Quantum TT&C waveform data compression method and system for heterogeneous communication

Abstract

The present application relates to the technical field of quantum computing heterogeneous communication, and particularly relates to a quantum measurement and control waveform data compression method and system for heterogeneous communication, which quantizes original waveform data, calculates the hash key of the quantized waveform, and stores the hash key in a waveform dictionary for deduplication, wherein the waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary; a playback instruction sequence is used to store the waveform identifier of each original waveform, the absolute starting time of the waveform on the global time axis, and the corresponding target channel identifier; the waveform dictionary, the playback instruction sequence, and the original waveform metadata are serialized into a byte stream, and the data packet is transmitted to a quantum measurement and control device through compression and packaging, so that the quantum measurement and control device recovers and acquires the waveform dictionary and the playback instruction sequence by deserializing the received data packet. The present application can realize low-delay transmission of waveform data in a heterogeneous fusion architecture, and provides key technical support for the scalability of large-scale quantum computing systems.

Description

Technical Field

[0001] This invention relates to the field of quantum computing heterogeneous communication technology, and in particular to a quantum measurement and control waveform data compression method and system for heterogeneous communication. Background Technology

[0002] As the scale of superconducting quantum computing systems expands, the amount of waveform data increases exponentially with the number of qubits, posing a severe challenge to the storage bandwidth and transmission paths (such as PCIe) of the measurement and control system. In the existing hierarchical architecture of "classical computing node - quantum measurement and control equipment - quantum chip", the massive amount of waveform data transmission not only consumes hardware resources but also introduces a significant transmission delay, severely restricting the timeliness of quantum operations.

[0003] Currently, quantum measurement and control systems mainly employ two waveform data transmission methods: one is transmitting parameterized pulse descriptions, and the other is transmitting raw waveform samples. Regarding compression technology, existing solutions have significant shortcomings. In the general compression field, algorithms such as Zlib and LZ4 are widely used for data compression, but these algorithms treat waveforms as ordinary floating-point sequences, failing to perceive the inherent repetitiveness and structure in quantum control logic. This results in low compression ratios in specific scenarios and high decompression computational overhead, making it difficult to meet the real-time response requirements of quantum measurement and control systems. In terms of dedicated data processing, although some systems reduce data volume by defining parameterized pulse libraries, this sacrifices flexibility and cannot adapt to arbitrary waveforms from pulse-level optimizations such as DRAG or optimal control. Real-time synthesis also consumes FPGA logic resources that could be used for feedback processing, limiting the complexity of the control sequences that can be generated. Furthermore, existing waveform storage often uses "continuous arrays" or "relative timestamp" models, which are prone to timing drift due to accumulated errors during decompression and reconstruction, affecting the synchronization accuracy of quantum gate operations. Frequency domain compression methods such as COMPAQT use Discrete Cosine Transform (DCT) to compress waveforms in the frequency domain, storing only significant coefficients to improve bandwidth. However, this approach targets reduced on-chip memory rather than optimized communication links, and the approximation errors introduced by its lossy frequency domain methods may accumulate in deep circuits, and it does not effectively utilize pulse repetition structures.

[0004] To address the issue of excessive memory usage during waveform instruction compilation, some waveform processing solutions have been proposed for quantum computing measurement and control experiments, such as Chinese patent (publication number 121390342A). This solution optimizes the quantum measurement and control experiment process by compressing the memory representation of waveform instructions and reducing the storage volume of spliced ​​waveforms through parameter separation. However, it does not deduplicate the waveform sample data itself. The transmission of waveform instructions relies on predefined parameterized waveform shapes (such as Constant, SquareEnvelop, etc.) and describes the waveform through parameters such as amplitude, frequency, phase, duration, and bias voltage. This parameterized description method cannot adapt to arbitrary waveforms generated by pulse-level optimizations such as DRAG (Derivative Removal by Adiabatic Gate) or optimal control, sacrificing the flexibility of waveform definition and limiting the complexity of the control sequences that can be generated. By separating and storing static timing data from scan parameters, avoiding the need to store a complete stitching waveform instruction for each scan point, instruction-level memory compression is achieved. Essentially, this involves parameter separation and dynamic replacement. The static timing data in the transmitted waveform instructions uses relative time description (e.g., the time in the waveshape field represents the duration), while the scan data stores multiple output parameter values ​​in array form. Timing accuracy depends on the sequential processing during compilation. Therefore, a dedicated compression scheme for quantum waveform data is urgently needed to meet the requirements of large-scale quantum computing tasks in terms of data redundancy, real-time transmission, and timing reconstruction accuracy. Summary of the Invention

[0005] To address the communication bottleneck caused by the surge in waveform data volume due to the expansion of bit size in large-scale quantum computing, this invention provides a quantum measurement and control waveform data compression method and system for heterogeneous communication. It reconstructs the underlying structure, matching granularity, and timing model, and deeply integrates quantum data characteristics and data compression technology through waveform dictionary, quantization matching, absolute time modeling, and hybrid coding. This enables low-latency transmission of waveform data in a "classical + quantum" heterogeneous fusion architecture, providing key technical support for the scalability of large-scale quantum computing systems.

[0006] According to the design scheme provided by the present invention, on the one hand, a quantum measurement and control waveform data compression method for heterogeneous communication is provided, comprising:

[0007] Read the raw waveform data generated after quantum circuit scheduling, and obtain the waveform identifier, target channel identifier, absolute start time and metadata of each waveform;

[0008] The original waveform data is quantized, the hash key of the quantized waveform is calculated, and the hash key is used to store the waveform data in the waveform dictionary for deduplication. The waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary.

[0009] The waveform identifiers of each original waveform, the absolute start time of the waveform on the global time axis, and the corresponding target channel identifier are stored using the playback instruction sequence.

[0010] The waveform dictionary, playback instruction sequence, and original waveform metadata are serialized into a byte stream and then compressed and encapsulated before being transmitted to the quantum measurement and control equipment. This allows the quantum measurement and control equipment to recover and obtain the waveform dictionary and playback instruction sequence by deserializing the received data packets.

[0011] As a quantum measurement and control waveform data compression method for heterogeneous communication according to the present invention, the original waveform data generated after reading the quantum circuit scheduling includes:

[0012] Based on the quantum chip hardware topology and quantum computing task logic circuit, construct a hardware-executable underlying quantum circuit;

[0013] By scheduling parameters to be assigned to each quantum gate operation in the underlying quantum circuit, a sequence of original pulse instructions is generated.

[0014] As a quantum measurement and control waveform data compression method for heterogeneous communication according to the present invention, the original waveform data is further quantized, including:

[0015] Extract the real and imaginary parts of the original waveform separately, and determine the normalization scaling factor based on the dynamic range of the waveform;

[0016] The real and imaginary parts are truncated from the original floating-point range and scaled to an integer range of a specified bit depth.

[0017] As a quantum measurement and control waveform data compression method for heterogeneous communication according to the present invention, further, the hash key of the quantized waveform is calculated, and the hash key is used to perform deduplication storage in the waveform dictionary, including:

[0018] The hash key of the quantized waveform is calculated using a hash function, and then the hash key is used to look up the waveform in the waveform dictionary.

[0019] If the search in the waveform dictionary fails, all waveform samples in the waveform dictionary are traversed, the L2 distance between the quantized waveform and the waveform sample is calculated, and a matching sample of the quantized waveform is determined in the waveform dictionary based on the distance threshold.

[0020] If no matching sample exists in the waveform dictionary, the current quantized model is used as the new waveform sample, and the new waveform sample is added to the waveform dictionary by assigning a new index.

[0021] As a quantum measurement and control waveform data compression method for heterogeneous communication according to the present invention, further, the method utilizes a playback instruction sequence to store the original waveform identifier and the absolute start time and position of the waveform on the global time axis, including:

[0022] Based on the timing of each pulse waveform duration, the absolute start time of the pulse waveform in the entire pulse waveform sequence is determined. The entire pulse waveform sequence consists of all pulses arranged on the same time axis when the quantum circuit executes the logic circuit once.

[0023] Each pulse waveform retains its waveform identifier, absolute start time, and destination channel identifier. All pulse waveforms are arranged in chronological order, and the relevant pulse waveform data is stored in the playback instruction sequence.

[0024] As a quantum measurement and control waveform data compression method for heterogeneous communication according to the present invention, further, the waveform dictionary, playback instruction sequence, and original waveform metadata are serialized into a byte stream, and the data packet is transmitted to the quantum measurement and control device through compression and encapsulation, including:

[0025] If the quantized waveform is stored as a quantum gate pulse waveform sample after deduplication in the waveform dictionary, then the dictionary index, playback instruction sequence and metadata corresponding to the waveform will be serialized into a byte stream and transmitted to the quantum measurement and control equipment through compression and encapsulation.

[0026] If, during the deduplication of the quantized waveform in the waveform dictionary, there are special non-quantum gate pulse waveform samples, the corresponding waveforms will be dynamically incrementally constructed and stored in the waveform dictionary to expand the waveform dictionary. The expanded waveform dictionary, playback instruction sequence, and metadata will then be serialized into a byte stream and transmitted to the quantum measurement and control equipment through compression and encapsulation.

[0027] As a quantum measurement and control waveform data compression method for heterogeneous communication according to the present invention, the quantum measurement and control device further includes, by deserializing the received data packets to recover and obtain the waveform dictionary and playback instruction sequence, the following:

[0028] Request a pre-allocated memory buffer and initialize it to zero;

[0029] Based on the waveform identifier and absolute start time, the corresponding waveform sample in the waveform dictionary is copied to the specified location in the buffer to perform waveform reconstruction.

[0030] Furthermore, this invention also provides a quantum measurement and control waveform data compression system for heterogeneous communication, comprising: a waveform acquisition module, a quantization storage module, an instruction generation module, and a compression transmission module, wherein...

[0031] The waveform acquisition module is used to read the raw waveform data generated after quantum circuit scheduling, and to acquire the waveform identifier, target channel identifier, absolute start time and metadata of each waveform;

[0032] The quantization storage module is used to quantize the original waveform data, calculate the hash key of the quantized waveform, and use the hash key to store the waveform in the waveform dictionary for deduplication. The waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary.

[0033] The instruction generation module is used to store the waveform identifiers of each original waveform and the absolute start time and corresponding target channel identifier of the waveform on the global time axis using the playback instruction sequence;

[0034] The compression transmission module is used to serialize the waveform dictionary, playback instruction sequence, and original waveform metadata into a byte stream, and transmit the data packet to the quantum measurement and control equipment through compression and encapsulation, so that the quantum measurement and control equipment can recover and obtain the waveform dictionary and playback instruction sequence by deserializing the received data packet.

[0035] The beneficial effects of this invention are:

[0036] 1. This invention decomposes the original continuous waveform array into two parts: a "waveform dictionary" and a "play instruction sequence," achieving separation of concerns: the dictionary stores the shape of the pulses, and the instruction sequence stores their occurrence time and location. This separation reduces the data for each execution from O(N×L) to O(M×L+N), where N is the number of gate operands, L is the average pulse length, and M is the number of different pulse shapes. Typically, M≪N, thus achieving a compression rate exceeding 99%. Based on this, the system analyzes the pulse scheduling of the quantum circuit, identifying all Play instructions. By extracting pulse sample data and performing quantization matching, all duplicate waveform segments are removed, retaining only unique waveform samples stored in the "waveform dictionary." Next, a series of concise "play instructions" are generated; each instruction only records the waveform ID and absolute start time, without storing the actual waveform data. This method leverages the reusability of quantum gates to significantly compress the data size.

[0037] 2. This invention utilizes a waveform-level matching mechanism to ensure the correct handling of bit-dependent waveform variations. The quantization process quantizes waveform samples from complex128 format (64 floating-point per component) to a configurable bit depth (default 12 bits). After quantization, a hash key is generated for O(1) dictionary lookup. Waveform-level matching is bit-independent and gate-type-independent. The algorithm does not need to know the target pulse bit; it simply compares waveforms based on sampled values, automatically adapting to hardware heterogeneity. The matching process employs a hierarchical strategy to balance speed and robustness. The first level uses SHA-256 hashing for fast matching, achieving O(1) complexity with a hit rate exceeding 99.5%. The second level uses L2 distance tolerance matching to handle quantization boundary conditions and floating-point errors, with a threshold set to ε=10⁻. 6The noise level is much higher than that of floating-point noise but much lower than the difference in physical meaning. If both levels of matching fail, a new dictionary entry is created. The overall time complexity averages O(N), where N is the total number of pulses.

[0038] 3. This invention utilizes an "absolute time model" to ensure timing accuracy during decompression and reconstruction. The "start time" field in each playback instruction points to an absolute moment on the global timeline, rather than a time interval relative to the previous instruction. At the decompression end, the waveform reconstruction process no longer depends on the execution order of instructions, but directly writes waveform samples from the dictionary to the specified time offset position in the target buffer. This mechanism eliminates the accumulated errors caused by sequential execution, ensuring nanosecond-level synchronization accuracy of the control pulses for each qubit. Absolute time encoding makes each instruction self-contained; the FPGA can read a single instruction and immediately determine the channel, pulse shape, and precise start time without referencing any other instructions. This achieves parallel multi-channel decoding, robust error isolation, and simplified hardware design.

[0039] 4. This invention employs a hybrid encoding strategy to improve compression efficiency. The instruction encoding format includes: header, channel-by-channel data, and instruction encoding. Incremental encoding is applied to the time difference of consecutive pulses on the same channel to minimize the instruction size. Zlib compression is applied to the large dictionary portion, and finally, compressed data packets are generated by encapsulation. Combining the advantages of "dedicated dictionary" and "general entropy encoding," extremely high compression rates are achieved while ensuring data integrity. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the quantum measurement and control waveform data compression process for heterogeneous communication in the embodiment;

[0041] Figure 2 This is a schematic diagram of the waveform data generation process in the embodiment;

[0042] Figure 3 This is a schematic diagram of the principle architecture of a waveform data compression algorithm for heterogeneous communication between classical and quantum systems, as shown in the example.

[0043] Figure 4 This is a schematic diagram of the waveform data compression algorithm in the embodiment. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this invention clearer and more understandable, the invention will be further described in detail below with reference to the accompanying drawings and technical solutions.

[0045] In practical large-scale quantum computing tasks, the process of transmitting waveform data from classical computing nodes to the underlying quantum measurement and control system faces multiple bottlenecks. These bottlenecks mainly stem from the contradiction between the special characteristics of quantum control logic and the limitations of general-purpose computing architectures. The waveform data generation process is as follows: Figure 2 As shown, by progressively compiling and translating high-level quantum logic into low-level pulse signals, and combining data compression, channel-specific processing, and IQ quadrature modulation to generate analog waveforms that can directly drive qubits, the following problems exist in terms of data redundancy, real-time transmission, and time sequence reconstruction accuracy.

[0046] Firstly, regarding data redundancy, quantum computing tasks consist of a large number of fundamental quantum gates (such as H-gates, X-gates, and CX-gates). However, in existing data transmission processes, this waveform data is often repeatedly transmitted and stored as quantum gates are repeated. The data volume can be calculated mathematically: let N be the number of gate operands, L be the average pulse length (sampling points), and the original waveform data volume be O(N×L×16) bytes, where 16 bytes are in complex128 format. For example, for a quantum circuit containing N identical CX gates, a classical computer needs to send the exact same waveform data N times to the measurement and control system, resulting in a significant waste of storage space and bus bandwidth, severely slowing down waveform loading speed.

[0047] Secondly, regarding real-time transmission, as the number of qubits increases, the amount of waveform data expands dramatically, leading to a significant increase in PCIe transmission time. In scenarios requiring rapid iterative calibration or real-time control, data transmission accounts for a large portion of the system's time consumption, making it difficult for the classical computing layer to complete the feedback control loop within the quantum state decoherence time, thus limiting the system's operating efficiency.

[0048] Thirdly, regarding timing reconstruction accuracy, traditional compression schemes often employ sequential storage or relative time models. During decompression and reconstruction, waveform segments need to be spliced ​​sequentially, and the processing delay of preceding instructions accumulates in subsequent instructions, causing a slight shift in the actual pulse emission time. For quantum gate operations requiring high synchronization accuracy, this timing drift disrupts the phase alignment between qubits, introduces control errors, and reduces quantum gate fidelity.

[0049] Furthermore, the same type of quantum gate can produce different waveforms when applied to different physical qubits because each qubit has unique calibration parameters, including anharmonicity, transition frequency, and coupling strength. These parameters vary between qubits, and the calibration process adjusts the pulse parameters to compensate for these differences, resulting in bit-dependent pulse characteristics. Validation on a 127-qubit back-end model shows that single-qubit gates exhibit consistent durations across all 127 qubits, but the situation is different for two-qubit gates. The ECR gate exhibits six different durations across 144 qubit pairs, with a maximum to minimum duration ratio of 2.58. This variation has a direct impact on pulse-level operation—circuit-level multiplexing methods cannot capture this variation, leading to waveform reconstruction errors.

[0050] The essence of the above problems lies in the fact that classical data compression techniques have failed to utilize the "logical reusability" and "time absoluteness" characteristics of quantum data, and have failed to properly handle the "bit-dependent variation" characteristics.

[0051] Therefore, in the embodiments of the present invention, see Figure 1 As shown, a quantum measurement and control waveform data compression method for heterogeneous communication is provided, comprising:

[0052] S101. Read the raw waveform data generated after quantum circuit scheduling, and obtain the waveform identifier, target channel identifier, absolute start time and metadata of each waveform.

[0053] Specifically, based on the quantum chip hardware topology and the quantum computing task logic circuit, a hardware-executable underlying quantum circuit can be constructed; by scheduling, parameters are allocated to each quantum gate operation in the underlying quantum circuit to generate the original pulse instruction sequence.

[0054] By traversing all qubit channels, the Play command is identified, non-zero pulse segments are extracted, and the start time, duration, and sample data of each segment are recorded, while invalid zero-filled intervals are filtered out. Specifically, this includes: command filtering, traversing all commands in the pulse scheduling and filtering Play commands; sample extraction, retrieving pulse objects and extracting their samples; and metadata collection, collecting start time, duration, and target channel.

[0055] S102. Quantize the original waveform data, calculate the hash key of the quantized waveform, and use the hash key to store duplicates in the waveform dictionary. The waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary.

[0056] The compression target is reduced to the waveform sample data itself. By constructing a waveform dictionary and utilizing the physical property that the same quantum gate produces the same waveform, the waveform samples are deduplicated using the dictionary, and only the dictionary index and instruction sequence are transmitted, thus achieving sample-level compression.

[0057] Specifically, the quantization process for the original waveform data can be designed to include:

[0058] Extract the real and imaginary parts of the original waveform separately, and determine the normalization scaling factor based on the dynamic range of the waveform;

[0059] The real and imaginary parts are truncated from the original floating-point range and scaled to an integer range of a specified bit depth.

[0060] The extracted sample data is quantized, converting the complex128 format to a 12-bit integer. A quantization hash key is calculated, and a dictionary lookup is performed. If a sample already exists in the dictionary, only its index ID is recorded; otherwise, it is added to the dictionary and a new ID is assigned. The algorithm's time complexity is O(T), where T is the total number of Play instructions. Each instruction requires one quantization operation and one hash lookup, both of which are O(1) operations. This process achieves efficient deduplication of quantum gate waveforms and automatically adapts to bit-dependent mutations.

[0061] The process of calculating the hash key of the quantized waveform and using this hash key to perform deduplication in the waveform dictionary may include:

[0062] The hash key of the quantized waveform is calculated using a hash function, and then the hash key is used to look up the waveform in the waveform dictionary.

[0063] If the search in the waveform dictionary fails, all waveform samples in the waveform dictionary are traversed, the L2 distance between the quantized waveform and the waveform sample is calculated, and a matching sample of the quantized waveform is determined in the waveform dictionary based on the distance threshold.

[0064] If no matching sample exists in the waveform dictionary, the current quantized model is used as the new waveform sample, and the new waveform sample is added to the waveform dictionary by assigning a new index.

[0065] like Figure 3 As shown, the proposed solution utilizes the physical property that identical quantum gates with the same bits in a quantum circuit produce the same waveform, breaking through the limitations of traditional general compression algorithms. It constructs a compression model of "waveform dictionary + absolute time instruction", which significantly reduces the data size and transmission time, providing key technical support for the scalability of large-scale quantum computing systems.

[0066] Instead of relying on predefined parameterized waveform shapes, it directly quantizes and constructs dictionaries from extracted waveform sample data, supporting arbitrary waveforms generated by pulse-level optimizations such as DRAG and optimal control. Through complex128 format quantization to 12-bit integers, quantization hash key calculation, and dictionary lookup mechanisms, it achieves efficient deduplication of arbitrary waveforms and automatically adapts to bit-dependent variations.

[0067] S103. Use the playback instruction sequence to store the waveform identifier of each original waveform, the absolute start time of the waveform on the global time axis, and the corresponding target channel identifier.

[0068] Specifically, based on the timing of each pulse waveform duration, the absolute start time of the pulse waveform in the entire pulse waveform sequence is determined. The entire pulse waveform sequence consists of all pulses arranged on the same time axis when the quantum circuit executes the logic circuit once. Each pulse waveform retains a waveform identifier, absolute start time, and destination channel identifier. All pulse waveforms are arranged in chronological order, and the relevant pulse waveform data is stored in the playback instruction sequence.

[0069] An "absolute time model" is used to ensure timing accuracy. The start time field in each playback instruction points to an absolute moment on the global timeline, rather than a time interval relative to the previous instruction. Instruction sequences are generated based on extracted segment information. The instruction format is designed as: (Channel ID, Waveform ID, Absolute Start Time). Simultaneously, the maximum end time of all channels is calculated as the total waveform length and included in the metadata.

[0070] S104. Serialize the waveform dictionary, playback instruction sequence, and original waveform metadata into a byte stream, and transmit the data packet to the quantum measurement and control device through compression and encapsulation, so that the quantum measurement and control device can recover and obtain the waveform dictionary and playback instruction sequence by deserializing the received data packet.

[0071] The waveform dictionary, instruction sequence, and metadata are serialized into a byte stream. Larger dictionary portions are compressed using Zlib, and finally, a compressed data packet is generated.

[0072] If the quantized waveform is stored as a quantum gate pulse waveform sample after deduplication in the waveform dictionary, then the dictionary index, playback instruction sequence and metadata corresponding to the waveform will be serialized into a byte stream and transmitted to the quantum measurement and control equipment through compression and encapsulation.

[0073] If, during the deduplication of the quantized waveform in the waveform dictionary, there are special non-quantum gate pulse waveform samples, the corresponding waveforms will be dynamically incrementally constructed and stored in the waveform dictionary to expand the waveform dictionary. The expanded waveform dictionary, playback instruction sequence, and metadata will then be serialized into a byte stream and transmitted to the quantum measurement and control equipment through compression and encapsulation.

[0074] like Figure 4 As shown, the contradiction between data scale and real-time performance is systematically resolved through innovations in four aspects: “waveform-level dictionary construction”, “quantization matching mechanism”, “absolute time modeling” and “hybrid coding”.

[0075] After receiving the compressed packet, the quantum measurement and control equipment deserializes and recovers the dictionary and instructions. It requests a pre-allocated memory buffer and initializes it to zero. The instructions are parsed, and the corresponding waveform samples from the dictionary are directly copied to the specified locations in the buffer based on the absolute start time, completing rapid waveform reconstruction. The FPGA decoding module consists of three components: a dictionary memory (stored in onboard BRAM, organized as a lookup table indexed by pulse_id), an instruction decoder (implemented as a three-state finite state machine: IDLE, FETCH, WRITE), and a waveform assembly buffer (dual-port RAM, supporting simultaneous read and write operations).

[0076] The decompression end directly writes waveform samples from the dictionary to the specified time offset position in the target buffer, eliminating the accumulated errors caused by sequential execution and ensuring nanosecond-level synchronization accuracy of the control pulses for each quantum bit. Absolute time encoding makes each instruction self-contained, allowing the FPGA to read a single instruction and immediately determine the channel, pulse shape, and precise start time, achieving parallel multi-channel decoding and robust error isolation.

[0077] Furthermore, based on the above method, this embodiment of the invention also provides a quantum measurement and control waveform data compression system for heterogeneous communication, comprising: a waveform acquisition module, a quantization storage module, an instruction generation module, and a compression transmission module, wherein,

[0078] The waveform acquisition module is used to read the raw waveform data generated after quantum circuit scheduling, and to acquire the waveform identifier, target channel identifier, absolute start time and metadata of each waveform;

[0079] The quantization storage module is used to quantize the original waveform data, calculate the hash key of the quantized waveform, and use the hash key to store the waveform in the waveform dictionary for deduplication. The waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary.

[0080] The instruction generation module is used to store the waveform identifiers of each original waveform and the absolute start time and corresponding target channel identifier of the waveform on the global time axis using the playback instruction sequence;

[0081] The compression transmission module is used to serialize the waveform dictionary, playback instruction sequence, and original waveform metadata into a byte stream, and transmit the data packet to the quantum measurement and control equipment through compression and encapsulation, so that the quantum measurement and control equipment can recover and obtain the waveform dictionary and playback instruction sequence by deserializing the received data packet.

[0082] Unless otherwise specifically stated, the relative steps, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0083] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0084] The units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations are not considered to be beyond the scope of this invention.

[0085] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiments can be implemented in hardware or as a software functional module. This invention is not limited to any particular combination of hardware and software.

[0086] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A quantum measurement and control waveform data compression method for heterogeneous communication, characterized in that, Include: Read the raw waveform data generated after quantum circuit scheduling, and obtain the waveform identifier, target channel identifier, absolute start time and metadata of each waveform; The original waveform data is quantized, the hash key of the quantized waveform is calculated, and the hash key is used to store the waveform data in the waveform dictionary for deduplication. The waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary. The waveform identifiers of each original waveform, the absolute start time of the waveform on the global time axis, and the corresponding target channel identifier are stored using the playback instruction sequence. The waveform dictionary, playback instruction sequence, and original waveform metadata are serialized into a byte stream and then compressed and encapsulated before being transmitted to the quantum measurement and control equipment. This allows the quantum measurement and control equipment to recover and obtain the waveform dictionary and playback instruction sequence by deserializing the received data packets.

2. The quantum measurement and control waveform data compression method for heterogeneous communication according to claim 1, characterized in that, The raw waveform data generated after reading the quantum circuit scheduling includes: Based on the quantum chip hardware topology and quantum computing task logic circuit, construct a hardware-executable underlying quantum circuit; By scheduling parameters to be assigned to each quantum gate operation in the underlying quantum circuit, a sequence of original pulse instructions is generated.

3. The quantum measurement and control waveform data compression method for heterogeneous communication according to claim 1, characterized in that, Quantization processing of the raw waveform data includes: Extract the real and imaginary parts of the original waveform separately, and determine the normalization scaling factor based on the dynamic range of the waveform; The real and imaginary parts are truncated from the original floating-point range and scaled to an integer range of a specified bit depth.

4. The quantum measurement and control waveform data compression method for heterogeneous communication according to claim 1, characterized in that, Calculate the hash key of the quantized waveform, and use the hash key to perform deduplication storage in the waveform dictionary, including: The hash key of the quantized waveform is calculated using a hash function, and then the hash key is used to look up the waveform in the waveform dictionary. If the search in the waveform dictionary fails, all waveform samples in the waveform dictionary are traversed, the L2 distance between the quantized waveform and the waveform sample is calculated, and a matching sample of the quantized waveform is determined in the waveform dictionary based on the distance threshold. If no matching sample exists in the waveform dictionary, the current quantized model is used as the new waveform sample, and the new waveform sample is added to the waveform dictionary by assigning a new index.

5. The quantum measurement and control waveform data compression method for heterogeneous communication according to claim 1, characterized in that, The original waveform identifier and the absolute start time and position of the waveform on the global timeline are stored using the playback command sequence, including: Based on the timing of each pulse waveform duration, the absolute start time of the pulse waveform in the entire pulse waveform sequence is determined. The entire pulse waveform sequence consists of all pulses arranged on the same time axis when the quantum circuit executes the logic circuit once. Each pulse waveform retains its waveform identifier, absolute start time, and destination channel identifier. All pulse waveforms are arranged in chronological order, and the relevant pulse waveform data is stored in the playback instruction sequence.

6. The quantum measurement and control waveform data compression method for heterogeneous communication according to claim 1, characterized in that, The waveform dictionary, playback command sequence, and original waveform metadata are serialized into a byte stream, and then compressed and encapsulated into data packets for transmission to the quantum measurement and control equipment. This includes: If the quantized waveform is stored as a quantum gate pulse waveform sample after deduplication in the waveform dictionary, then the dictionary index, playback instruction sequence and metadata corresponding to the waveform will be serialized into a byte stream and transmitted to the quantum measurement and control equipment through compression and encapsulation. If, during the deduplication of the quantized waveform in the waveform dictionary, there are special non-quantum gate pulse waveform samples, then the corresponding waveforms will be dynamically incrementally constructed and stored in the waveform dictionary to expand the waveform dictionary. The expanded waveform dictionary, playback instruction sequence, and metadata will be serialized into a byte stream and transmitted to the quantum measurement and control equipment through compression and encapsulation.

7. The quantum measurement and control waveform data compression method for heterogeneous communication according to claim 1 or 6, characterized in that, The quantum measurement and control equipment recovers and obtains the waveform dictionary and playback instruction sequence by deserializing the received data packets, and also includes: Request a pre-allocated memory buffer and initialize it to zero; Based on the waveform identifier and absolute start time, the corresponding waveform sample in the waveform dictionary is copied to the specified location in the buffer to perform waveform reconstruction.

8. A quantum measurement and control waveform data compression system for heterogeneous communication, characterized in that, It includes: a waveform acquisition module, a quantization and storage module, an instruction generation module, and a compression and transmission module. The waveform acquisition module is used to read the raw waveform data generated after quantum circuit scheduling, and to acquire the waveform identifier, target channel identifier, absolute start time and metadata of each waveform; The quantization storage module is used to quantize the original waveform data, calculate the hash key of the quantized waveform, and use the hash key to store the waveform data in the waveform dictionary for deduplication. The waveform dictionary is a pre-constructed quantum gate pulse waveform sample dictionary. The instruction generation module is used to store the waveform identifiers of each original waveform and the absolute start time and corresponding target channel identifier of the waveform on the global time axis using the playback instruction sequence; The compression transmission module is used to serialize the waveform dictionary, playback instruction sequence, and original waveform metadata into a byte stream, and transmit the data packet to the quantum measurement and control equipment through compression and encapsulation, so that the quantum measurement and control equipment can recover and obtain the waveform dictionary and playback instruction sequence by deserializing the received data packet.

9. An electronic device, characterized in that, include: At least one processor, and a memory coupled to said at least one processor; The memory stores a computer program that can be executed by the at least one processor to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed, enables the implementation of the method as described in any one of claims 1 to 7.

Images