for Time waveform amplitude, pulse waveform function middle is 0, is 1, , N is the total number of data samples, k and n are positive integers from 0 to N-1.

[0093] In this embodiment, The edge data points in the image are converted into convolution data according to the following formula, which is the final calibration data. , the algorithm formula is as follows:

[0094]

[0095]

[0096]

[0097]

[0098] refer to Figure 7 , where the edge data is … There are M data in total, and , , ≠0, , , Indicates the edge data The amplitude of the data. Figure 7 As shown, assume that the data The edge of At the beginning, there are 5 data on the edge (M=5), among which, , , , , .

[0099] In this embodiment, when the pulse waveform is corrected using the calibration data, the pulse waveform can be directly convolved with the calibration data to correct the data. , the algorithm formula is as follows:

[0100]

[0101] in, is the pulse waveform function to be output, is the calibration data, N is the total number of data samples, and k and n are positive integers from 0 to N-1.

[0102] In this embodiment, through the above-mentioned convolution calculation, there is no need to determine the position of the rising edge or the falling edge during calibration and correction, so there is no problem of missed edges or over-correction, and the use of convolution can make good use of the parallel computing advantages in the programmable gate array to achieve the purpose of real-time processing, so that the corrected waveform is smoother, effectively avoiding the problem of missed edges or over-correction caused by manual calibration.

[0103] Reference Figure 6 Driven by the logic clock, the pulse waveform data points keep moving and are multiplied with the convolution kernel at the same time, so the convolution result can be output quickly. No additional processing time will be added due to data calculation. The calculation efficiency is improved while real-time processing is realized, which effectively improves the calculation efficiency and calibration accuracy.

[0104] Reference Figure 8 The present invention also includes a quantum control waveform overshoot correction system based on frequency response measurement, and the correction system is used to execute the correction method of any one of the above technical solutions, including:

[0105] Output module 1 is used to generate quantum control waveforms and play and output after loading extremely narrow pulses;

[0106] The measurement and display module 2 collects the pulse waveform and the pulse response of the measuring system, saves it as data points and sends it to the calculation and processing module to display the pulse waveform before and after correction;

[0107] The processing and calculation module 3 obtains the system frequency response function by calculation, and obtains the correction data by convolution processing of the frequency response function;

[0108] The transceiver control module 4 receives the set waveform and sampling parameters, sends them to the generation output module, receives the correction data of the processing calculation module, and cooperates with the programmable gate array to perform parallel calculation and waveform correction.

[0109] In this embodiment, the operating program of the correction method described above is executed by the correction system: the generation output module 1 generates a quantum control waveform, which is played and output after loading an extremely narrow pulse; these waveforms have specific spectral characteristics and modulation methods, and the pulse waveform and the pulse response of the measurement system are collected by the measurement display module 2, and are saved as data points and then sent to the calculation processing module; after receiving the measurement data, the processing and calculation module 3 performs data processing and calculation to obtain the frequency response function of the system, and then performs two convolution processes on the frequency response function to obtain the correction data; the transceiver control module 4 is responsible for receiving the set waveform and sampling parameters, sending them to the generation output module 1 to generate the waveform, and after receiving the correction data, it cooperates with the programmable gate array to perform parallel calculations to correct the waveform. The waveforms before and after the correction will be displayed in real time through the measurement display module 2 to ensure that the correction effect meets expectations. Fig. 9As shown, it can be seen that the final corrected waveform is smoother, without overshoot, and the correction effect is better.

[0110] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A quantum control waveform overshoot correction method based on frequency response measurement, characterized in that: include: S1, setting the pulse parameters and output mode of the quantum control signal generation system, loading the pulse signal and outputting it for playback, collecting the output waveform to obtain the data of the pulse response of the quantum control signal generation system; S2, performing data transformation processing on the data points of the impulse response of the system to obtain the time domain waveform function of the system; S3, setting a pulse waveform function, performing a first convolution process on the time domain waveform function and the pulse waveform function, converting them into calibration data, performing a second convolution process on the pulse waveform function and the calibration data, and obtaining a corrected data; S4, loading correction data to correct the waveform, and cooperating with the programmable gate array to perform parallel calculation; In the step S3, it includes: S31, pulse waveform function and time domain waveform function Perform the first convolution process and calculate the rising edge function ; S32, take the rising edge function Edge data points are converted into calibration data ; S33, pulse waveform function for output and calibration data Perform the second convolution process to calculate the final corrected data .

2. The correction method according to claim 1, characterized in that: In the step S2, it includes: S21, perform Fourier transform processing on the pulse response data of the quantum control signal generation system to calculate the frequency response function of the system ; S22, frequency response function of the system Take the inverse and calculate the frequency response function of frequency domain compensation ; S23, frequency response function for frequency domain compensation Perform inverse Fourier transform processing to calculate the system's time domain waveform function .

3. The correction method according to claim 2, characterized in that: The frequency response function of the system , the algorithm formula is as follows: in, is the frequency response function The discrete form of is the equally spaced discrete sample data of the collected input signal, N is the total number of data samples, and n is a positive integer from 0 to N-1, is the frequency response angular frequency parameter, are equally spaced discrete values ​​of the angular frequency parameter, is a natural constant, Is an imaginary unit.

4. The correction method according to claim 2, characterized in that: The frequency response function of the frequency domain compensation , the algorithm formula is as follows: Where N is the total number of data samples, n is a positive integer from 0 to N-1, are equally spaced discrete values ​​of the angular frequency parameter, Is an imaginary unit.

5. The correction method according to claim 2, characterized in that: The time domain waveform function of the system , the algorithm formula is as follows: in, are the discrete data points of the time domain waveform obtained after conversion, represents a time series, represents the frequency sequence, is the frequency response function of the complementary filter frequency domain compensation, N is the total number of data samples, and k and n are positive integers from 0 to N-1.

6. The correction method according to claim 1, characterized in that: The rising edge function , the algorithm formula is as follows: in, for Time waveform amplitude, pulse waveform function middle is 0, is 1, , N is the total number of data samples, k and n are positive integers from 0 to N-1.

7. The correction method according to claim 1, characterized in that: The calibration data , the algorithm formula is as follows: Among them, the edge data is … There are M data in total, and , , , , , Indicates the edge data The range of the data.

8. The correction method according to claim 1, characterized in that: The correction data , the algorithm formula is as follows: in, is the pulse waveform function to be output, is the calibration data, N is the total number of data samples, and k and n are positive integers from 0 to N-1.

9. A quantum control waveform overshoot correction system based on frequency response measurement, characterized in that: The correction system is used to perform the correction method according to any one of claims 1 to 8, comprising: The output module (1) generates a quantum control waveform and plays and outputs it after loading the pulse; The measurement and display module (2) collects the pulse waveform and the pulse response of the measurement system, saves them as data points and sends them to the calculation and processing module, and displays the pulse waveform before and after correction; A processing and calculation module (3) obtains a system frequency response function by calculation, and obtains correction data by convolution processing of the frequency response function; The transceiver control module (4) receives the set waveform and sampling parameters, sends them to the generating output module, receives the correction data from the processing and calculation module, and cooperates with the programmable gate array to perform parallel calculation and correct the waveform.

Images