Information processing device, waveform data generation method, and program

Abstract

To easily describe the desired pulse sequence. [Solution] An information processing device for generating waveform data used to generate microwave pulses to irradiate a qubit, comprising a conversion unit that takes a string representing the waveform of the microwave pulse as input and converts the string into waveform data.

Description

Technical Field

[0001] The present invention relates to the technical field of quantum computers.

Background Art

[0002] A quantum computer is a technology that performs calculations by utilizing the principle of superposition in quantum mechanics. If a sufficiently large-scale quantum computer is constructed, it is expected to show much higher performance than currently widely used computers (classical computers) for basic calculation tasks related to material analysis and discovery of periodicity, which quantum computers are good at. Therefore, the development of practical-scale quantum computers is actively underway worldwide.

[0003] As physical systems for realizing qubits, there are superconducting circuits, ions, atoms, color centers, silicon dots, optical quanta, floating electrons, etc. Among these physical systems, superconducting qubits realized by superconducting circuits have been successfully scaled up to hundreds of qubits, and are thus regarded as one of the standard qubits.

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Controlling superconducting qubits requires shaping microwave pulses and irradiating them in the appropriate form. This necessitates a unified method for describing the pulse shape. A pulse sequence is constructed by adding multiple waveforms, such as Gaussian waveforms or square waves with rounded edges, and modulating them at a specific frequency. Furthermore, when simultaneously irradiating multiple qubits with microwave pulses, it is necessary to synchronize the pulse positions in an appropriate relative position.

[0006] However, conventional technology had the problem of not being able to easily describe the desired pulse sequence.

[0007] This invention has been made in view of the above points, and aims to provide a technology that enables the simple description of a desired pulse sequence. [Means for solving the problem]

[0008] According to the disclosed technology, an information processing device that generates waveform data used to generate microwave pulses to irradiate a qubit, A conversion unit that takes a string representing the waveform of the microwave pulse as input and converts the string into waveform data. An information processing device equipped with [this feature] is provided. [Effects of the Invention]

[0009] The disclosed technology provides a technique that enables the convenient description of a desired pulse sequence. [Brief explanation of the drawing]

[0010] [Figure 1] This figure shows an example of a system configuration in an embodiment of the present invention. [Figure 2] This figure shows an example of waveform data. [Figure 3] The figure shows an example of string data. [Figure 4] This is a flowchart illustrating the operation of the conversion unit 110. [Figure 5] This is a flowchart illustrating the operation of the conversion unit 110. [Figure 6] This is a flowchart illustrating the operation of the conversion unit 110. [Figure 7] This figure shows an example of the hardware configuration of the information processing device 100. [Modes for carrying out the invention]

[0011] Hereinafter, embodiments of the present invention (this embodiment) will be described with reference to the drawings. The embodiments described below are merely examples, and the embodiments to which the present invention is applied are not limited to the embodiments described below.

[0012] In the following, to facilitate understanding of the technology of this embodiment, we will first describe the prior art related to quantum computers that is relevant to this embodiment.

[0013] [About quantum computing] A quantum computer (or quantum computing machine) is a technology that performs calculations by utilizing the superposition principle of quantum mechanics. It is expected that a sufficiently large-scale quantum computer will perform far better than currently used computers (classical computers) in fundamental computational tasks related to material analysis and the discovery of periodicity, areas in which quantum computers excel. Therefore, the development of practical-scale quantum computers has been actively pursued worldwide.

[0014] Classical bits, which are the elements that make up classical computers, can take values ​​of 0 or 1. On the other hand, quantum bits, which are the elements that make up quantum computers, can take on not only 0 and 1, but also a continuous superposition state of 0 and 1. By effectively utilizing a property called coherence that occurs in this superposition state, it is possible to suppress the probability of obtaining an undesirable answer in problems with a periodic structure, and thus enable the high-speed calculations described above.

[0015] When performing calculations using qubits, not only is it necessary to perform operations that transition a qubit from one state to another, but also operations to read the result of the calculation as a normal bit string that can be recognized by humans. This operation of extracting a classical bit string from a quantum bit is called "measurement" or "observation" in technical terms. When a measurement is performed on a qubit that takes a superposition of two states, 0 and 1, its value is probabilistically determined to be 0 or 1 according to the square of the absolute value of the components of the superposition. At this time, as a side effect of the measurement, the state of the qubit changes according to the measured value.

[0016] [Noise and Operations of Quantum Computers] The superposition state of qubits is known to be sensitive to environmental noise and easily break in a naive implementation. A quantum device refers to a device that can maintain a quantum superposition state for a long time and is specifically controllable. In particular, superconducting qubits created using superconducting circuits have been successfully scaled up to hundreds of qubits, and are regarded as one of the standard qubits. A superconducting qubit typically sets the state with the lowest energy (ground energy state) of the created circuit as 0, and the state with the next lowest energy as 1.

[0017] To perform operations using qubits, it is sufficient to implement operations that change the qubits to different states and measurements that extract the information of the qubits as classical bits. The implemented superconducting qubit itself is a passive element, and its operations and measurements are realized by irradiating a microwave pulse with a programmed waveform through a transmission line. A superconducting qubit is typically controlled by irradiating an external microwave pulse with a frequency of several GHz.

[0018] Regarding the operation of qubits, it can be realized by irradiating a microwave pulse with a predetermined envelope that has the resonance frequency of the qubit as the carrier frequency. Regarding measurement, a method called dispersive readout is currently commonly used. This is a method of indirectly examining the state of the qubit from its reflection by externally exciting a resonator having an appropriate resonance frequency coupled to the qubit, rather than directly irradiating the qubit with a signal.

[0019] [Signal generation] In order to programmably control qubits, it is necessary to construct a pulse generation system that has the resonance frequency of the superconducting qubit as the carrier frequency and whose envelope shape can be designed. In a typical waveform generator, it is difficult to directly generate an arbitrary waveform in the several GHz band required by superconducting qubits. Therefore, it is generated by combining a low-frequency arbitrary waveform generator and a signal source that outputs a high-frequency sine wave with a mixer and filtering. When receiving a signal for reading out a qubit, this is realized by following the reverse procedure and converting the received analog signal into a digital signal.

[0020] (Overview of the embodiment) As described above, in order to control a superconducting qubit, it is necessary to shape a microwave pulse and irradiate the qubit with the pulse in an appropriate form. At this time, a method for uniformly describing the shape of the pulse is required. The pulse sequence is composed of adding a plurality of waveforms such as a Gaussian-shaped waveform or a rectangular wave with rounded edges, and modulating at a certain frequency. Also, when irradiating a plurality of qubits with microwave pulses simultaneously, it is necessary to synchronize the positions of the pulses in an appropriate positional relationship.

[0021] However, in the prior art, it was not possible to simply describe a desired pulse sequence.

[0022] Therefore, the technology according to this embodiment makes it possible to specify the shape of a user-defined pulse or to describe pulses generated based on complex parameter arguments using a script consisting only of strings. The configuration and operation of the system according to this embodiment will be described in detail below.

[0023] (Example system configuration) Figure 1 shows an example of the configuration of a quantum computer (which may also be called a quantum computing system, quantum computer, etc.) in this embodiment.

[0024] As shown in Figure 1, the quantum computer in this embodiment includes quantum hardware 200, a pulse irradiation device 310, a measurement device 320, and an information processing device 100. The information processing device 100 also includes a communication function for sending and receiving data.

[0025] In this embodiment, it is assumed that superconducting qubits are used as qubits. The quantum hardware 200 includes an integrated qubit (superconducting circuit) and the entire system inside the dilution refrigerator surrounding it.

[0026] As shown in Figure 1, the information processing device 100 has a conversion unit 110 and a storage unit 120. The conversion unit 110 in the information processing device 100 converts string data into waveform data and transmits the converted waveform data to the pulse irradiation device 310. The storage unit 120 stores, for example, waveform data created in the past. The conversion unit 110 can read the waveform data from the storage unit 120 and use it. The storage unit 120 also stores data necessary for processing the flows shown in Figures 4 to 6, which will be described later.

[0027] The pulse irradiation device 310 generates microwave pulses to irradiate the qubits from waveform data using a sideband modulation method with a frequency mixer. The microwave pulses may be used to manipulate the qubits or to measure them.

[0028] More specifically, the pulse irradiation device 310 includes an arbitrary waveform generator, a frequency mixer, and a plurality of local oscillators (e.g., generating a 6 GHz waveform).

[0029] The arbitrary waveform generator generates an appropriate pulse waveform (a waveform created by combining the I-phase waveform and the Q-phase waveform at a specified carrier frequency) according to the waveform data (including the I-phase waveform data and the Q-phase waveform data).

[0030] The frequency mixer receives a high-frequency waveform from the local oscillator that corresponds to the resonance frequency of the qubit. In the frequency mixer, the pulse generated by the arbitrary waveform generator is modulated to obtain the desired microwave pulse to irradiate the qubit. In the generated waveform with a high-frequency carrier frequency, a high-frequency cosine wave is multiplied into the I-phase waveform, and a high-frequency sine wave is multiplied into the Q-phase waveform, and a microwave output with the combined shape of these is obtained.

[0031] The measuring device 320 shown in Figure 1 is used for measurement (readout of superconducting qubits). The measurement is performed using the distributed readout method. In the distributed readout method, the measuring device 320 determines the state of the qubit by measuring the reflected signal from a resonator irradiated with microwave pulses.

[0032] (Regarding the basic functions of the conversion unit 110) The conversion unit 110 takes string data representing a pulse sequence as input and converts the string data into waveform data.

[0033] String data is data created by concatenating one-dimensional sequences of command characters separated by spaces. For example, "A0.5 T B100 G100 B100" is a sequence of five commands. The conversion unit 110 converts such string data into waveform data.

[0034] More specifically, in string data, a concept called a cursor is used to represent the current time position. The initial position of the cursor is 0. In this state, the conversion unit 110 interprets the commands from left to right. Each command is a predetermined string followed by a comma-separated sequence of real numbers. For example, "A0.5" means "set the amplitude to 0.5". "B100" means to advance the cursor 100ns ahead. "G100" is a command to place a Gaussian waveform with a width of 100ns at the current cursor position. "T" means trigger and is used to synchronize with other sequences (other channels).

[0035] Therefore, the above "A0.5 T B100 G100 B100" is an instruction to "place a Gaussian waveform with a width of 100ns and a height of 0.5 at a position 100ns after the trigger."

[0036] (Example of waveform data) Figure 2 shows an example of a waveform generated from string data by the conversion unit 110. Figure 2 shows the waveforms corresponding to four microwave pulses arranged vertically. The waveforms (or distinctions between them) of the four microwave pulses are called "channels".

[0037] As mentioned above, a single microwave pulse is generated from the waveform of the I phase and the waveform of the Q phase. Therefore, the waveform of the I phase (solid line) and the waveform of the Q phase (dotted line) are shown for each channel. Figure 2 shows four channels (q1r, q1q, q0q, q0r). For example, q1rI means the waveform of the I phase in the q1r channel. The same applies to the others.

[0038] Note that q1r represents the pulse from port r to qubit q1, and q1q represents the pulse from port q to qubit q1. The same applies to q0.

[0039] Furthermore, in Figure 2, the trigger (synchronization signal) mentioned above is shown as a vertical dotted line. In addition, in Figure 2, a measurement window indicating the measurement period (time width) is shown as a solid line. The measurement window represents the period during which the measuring device 320 measures the reflected signal (reflected wave).

[0040] The waveform shown in Figure 2 is, for example, a waveform converted by the conversion unit 110 from the string data shown in Figure 3.

[0041] In Figure 3, seq.q0.r.seq represents a variable that stores the string data of the q0r waveform.

[0042] seq.q0.r.seq="T MEAS TT MEAS" means that the measurement (pulse irradiation and measurement) will be performed with a trigger in between, as shown in q0r in Figure 2. MEAS is a user-defined command, and as shown in Figure 3, it is "A1.0 M F3000 B1000". Here, M indicates the starting point of the measurement window (data acquisition window). The delay and duration from the starting point of the measurement window are also defined as strings. F3000 means that a 3000ns flat-top waveform will be added.

[0043] The command `seq.q0.q.seq="R3 PIV0V, 200 TT PI200,200 T1 T"` shows the waveform of q0q in Figure 2. R3 means that PIV0V is repeated three times. Here, VvV (e.g., V0V) means that it will be replaced by the v-th variable argument, which is defined separately. PI is a user-defined command and is defined as shown in Figure 3. Tv (e.g., T1) means that only sequences (pulses) with Tv will be synchronized at this point.

[0044] seq.q1.q.seq="TT T1 R3 C0CV1V T" shows the waveform of q1q in Figure 2. CvC (e.g., C0C) means that it will be replaced by the v-th variable command, which is defined separately.

[0045] The expression seq.q1.r.seq = seq.q0.r.seq shows the waveform of q1r in Figure 2, indicating that it is the same as the waveform of q0r.

[0046] (Example of operation of the conversion unit 110) An example of the operation of the conversion unit 110 will be explained with reference to the flowcharts in Figures 4 to 6. Before the start of the flow, it is assumed that string data has been input to the conversion unit 110.

[0047] Steps S1 to S7 in Figure 4 are performed for each channel. In step S1 (Step 1) of Figure 4, the conversion unit 110 splits the string using a regular expression and performs tokenization. For example, the conversion unit 110 converts "B100 GAUSS100,200 F100" to [("B",100), ("GAUSS", 100, 200), …].

[0048] In S2, the conversion unit 110 interprets the token generated in S1 and converts the token into command format. For example, the conversion unit 110 converts ("B", 100) to Command("B", 100) (a class object).

[0049] In S3, the conversion unit 110 replaces the sweep argument value variable with a specific value. The nth variable to which the value to be swept by the measurement is assigned is represented as VnV. For example, the conversion unit 110 changes "AV0V" to "A0.5".

[0050] In S4, the conversion unit 110 processes the repeating instruction. For example, the conversion unit 110 expands "R3 COM" to "COM COM COM".

[0051] In S5, the conversion unit 110 expands the sweep command variable. The nth command variable, to which the command name to be swept by the measurement is assigned, is represented as CnC. For example, the conversion unit 110 expands "C0C C1C" to "X90 CNOT".

[0052] In S6, the conversion unit 110 processes user-defined macros (which may also be called user-defined commands). That is, it converts a user-defined string into a predetermined command sequence. For example, the conversion unit 110 expands "X90" into "A0.5 B50 G50 B50".

[0053] In S7, the conversion unit 110 obtains a list of local trigger times from the storage unit 120. Specifically, the conversion unit 110 obtains the time position (channel internal time) for each synchronization trigger instruction in the channel currently being processed.

[0054] For example, in the example in Figure 2, ultimately, the time positions of the triggers for each of the four channels need to be synchronized. As preparation for this, S7 obtains the time (channel internal time) at which each trigger instruction can be used as a trigger for the current target channel.

[0055] For example, if there are no instructions other than a trigger instruction at the beginning of the sequence, the trigger can be placed at the beginning (0ns). Next, if there is a trigger instruction after a 200ns waveform instruction, the next trigger will be placed 200ns later. In this way, the position (channel internal time) of each trigger instruction for each channel is obtained. In other words, for each trigger for each channel, the earliest possible time position is obtained as the channel internal time.

[0056] Next, the conversion unit 110 determines the synchronization point (trigger time position) for all channels. For example, consider the case where there are four channels. If we can place the trigger for trigger 1 (the first trigger) at 10ns on channel 1, 50ns on channel 2, 100ns on channel 3, and 50ns on channel 4, then the time position of trigger 1 will be at 100ns on channel 3, which is the latest of these. In other words, the synchronization point for trigger 1 will be at 100ns. In this case, the other channels are set to 100ns by adding a wait (blank) period.

[0057] If the synchronization point cannot be determined for reasons such as the number of triggers "T" that synchronize the entire system not matching, or the absence of trigger commands and insufficient information to determine the relative positions between channels, the judgment in S8 will be No, and the conversion unit 110 will send an error. If the synchronization points for all channels are successfully determined, the judgment in S8 in Figure 5 will be Yes, and the process will proceed to S9.

[0058] In S9, the transformation unit 110 performs a topological sort of the instructions. If a topological sort is possible under the trigger constraints, the process proceeds to S10; otherwise, the transformation unit 110 sends an error. In the topological sort, the positions of the instructions and triggers are determined so as to minimize the experiment time (measurement length) (i.e., to minimize the longest path) while satisfying the dependency relationships of the instructions and under the trigger constraints.

[0059] In S10, the timing of all triggers is determined. The following steps S11-S13 are performed for each channel.

[0060] In S11, the conversion unit 110 inserts a blank space (wait) in the command sequence of the target channel based on the trigger information to synchronize.

[0061] In S12, the conversion unit 110 calculates the total length (wavelength) of the waveform of the target channel.

[0062] In S13, the conversion unit 110 determines the reception time (i.e., measurement window) of the measurement device for the target channel.

[0063] In S14, the conversion unit 110 determines the measurement length (experimental time length) from the waveform length of the channel with the maximum waveform length. However, if the waveform length is only allowed to be a constant multiple of a certain integer, the measurement length is appropriately increased.

[0064] In S15 of Figure 6, the conversion unit 110 generates a specific waveform signal array from the "processed channel command sequence, the length of the measurement (experiment) to be performed, and the length of time the measurement device is enabled" generated in the processing up to S14, in accordance with the sampling rate of the pulse irradiation device 310. In other words, the conversion unit 110 generates waveform data from the string generated in the processing up to S14.

[0065] Steps S16 and S17 are performed for each channel. Steps S16 and S17 are frequency correction processes performed when the frequency of the local oscillator cannot be precisely specified for the pulse irradiation device 310. The inability to precisely specify the frequency means, for example, that only integer multiples of 50 MHz can be output. Here, we assume that the frequency of the local oscillator cannot be precisely specified. Frequencies that cannot be adjusted by the pulse irradiation device 310 are absorbed by correcting the waveform data being transmitted.

[0066] For example, if we want to generate a 6001MHz microwave pulse for irradiating a qubit using a local oscillator that can output only integer multiples of 50MHz with 6GHz as the reference, this can be achieved by mixing the 1MHz frequency difference between 6GHz and 6001MHz into the waveform data. S16 and S17 show an overview of this waveform data correction process. The frequency correction process itself is a generally known technique.

[0067] In S16, the conversion unit 110 corrects the waveform data by integrating the difference frequency (a 1MHz wave in the above example) into the waveform data, and outputs the corrected waveform and measurement window data for each channel, without considering multiplexing.

[0068] In S17, the conversion unit 110 corrects the difference frequency for the multiplexed channels and adds them together to output waveform and measurement window data for each channel that takes multiplexing into account.

[0069] (Example hardware configuration) The information processing device 100 described in this embodiment can be realized, for example, by having a computer execute a program. This computer may be a physical computer or a virtual machine on the cloud.

[0070] In other words, the information processing device 100 can be realized by using hardware resources such as the CPU and memory built into the computer to execute a program corresponding to the processing performed by the information processing device 100. The above program can be recorded on a computer-readable recording medium (such as portable memory), saved, and distributed. It is also possible to provide the above program via a network such as the Internet or email.

[0071] Figure 7 shows an example of the hardware configuration of the computer described above. The computer in Figure 7 has a drive device 1000, an auxiliary storage device 1002, a memory device 1003, a CPU 1004, an interface device 1005, a display device 1006, an input device 1007, an output device 1008, etc., all of which are interconnected by bus B. The computer may also be equipped with a GPU.

[0072] The program that enables processing on the computer is provided, for example, on a recording medium 1001 such as a CD-ROM or memory card. When the recording medium 1001 containing the program is set in the drive device 1000, the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000. However, the program does not necessarily have to be installed from the recording medium 1001; it may also be downloaded from another computer via a network. The auxiliary storage device 1002 stores the installed program as well as necessary files and data.

[0073] The memory device 1003 reads and stores a program from the auxiliary storage device 1002 when a program startup command is received. The CPU 1004 implements the functions related to the information processing device 100 according to the program stored in the memory device 1003. The interface device 1005 is used as an interface for connecting to a network, etc. The display device 1006 displays a GUI (Graphical User Interface) etc., generated by a program. The input device 1007 consists of a keyboard and mouse, buttons, or a touch panel, etc., and is used to input various operation commands. The output device 1008 outputs the calculation results.

[0074] (Regarding the effects of the technology related to the embodiment) As described above, the technology according to this embodiment makes it possible to easily describe a desired pulse sequence, enabling efficient measurement of microwaves. Furthermore, because the description is written as a string, it is easier to save and reuse compared to pulse generation technologies using general SDKs, and it has excellent readability.

[0075] Furthermore, Non-Patent Document 1 (IBM's OpenPulse) discloses software for generating pulse sequences. However, because this technology enables class-based pulse design, serialization in formats such as scripts is difficult, and it is also difficult for users to define functions with variables. In addition, it is not possible to incorporate advanced functions such as loop statements. The technology according to this embodiment differs from the technology in Non-Patent Document 1 in that it can easily achieve these goals.

[0076] The following additional information is disclosed regarding the embodiments described above.

[0077] <Note> (Additional note 1) An information processing device that generates waveform data used to generate microwave pulses to irradiate a qubit, A conversion unit that takes a string representing the waveform of the microwave pulse as input and converts the string into waveform data. An information processing device equipped with the following features. (Additional note 2) The aforementioned string includes a command indicating the type of waveform and the shape of the waveform, and a trigger indicating the point where multiple channels should be synchronized. The information processing device described in Appendix 1. (Additional note 3) The string further includes the start time position of the measurement of a predetermined time length. The information processing device described in Appendix 2. (Additional note 4) The conversion unit adjusts the timing of commands so that the triggers are at the same timing across multiple channels. The information processing device described in Appendix 2. (Additional note 5) The string includes user-defined commands registered as a sequence of multiple commands, and the conversion unit expands the user-defined commands into the sequence of multiple commands. The information processing device described in Appendix 1. (Additional note 6) The conversion unit corrects the waveform data using the difference between the frequency of the microwave generated by the local oscillator of the pulse irradiation device and the desired frequency, so that a microwave pulse of the desired frequency is generated. The information processing device described in Appendix 1. (Additional note 7) A waveform data generation method performed by an information processing device that generates waveform data used to generate microwave pulses to irradiate a qubit, A step of taking a string representing the waveform of the microwave pulse as input and converting the string into waveform data. A waveform data generation method comprising the following features. (Additional note 8) A non-temporary storage medium storing a program for causing a computer to function as a conversion unit in an information processing device described in any one of the appendices 1 through 6.

[0078] Although this embodiment has been described above, the present invention is not limited to this specific embodiment, and various modifications and changes are possible within the scope of the gist of the invention as described in the claims. [Explanation of Symbols]

[0079] 100 Information Processing Devices 110 Conversion Unit 120 Storage Unit 200 Quantum Hardware 310 Pulse irradiation device 320 Measuring devices 1000 drive unit 1001 Recording media 1002 Auxiliary storage device 1003 Memory device 1004 CPU 1005 Interface device 1006 Display device 1007 Input device 1008 Output device

Claims

1. An information processing device that generates waveform data used to generate microwave pulses to irradiate a qubit, A conversion unit that takes a string representing the waveform of the microwave pulse as input and converts the string into waveform data. An information processing device equipped with the following features.

2. The aforementioned string includes a command indicating the type of waveform and the shape of the waveform, and a trigger indicating the point where multiple channels should be synchronized. The information processing apparatus according to claim 1.

3. The string further includes the start time position of the measurement of a predetermined time length. The information processing apparatus according to claim 2.

4. The conversion unit adjusts the timing of commands so that the triggers are at the same timing across multiple channels. The information processing apparatus according to claim 2.

5. The string includes user-defined commands registered as a sequence of multiple commands, and the conversion unit expands the user-defined commands into the sequence of multiple commands. The information processing apparatus according to claim 1.

6. The conversion unit corrects the waveform data using the difference between the frequency of the microwave generated by the local oscillator of the pulse irradiation device and the desired frequency, so that a microwave pulse of the desired frequency is generated. The information processing apparatus according to claim 1.

7. A waveform data generation method performed by an information processing device that generates waveform data used to generate microwave pulses to irradiate a qubit, A step of taking a string representing the waveform of the microwave pulse as input and converting the string into waveform data. A waveform data generation method comprising the following features.

8. A program for causing a computer to function as a conversion unit in an information processing device according to any one of claims 1 to 6.

Images