Quantum processing unit, quantum computing system and method for reading out the state of a qubit
The quantum processing unit uses swap gates to efficiently read out multiple qubits by transferring states to a central resonator, addressing space and cost challenges while maintaining coherence, thus optimizing quantum computing operations.
Patent Information
- Application Number
- JP2024528554
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2021-11-15
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing quantum computing systems face challenges in reading out the states of multiple qubits without requiring large space in readout-related circuitry, while also managing heat and cost associated with wiring, especially as the number of qubits increases.
A quantum processing unit with a readout resonator that can sequentially read out multiple qubits far from the resonator using swap gates to transfer quantum states to a qubit closest to the resonator, allowing dynamic adaptation of readout order and grouping to suit computing operations.
This approach enables efficient reading of a large number of qubits without significant space or cost overhead, maintaining coherence time and allowing flexible operation tailored to specific computing tasks.
Smart Images

Figure 2024541372000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to the technical field of quantum computing, in particular to reading out the state of multiple qubits in a quantum processing unit. [Background technology]
[0002] The basic functional unit of quantum computing is the qubit, and there may be many qubits on a quantum processing unit. Throughout this specification, the term quantum processing unit and its acronym QPU refer to hardware at least a portion of which is suitable for quantum computing and in which multiple circuit elements designed for quantum computing are present in a physical form suitable for operating in the cryogenically cooled environment required for quantum computing. The term quantum circuit refers to a configurable abstraction of the quantum gates that are executed during a quantum computation. The term quantum computing system refers to a larger entity that includes one or more QPUs, a controller outside the cryogenically cooled environment, and the signal paths between the two.
[0003] Each qubit used in quantum computing assumes a superposition of two basis states. For concise reference, the superposition is often referred to as the quantum state of the qubit, or simply the state of the qubit. In general, multi-qubit systems are in a superposition of multi-qubit eigenstates.
[0004] To obtain a useful result of a quantum computation, a read operation must be performed. A read operation collapses the quantum state of a single qubit into one of the possible basis states, resulting in a classical state that can be represented as a digital 1 or a digital 0. A representative property of any quantum circuit is the coherence time, during which the read operation must be performed to avoid loss of the information represented by the quantum state.
[0005] A known method for reading out a qubit involves the use of a readout resonator. The qubit is weakly coupled to an adjacent readout resonator, and the energy in the qubit induces a small shift in the scattering parameters of the coupled system consisting of the qubit and the readout resonator, due to the nonlinearity of the qubit. This shift can be detected by transmitting a so-called readout signal, which is a microwave pulse resonant with the readout resonator.
[0006] Assuming that the qubits are transmons, the interaction between the qubit state and the readout signal injected into the readout resonator produces an observable effect on the amplitude and phase of the transmitted signal. This effect is indicative of the classical state observed upon readout. From the readout resonator, there is a further signal path that finally transfers the retrieved classical state out of the cryogenically cooled environment in which the quantum processing unit resides.
[0007] The readout resonator must be located close to the qubits whose states are to be read out. Building a signal path between the readout resonator in a cryogenically cooled environment and the processing electronics in the ambient room temperature environment is not a trivial task, as it requires transmission lines capable of operating at gigahertz frequencies with appropriate filtering and thermal anchoring to the cold object in the cryostat. In practice, frequency multiplexing can be used to share a common transmission line between about 10 readout resonators, with the limit related to the gate speed and available bandwidth. The slower the gates, if the hardware and software support it, the more channels can be frequency multiplexed, but slow gates work against the overall goal of making the most of the limited coherence time of the qubits. In systems where the quantum processing unit contains only a small number of qubits, these are not a big problem. However, as the number of qubits in the quantum processing unit increases, it has been found that the physical space requirements of known readout systems, as well as the conduction heat and costs associated with large amounts of wiring, can become prohibitive. [Prior art documents] [Patent documents]
[0008] [Patent Document 1] WO2021 / 156542 Summary of the Invention [Problem to be solved by the invention]
[0009] It is an object to provide a quantum processing unit, a quantum computing system, and a method that allows the state of multiple qubits to be read out without requiring a large amount of space for readout-related circuitry. Another object is to provide a quantum processing unit, a quantum computing system, and a method that allows the readout operation to be tailored to specific needs in a system with a large number of qubits. A further object is to reduce the amount of conductive heat and costs associated with wiring in a quantum computing system.
[0010] These and other advantageous objects are accomplished by providing a readout resonator that can be used to sequentially read out multiple qubits, most of which are far away from the readout resonator, where swap gates are utilized to sequentially transfer the quantum state of each qubit to its closest location to a common readout resonator. [Means for solving the problem]
[0011] According to a first aspect, there is provided a quantum processing unit comprising a plurality of qubits and a readout resonator, each of the plurality of qubits configured to acquire a quantum state specific to that qubit as a result of a quantum computing operation, a first qubit of the plurality of qubits being located closest to the readout resonator. The quantum processing unit comprises a plurality of couplers configured to selectively execute swap gates between respective pairs of the plurality of qubits in response to respective swap gate control signals to swap states of the qubits of the respective pairs. The couplers are configured to cause the acquired quantum state of each of the plurality of qubits to appear in sequence to the first qubit by repeatedly executing the swap gates.
[0012] According to one embodiment, there are more than 10 qubits in said plurality of qubits, preferably more than 100 qubits, most preferably more than 1000 qubits. This is particularly with the advantage that the advantageous properties of this particular readout scheme become increasingly evident, since when applied with an increasing number of qubits, more substantial savings in space, manufacturing costs, hardware complexity and heat load of cryogenic cooling equipment are achieved.
[0013] According to one embodiment, the coupler is configured to controllably vary the order in which the acquired quantum states of each of the plurality of qubits are presented to the first qubit, with the particular advantage that the order in which the acquired quantum states of the qubits are read out can be dynamically adapted to the requirements of the quantum computing operation to be performed.
[0014] According to one embodiment, the plurality of qubits constitutes a first subset of all qubits in the quantum processing unit, the quantum processing unit comprising one or more other subsets of qubits. The quantum processing unit may then comprise a plurality of readout resonators, at least one of each of which is dedicated to each of the subsets of qubits. This entails the advantage in particular that even a very large number of qubits can be read out without spending more time and without losing more fidelity than is acceptable.
[0015] According to one embodiment, the coupler is configured to controllably vary the assignment of readout resonators of the plurality of readout resonators to subsets of qubits, with the particular advantage that the grouping of qubits, from which the acquired quantum states of the qubits are read out, can be dynamically adapted to the requirements of the quantum computing operation to be performed.
[0016] According to one embodiment, the quantum processing unit comprises a two-dimensional area on a substrate, said subset of qubits being spread across said two-dimensional area, which entails the particular advantage of being able to implement a very large number of qubits and read out their acquired quantum states in an appropriate space.
[0017] According to one embodiment, the readout resonators are located at one or more edges of the two-dimensional area, which entails the advantage in particular that wiring to the readout resonators is relatively simple and that the readout resonators do not take up space from a central part of the two-dimensional area.
[0018] According to one embodiment, at least part of the plurality of readout resonators are distributed over the two-dimensional area, which entails in particular the advantage that the grouping of qubits can be more freely determined.
[0019] According to one embodiment, the first subset includes tuned qubits and either parked qubits or fixed frequency qubits in an alternating pattern, where the parked qubits are qubits held at their respective resonant frequencies. The resonant frequencies of the parked qubits or fixed frequency qubits in the alternating pattern may follow a rotation pattern of three different frequencies. The second subset may then include tuned qubits and either parked qubits or fixed frequency qubits in an alternating pattern, where the resonant frequencies of the parked qubits or fixed frequency qubits in the alternating pattern follow the same rotation pattern of three different frequencies as the first subset. The tuned qubits and parked qubits or fixed frequency qubits of the first and second subsets may be located at the vertices of a hexagonal grid, where the three nearest neighbors of each parked qubit or fixed frequency qubit in the grid are tuned qubits, and the three nearest neighbors of each tuned qubit in the grid are parked qubits or fixed frequency qubits, each of which has a unique one of the three different frequencies. This entails, among other things, the advantage that a relatively simple readout scheme can be implemented without the risk of introducing unwanted interference between qubits of neighbouring subsets.
[0020] According to a second aspect, there is provided a quantum computing system comprising a quantum processing unit of any kind as introduced above, and a controller, and a number of signal paths between said quantum processing unit and said controller, said controller being configured to provide said swap gate control signal to a coupler in said quantum processing unit.
[0021] According to one embodiment, the control device is configured to dynamically determine a read-out order of at least a portion of the plurality of qubits in the quantum processing unit and to provide the swap control signals to couplers according to the dynamically determined read-out order, such that the obtained quantum states of each of the at least a portion of the plurality of qubits appear at the first qubit in the determined read-out order, which entails in particular the advantage that the operation of the quantum computing system can be flexibly optimized for each quantum computing task to be performed.
[0022] According to one embodiment, the plurality of qubits in the quantum processing unit includes computation qubits and ancilla qubits. The controller may then be configured to read out the state of at least some of the ancilla qubits multiple times before reading out the state obtained by the computation qubits. This has the advantage in particular of enabling effective error detection without sacrificing any other advantageous properties of the system.
[0023] According to one embodiment, the computation qubits arrange a first matrix pattern on a surface of a quantum processing unit and the ancilla qubits arrange a second matrix pattern on the surface of the quantum processing unit that is intertwined with the first matrix pattern, with the particular advantage that all computation qubits, or at least a large proportion of computation qubits, can be subjected to error correction using ancilla qubits.
[0024] According to a third aspect, there is provided a method for reading out a state of a qubit in a quantum computing system, the method comprising repeatedly performing a swap gate among a plurality of qubits to sequentially present to a first qubit of the plurality of qubits a state acquired by each of the plurality of qubits as a result of a quantum computing operation, the method also comprising repeatedly performing a read operation on the first qubit, thereby sequentially reading out the acquired state of each of the plurality of qubits that has been sequentially presented to the first qubit.
[0025] The accompanying drawings are included to provide a further understanding of the invention, constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention. [Brief description of the drawings]
[0026] [Figure 1] FIG. 2 illustrates a quantum processing unit. [Diagram 2] FIG. 13 illustrates an example of a read and swap gate operation sequence. [Diagram 3] FIG. 2 illustrates a quantum processing unit. [Figure 4] FIG. 1 illustrates a possible readout order of qubits in a quantum processing unit. [Diagram 5] FIG. 1 illustrates a possible readout order of qubits in a quantum processing unit. [Figure 6] FIG. 1 illustrates a possible readout order of qubits in a quantum processing unit. [Figure 7] FIG. 1 illustrates an example of the use of an ancilla qubit. [Figure 8] FIG. 1 illustrates a quantum computing system. [Figure 9] FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0027] 1 is a simplified schematic diagram of a quantum processing unit 101. It includes a plurality of qubits, here shown as rectangles with rounded corners numbered 0 through 9. In addition, the quantum processing unit 101 comprises a readout resonator 102. Each of the plurality of qubits is configured to obtain a quantum state specific to that qubit as a result of a quantum computing operation. In order to obtain useful results from the quantum computing operation, a readout operation needs to be performed on each qubit, or at least a selected group of qubits.
[0028] Of the multiple qubits, one qubit is located closest to the readout resonator 102. This closest qubit can be referred to as the first qubit, although it should be noted that the numerical designations first, second, etc. are used for clear reference only, without necessarily indicating any geometric or other defined order. In the schematic of FIG. 1, the first qubit is the leftmost one, numbered 0.
[0029] In addition, the quantum processing unit includes a number of couplers, of which coupler 103 is shown as an example, located adjacent to the leftmost qubit and the immediately adjacent second-left qubit. The couplers are configured to selectively perform swap gates between respective pairs of qubits in response to respective swap gate control signals 104. A swap gate between a pair of qubits is an operation that swaps the quantum states of these two qubits. The swap gate control signal 104 provided to the leftmost coupler 103 in FIG. 1 swaps the states of the leftmost qubit and the second-left qubit in FIG. 1.
[0030] In particular, the coupler is configured to cause the obtained quantum state of each of the multiple qubits to appear in turn at the first (i.e. leftmost) qubit in Figure 1 by repeatedly executing the swap gate. In this manner, only one readout resonator is required to perform all readout operations required to read out the states of the ten qubits in quantum processing unit 101.
[0031] As an example, the captured state of the first (leftmost) qubit in Figure 1 can be read out using readout resonator 102 without performing any swap gates. A single swap gate can then be performed using coupler 103 to make the quantum state first captured by the second-left qubit appear at the first qubit for readout using readout resonator 102. Then, two swap gates can be performed in sequence using the second-left and leftmost couplers, after which the quantum state first captured by the third-left qubit appears at the first qubit for readout, and so on.
[0032] A notation may be adopted in which Rx means to read out the state of the x'th qubit in FIG. 1. In this example x=0 since the readout resonator 102 is adjacent to only the 0th qubit. Similarly, the notation adopted Syz (y∈[0,8],z=y+1) means a swap gate between the y'th and z'th qubits. Using this notation, successive cycles of readout and swap gate operations can be shown as in the table of FIG. 2.
[0033] In FIG. 2, the hatched edges of the table cells highlight the readout operations. In row number 0, the consecutive numbers from 0 to 9 indicate the acquired quantum states of the qubits in the exemplary 10-qubit quantum processing unit of FIG. 1. Executing R0, i.e., reading out the quantum state of the 0th qubit, collapses it to a classical state. This is shown as 0' in FIG. 2 from row 2 onwards. Row number 1 of the table shows how the states of the two leftmost qubits are swapped by executing a swap gate S01, so that the acquired quantum state 1 of the 1st qubit appears in the leftmost qubit and the collapsed classical state 0' of the 0th qubit then appears in the second qubit from the left. In row number 2, the readout operation R0 reads out the quantum state 1 of the 1st qubit and collapses it to the classical state 1'. In the same row, a swap gate S12 swaps the states of the second and third qubits from the left. Similar operations continue so that each quantum state is transferred to the leftmost qubit and read out in turn, and the resulting classical state begins to move to the right in the column of qubits.
[0034] Row number 10 of the table in Figure 2 shows how the classical states 1' and 0' have been transferred to the two rightmost qubits. Note that the operations R0-S12-S34-S56-S78 in row number 10 are the same as in row number 8. The table can simply continue, with row number 11 repeating the operations in row number 7, row number 12 repeating the operations in row number 6, and so on. After a total of 18 rows, the quantum states previously obtained by all qubits have been read out and the resulting classical states will appear to the qubits in a mirrored order relative to row number 0 of the table.
[0035] The single row of qubits in Figure 1 can be thought of as representing a subset of qubits in a larger quantum processing unit. As an example, one might think of 1024 qubits arranged in a rectangular array of 32 rows and 32 columns, with each row having a row-specific readout resonator at the end of the array. Furthermore, assuming that a common transmission line can be shared among the eight resonators through frequency multiplexing, only four input lines and four output lines are required to read out all 1024 qubits, in addition to the control lines needed to implement the swap gates.
[0036] Furthermore, assuming that all rows can be read and all swap gates can be performed in parallel, we can estimate the time required for a read. If we tabulate the read sequence for a single row, as in the table in Figure 2, this means that 32 rows contain both read operations and swap gates, and an additional 32 rows contain only swap gates. If a swap gate can be performed in 50 nanoseconds and a read operation takes 100 nanoseconds, then the 32+32 rows of the table can be executed in 4.8 microseconds. Given that coherence times of the order of 100 microseconds have been demonstrated up to the time of writing of this text, this is more than fast enough to perform a read.
[0037] In the above example, the captured quantum state of the last qubit in the row will undergo 32 swap gates before readout. Assuming a fidelity of 99.95 percent per swap gate, this means that the worst-case fidelity loss will be on the order of less than 2 percent (0.9995^32 ≒ 0.9841).
[0038] As mentioned above, the advantages achieved by the proposed readout method are most pronounced when the number of qubits in the quantum processing unit is large: there may be more than 10 qubits, more than 100 qubits, or more than 1000 qubits in the quantum processing unit.
[0039] Most preferably, the couplers configured to perform the swap gates are configured to controllably change the order in which the acquired quantum states of each qubit are presented to the qubits adjacent to the readout resonator. That is, by selecting the order in which the swap gates are performed, qubits sharing a common readout resonator can be subjected to readout operations in a desired order. As a simple example, if the first four qubits in FIG. 1 (qubits 0, 1, 2, and 3) are to be read out in the order 0-3-1-2, the operations R0, S01-S23, S12, S01-S23, R0, S01, R0-S12, S01, R0 can be performed.
[0040] As already shown in the example of 1024 qubits in a 32x32 array, there may be an initial subset of qubits and one or more other subsets. The quantum processing unit may then comprise multiple readout resonators, at least one of each of which is dedicated to each subset of qubits. The example in which each subset is a row of qubits in a rectangular array, with a dedicated subset-specific readout resonator at one end of the row, is for illustrative purposes only. Further examples for geometrically arranging the subsets and readout resonators are provided below.
[0041] The coupler may be configured to controllably change the allocation of readout resonators to subsets of qubits. This means that the chain of qubits may be different at different times in which the acquired quantum state leads to a particular qubit for readout. In a very simple example, the quantum processing unit of FIG. 1 may have two readout resonators, one at each end of a row of qubits. At one time, the coupler may, for example, make the 0th, 1st, 2nd, 3rd, and 5th acquired quantum states appear in sequence to the 0th qubit, and the 4th, 6th, 7th, 8th, and 9th qubits appear in sequence to the 9th qubit, by repeatedly executing swap gates. At another time, this division into subsets may be done in a different way.
[0042] FIG. 3 illustrates an example of a quantum processing unit that includes a two-dimensional area on a substrate 301. A subset of qubits spans the two-dimensional area. In this embodiment, the qubits form a rectangular array. Couplers are not shown in FIG. 3 for clarity. A plurality of readout resonators 302 are located at one end of the two-dimensional area. In this embodiment, a second plurality of readout resonators 303 are located at a second end of the two-dimensional area.
[0043] The principles illustrated in Figure 3 are well suited to controllably varying the assignment of readout resonators to subsets of qubits. For example, at one point in time, the subset of qubits dedicated to one of the readout resonators 302 at the left-hand side of Figure 3 may include rows of qubits such as qubits 100, 101, 102, 103, 104, etc. At another point in time (or even the same time), the subset of qubits dedicated to one of the readout resonators 303 at the top of Figure 3 may include columns of qubits such as qubits 001, 101, 201, 301, 401, etc.
[0044] The qubits need not form a rectangular array on the surface of the substrate. Even if they do, the subset of qubits need not be defined as a linear subset, such as a row or column. FIG. 4 shows an example where the subset of qubits includes a 4×4 array of 16 qubits, with a readout resonator 401 located adjacent to a qubit 402 in the top left corner of the array. In FIG. 4, for clarity of illustration, couplers are not shown. However, couplers are present and are configured to sequentially force the acquired quantum state of each of the 16 qubits to appear in the top left qubit 402 by repeatedly executing swap gates. The order in which the readout occurs follows the zigzag pattern of the arrows in FIG. 4. For example, to read out the acquired quantum state of the bottom right qubit, the quantum state is first swapped one qubit to the left, then diagonally to the top right, then straight up, then diagonally to the bottom left, and so on.
[0045] FIG. 5 illustrates some exemplary principles that are independent of each other in themselves and can be applied in combination with other principles as well. First, the qubits in FIG. 5 do not form a rectangular array, but some other, here hexagonal, array. Thus, it is not necessary to arrange the qubits in any kind of regular array. Second, at least some of the multiple readout resonators are distributed over the two-dimensional area in which the qubits span. The readout resonators are represented diagrammatically as circles, see for example readout resonator 501. The individual readout resonators can be present, for example, in a checkerboard arrangement, or generally in a central area of the quantum processing unit, with the subset of qubits targeted to each readout resonator typically located in the immediate surrounding area of such readout resonator.
[0046] Third, the (default) subsets of qubits are not linear, but form a compact (here regular hexagonal) pattern around a dedicated readout resonator. Fourth, the division of qubits into subsets need not be constant; for example, qubit 502 in Figure 5 could be read out as part of the bottom subset, as indicated by the solid arrow, or as part of the top right subset, as indicated by the dashed arrow, by appropriately controlling swap gates implemented by couplers (not shown in Figure 5 for clarity).
[0047] FIG. 6 shows an example of an embodiment of a drive in which half of the qubits are tuned qubits and the other half are parked qubits or fixed frequency qubits. Parked qubits are qubits that can be tuned to different resonant frequencies by their structure and associated circuitry, but are maintained at their respective resonant frequencies for the time being. Here, the parked qubits have three different frequencies, marked A, B, and C, respectively. Fixed frequency qubits are qubits that have a single fixed resonant frequency A, B, or C. The use of fixed frequency qubits entails several advantages in simpler structure and electrical design, while the use of parked qubits entails the advantage of not having to stick to a predefined frequency.
[0048] Tuned qubits are marked with an X. A first subset of qubits 601 is highlighted with a dashed oval. The first subset is formed of a chain of qubits and includes tuned qubits and park or fixed frequency qubits in an alternating pattern, such that every third qubit is a tuned qubit and the other qubits are park or fixed frequency qubits. The resonant frequencies of the park or fixed frequency qubits follow a rotational pattern ABCABC... so that overall the first subset of qubits 601 is AXBXCXAXBXCX....
[0049] The second subset of qubits is similar to the first subset of Figure 6. In other words, the second subset includes tuned qubits and park or fixed frequency qubits in an alternating pattern, where the resonant frequencies of the park or fixed frequency qubits in the alternating pattern follow the same rotation pattern of three different frequencies as the first subset.
[0050] In the embodiment of Figure 6, the tuned qubits and parked qubits or fixed frequency qubits of the first and second subsets (as well as further subsets) are located at the vertices of a hexagonal grid. In particular, the three nearest neighbors of each parked qubit or fixed frequency qubit in said grid are tuned qubits, and the three nearest neighbors of each tuned qubit in said grid are parked qubits or fixed frequency qubits, each of which has a unique one of three different frequencies A, B, and C.
[0051] This arrangement of tuned qubits and park or fixed frequency qubits allows a scheme where successive tuning steps of the tuned qubit bring it into resonance with the appropriate park qubit at each instant in time. For example, considering tuned qubit 602, it may be tuned to resonate with the park or fixed frequency qubit at resonant frequency C in the top right in the same subset 601. The tuned qubit 602 may then be tuned to resonate with the park or fixed frequency qubit at resonant frequency A below it in the same subset 601. None of these tuning steps cause unwanted resonance with the nearest park or fixed frequency qubit 603 of a different subset, which is the park or fixed frequency qubit at resonant frequency B in the top left. By combining flux and drives, this scheme requires 1024 control lines for a grid of 1024 qubits, and 8 lines to read them all out (assuming 8x frequency multiplexing with 4 input and 4 output lines to read out 32 qubits simultaneously).
[0052] In an embodiment such as FIG. 6, or indeed in any embodiment, the number of control lines required to execute a swap gate can be effectively reduced by applying the additive control principle described in the co-pending patent application published as WO2021 / 156542. In essence, the additive control principle involves sending a weak control signal through an array of control lines of a quantum processing unit. A weak control signal alone will not have a significant effect on a qubit (or other controllable circuit element) close to that control line. However, a qubit (or other controllable circuit element) that has two or more control lines passing nearby will experience the combined effect of those control signals as the weak control signal propagates through each of said two or more control lines simultaneously, and will be affected as intended. For example, assuming a 32×32 rectangular grid of qubits, 64 control lines (32 along the rows of qubits and 32 along the columns) may be sufficient. For simplicity, this calculation excludes terminations. As previously mentioned, a total of eight additional lines for readout may be sufficient (assuming 32 readout resonators and 8x frequency multiplexing).
[0053] Repeated swap gates between qubits cause only a small degradation in fidelity, but such degradation cannot be completely avoided. However, the readout scheme described above allows many more qubits to be incorporated into a quantum computing system that allows only a certain number of signal paths between the cryogenically cooled and room temperature environments. The quantum error correction provided by the appropriate use of such a large number of qubits may well outweigh the potential sources of error caused by the degradation in fidelity.
[0054] Figure 7 shows an example of a quantum processing unit that includes so-called ancilla qubits as well as actual computation qubits. As is well known in the art, ancilla qubits can be used for parity measurements as part of quantum error correction. In the embodiment of Figure 7, the ancilla qubits and computation qubits form a checkerboard pattern in which every other qubit is an ancilla qubit, as shown by the shading in Figure 7. In other words, the computation qubits form a first matrix pattern on the surface of the quantum processing unit, and the ancilla qubits form a second matrix pattern on the surface of the quantum processing unit that is intertwined with the first matrix pattern.
[0055] The ancilla qubits are the ones that need to be read out frequently, whereas the computation qubits only need to be read out at the end of the quantum computation. Reading out the ancilla qubits does not disturb the actual data present in the quantum state of the computation qubits. The quantum processing unit may include a dedicated readout resonator for the ancilla qubits to read out as fast as possible. Alternatively, there may be a relatively small subset of the ancilla qubits, each with its own dedicated readout resonator. At the end of the quantum computation, a swap gate may be performed and the state of the computation qubits may finally appear in the ancilla qubits with their readout resonators.
[0056] 8 shows a schematic representation of a quantum computing system. It comprises a quantum processing unit 801, which may be of any kind described above. The quantum processing unit 801 is located in a cryogenically cooled environment 802. The quantum computing system comprises a controller 803 located (at least partially) outside of the cryogenically cooled environment 802. In addition, the quantum computing system comprises a number of signal paths 804 between the quantum processing unit 801 and the controller 803. In accordance with the above description, the controller 803 is configured to provide swap gate control signals to couplers in the quantum processing unit 801 to cause the captured quantum states of the qubits in each subset to appear to subset-specific qubits located sufficiently close to the nearest readout resonator.
[0057] Controller 803 may be configured to dynamically determine a readout order for at least a portion of the plurality of qubits in quantum processing unit 801. Dynamically determining the readout order provides flexibility to accommodate different needs for reading out the states of various qubits that may arise as the quantum computation progresses. Controller 803 may then be configured to provide the swap control signals to a combiner in quantum processing unit 801 according to the dynamically determined readout order. This causes the obtained quantum states of each of the qubits (or at least a desired portion of the qubits) to appear in one or more "readout qubits" in the determined readout order.
[0058] Where quantum processing unit 801 has computation qubits and ancilla qubits, the controller may be configured to read out the states of at least some of the ancilla qubits multiple times before reading out the states acquired by the computation qubits. As explained above, frequently reading out the states of the ancilla qubits may help to perform quantum error correction without disturbing the information present in the quantum states of the computation qubits.
[0059] The quantum computing system may include a user interface 805 for local control of its operation and / or an external connection 806 for remote control of its operation, for exchanging information with other systems.
[0060] FIG. 9 illustrates a method for reading out the state of a qubit in a quantum computing system. The method includes repeatedly performing a swap gate between a plurality of qubits to sequentially present a state acquired by each of the plurality of qubits to a first qubit of the plurality of qubits as a result of a quantum computing operation. In addition, the method includes repeatedly performing a read operation on the first qubit, thereby sequentially reading out the acquired state of each of the plurality of qubits that has been sequentially presented to the first qubit. In the diagram of FIG. 9, a swap gate is performed in step 903 and a read operation is performed in step 901. These iterations result from a decision step 902 in which it is checked whether the states of all desired qubits in a subset have already been read out. The loop consisting of steps 901, 902, and 903 is repeated until a positive finding is made in step 902, at which point the method proceeds to end.
[0061] It is obvious to those skilled in the art that with the advancement of technology, the basic concept of the present invention can be implemented in various ways. Therefore, the present invention and its embodiments are not limited to the above examples, but can instead be modified within the scope of the claims. [Explanation of symbols]
[0062] 101 Quantum Processing Unit 102 Readout resonator 103 Coupler 104 Swap Gate Control Signal 301 Substrate 302 Readout resonator 303 second plurality of readout resonators 401 Readout resonator 402 Cubit 501 Readout resonator 502 Cubit 601 Cubit 602 Adjusted Cubit 603 Fixed Frequency Qubits 801 Quantum Processing Unit 802 Cryogenic cooling environment 803 Control device 804 Signal Path 805 User Interface 806 External Connection
Claims
1. a quantum processing unit comprising a plurality of qubits and a readout resonator; each of the plurality of qubits is configured to obtain a quantum state specific to that qubit as a result of a quantum computing operation; a quantum processing unit, wherein a first qubit of the plurality of qubits is located closest to the readout resonator; the quantum processing unit comprises a plurality of couplers configured to selectively execute swap gates between respective pairs of the plurality of qubits in response to respective swap gate control signals to swap states of the qubits in the respective pairs; the coupler is configured to cause the acquired quantum state of each of the plurality of qubits to appear in sequence to the first qubit by repeatedly executing the swap gate.
2. 2. The quantum processing unit of claim 1, wherein the plurality of qubits is greater than 10 qubits, preferably greater than 100 qubits, and most preferably greater than 1000 qubits.
3. 3. The quantum processing unit of claim 1, wherein the coupler is configured to controllably vary the order in which the acquired quantum states of each of the plurality of qubits are presented to the first qubit.
4. the plurality of qubits constituting a first subset of all qubits in the quantum processing unit; the quantum processing unit includes one or more other subsets of qubits; the quantum processing unit comprises a plurality of readout resonators, at least one of which is available for dedication to each of said subsets of qubits; 4. A quantum processing unit according to any one of claims 1 to 3.
5. the coupler is configured to controllably vary an assignment of readout resonators of the plurality of readout resonators to a subset of qubits.
5. The quantum processing unit of claim 4.
6. the quantum processing unit comprises a two-dimensional area on a substrate; the subset of qubits is spread across the two-dimensional area; 6. A quantum processing unit according to claim 4 or 5.
7. 7. The quantum processing unit of claim 6, wherein the plurality of readout resonators are located at one or more ends of the two-dimensional area.
8. 7. The quantum processing unit of claim 6, wherein at least a portion of the plurality of readout resonators are distributed across the two-dimensional area.
9. the first subset includes tuned qubits and either park qubits or fixed frequency qubits in an alternating pattern, the park qubits being qubits held at their respective resonant frequencies; the resonant frequencies of the park qubits or the fixed frequency qubits in the alternating pattern follow a rotation pattern of three different frequencies; a second subset includes tuned qubits and either park qubits or fixed frequency qubits in an alternating pattern, the resonant frequencies of the park qubits or fixed frequency qubits in the alternating pattern following the same rotation pattern of three different frequencies as the first subset; the tuned qubits and the park qubits or fixed frequency qubits of the first and second subsets are located at the vertices of a hexagonal grid, and the three nearest neighbors of each park qubit or fixed frequency qubit in the grid are tuned qubits, and the three nearest neighbors of each tuned qubit in the grid are park qubits or fixed frequency qubits, each of which has a unique one of the three distinct frequencies.
9. A quantum processing unit according to any one of claims 6 to 8.
10. 1. A quantum computing system, comprising: A quantum processing unit according to any one of claims 1 to 9; A control device; a plurality of signal paths between said quantum processing units and said controller; Equipped with the controller is configured to provide the swap gate control signal to the coupler in the quantum processing unit. Quantum computing systems.
11. The control device, dynamically determining a readout order of at least a portion of the plurality of qubits in the quantum processing unit; providing the swap control signal to the coupler in accordance with the dynamically determined readout order to cause the acquired quantum state of each of the at least some of the plurality of qubits to appear to the first qubit in the determined readout order.
11. The quantum computing system of claim 10, configured to:
12. the plurality of qubits in the quantum processing unit includes computation qubits and ancilla qubits; the controller is configured to read out the states of at least some of the ancilla qubits multiple times before reading out the states acquired by the computation qubits.
12. A quantum computing system according to claim 10 or 11.
13. 13. The quantum computing system of claim 12, wherein the computational qubits form a first matrix pattern on a surface of the quantum processing unit and the ancilla qubits form a second matrix pattern on the surface of the quantum processing unit that is intertwined with the first matrix pattern.
14. 1. A method for reading out the state of a qubit in a quantum computing system, comprising: repeatedly performing a swap gate among a plurality of qubits to sequentially cause a state acquired by each of the plurality of qubits to appear at a first qubit of the plurality of qubits as a result of a quantum computing operation; repeatedly performing a read operation on the first qubit, thereby sequentially reading out the acquired states of each of the plurality of qubits that have been sequentially presented to the first qubit; A method comprising:
Citation Information
Patent Citations
Additive control of qubits for enabling time-domain and frequency-domain multiplexing
WO2021156542A1
