Signal integrity analysis method and device and electronic equipment

Abstract

The invention provides a signal integrity analysis method and device and electronic equipment, and the method comprises the steps: collecting signal data, and converting the signal data into corresponding quantum state data; key features in the quantum state data are extracted through a linear discriminant analysis algorithm; performing iterative training on a quantum machine learning model by using the key feature value to obtain a quantum channel model; the quantum channel model is used for representing the physical limitation and evolution law of the signal in the process from sending to receiving; and inputting a signal to be analyzed into the quantum channel model to obtain a signal integrity analysis result. According to the invention, accurate quantum state calculation analysis is carried out on a large amount of high-speed and high-frequency signal data through a quantum calculation technology, and the influence of noise, signal attenuation, coherence weakening and other factors on signal transmission is effectively solved.

Description

Technical Field

[0001] The present disclosure relates to the field of signal processing, and in particular to a signal integrity analysis method, device and electronic equipment. Background Art

[0002] In computing devices, signal integrity analysis is the key to improving device performance, reducing signal interference, and enhancing user experience. In related technologies, signal integrity analysis methods are mostly based on linear system theory and digital signal processing technology, such as eye diagram method, Fourier transform method, etc. However, these methods have certain limitations when processing high-speed, high-frequency complex signals, especially advanced interface signals such as PCIE4.0. Because the amount of high-speed and high-frequency signal data is huge, traditional processing methods cannot effectively achieve real-time analysis and processing. Secondly, with the increase in chip and device density, noise interference, signal attenuation and other problems have become more serious, and traditional signal integrity analysis methods cannot accurately analyze and predict signals. Summary of the invention

[0003] The present application provides a signal integrity analysis method, device and electronic device to at least solve the above technical problems existing in the prior art.

[0004] According to a first aspect of the present application, a signal integrity analysis method is provided, the method comprising:

[0005] Collecting signal data, and converting the signal data into corresponding quantum state data;

[0006] Using a linear discriminant analysis algorithm to extract key features from the quantum state data;

[0007] Iteratively training the quantum machine learning model using the key eigenvalues ​​to obtain a quantum channel model; the quantum channel model is used to characterize the physical limitations and evolution laws of the signal from sending to receiving;

[0008] The signal to be analyzed is input into the quantum channel model to obtain a signal integrity analysis result.

[0009] In one possible implementation, it further includes:

[0010] Based on the signal integrity analysis result, the signal to be analyzed is optimized to obtain an optimized signal.

[0011] In one possible implementation, converting the signal data into corresponding quantum state data includes:

[0012] Standardizing the signal data using a matrix equation calculation method to obtain standardized data;

[0013] The standardized data is mapped onto a quantum state to obtain a quantum state of at least one quantum bit.

[0014] In one possible implementation, the extracting key features from the quantum state data using a linear discriminant analysis algorithm includes:

[0015] Classifying the quantum state data to obtain multiple categories of quantum state data;

[0016] Calculate the first mean vector of samples in each class, calculate the first deviation of each sample in the class to the first mean vector, and transpose the first deviation to obtain an intra-class scatter matrix;

[0017] Calculate the second mean vector of all samples, calculate the second deviation from the first mean vector to the second mean vector in each category, and transpose the second deviation to obtain an inter-class scatter matrix;

[0018] Determining a target projection vector based on the intra-class scatter matrix and the inter-class scatter matrix; the target projection vector is used to characterize intra-class targets and inter-class targets;

[0019] The sample is projected based on the target projection vector to obtain low-dimensional key features.

[0020] In one possible implementation, the quantum machine learning model includes:

[0021] A quantum error correction mechanism is used to encode the quantum state data using an error correction code to obtain error-corrected quantum state data

[0022] Quantum control mechanisms for managing quantum states.

[0023] In one possible implementation manner, the iterative training of the quantum machine learning model using the key eigenvalue to obtain the quantum channel model includes:

[0024] Inputting the key feature value into the quantum machine learning model to obtain the prediction accuracy;

[0025] Determine a loss value of a loss function of the quantum machine learning model based on the prediction accuracy;

[0026] Determining a target update gradient based on a loss value of the loss function;

[0027] Based on the target update gradient, the model parameters of the quantum machine learning model are updated, and the training is performed to a preset convergence condition;

[0028] The quantum machine learning model trained to the preset convergence condition is used as the output quantum channel model.

[0029] In one possible implementation manner, the acquisition signal data includes:

[0030] Using an appropriate network analyzer to collect signal data of signal parameters;

[0031] The signal parameters include: at least one of signal strength, noise level, signal loss, bit error rate, impedance value and operating temperature.

[0032] In one possible implementation manner, the optimizing the signal to be analyzed based on the signal integrity analysis result to obtain an optimized signal includes:

[0033] When there is signal loss in the signal to be analyzed under the target working condition, the signal to be analyzed is optimized according to the corresponding processing rules under the target working condition until the optimization target is reached, and then the optimized signal is output.

[0034] According to a second aspect of the present application, a signal integrity analysis device is provided, the device comprising:

[0035] An acquisition module, used to acquire signal data and convert the signal data into corresponding quantum state data;

[0036] An extraction module, used to extract key features from the quantum state data using a linear discriminant analysis algorithm;

[0037] A training module, used to iteratively train the quantum machine learning model using the key eigenvalues ​​to obtain a quantum channel model; the quantum channel model is used to characterize the physical limitations and evolution laws of the signal from sending to receiving;

[0038] The analysis module is used to input the signal to be analyzed into the quantum channel model to obtain a signal integrity analysis result.

[0039] According to a third aspect of the present application, an electronic device is provided, including:

[0040] at least one processor; and

[0041] a memory communicatively connected to the at least one processor; wherein,

[0042] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the present application.

[0043] According to a fourth aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the present application.

[0044] According to a fifth aspect of the present application, a computer program product is provided, comprising a computer program or instructions, which implement the method described in the present application when executed by a processor.

[0045] By utilizing the technical solution of the present application, it is possible to perform accurate quantum state calculation and analysis on a large amount of high-speed, high-frequency signal data through quantum computing technology, effectively solving the impact of noise, signal attenuation, coherence reduction and other factors on signal transmission.

[0046] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which:

[0048] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.

[0049] Figure 1 The implementation process of the signal integrity analysis method in the embodiment of the present application is shown in FIG. Figure 1 ;

[0050] Figure 2 The implementation process of the signal integrity analysis method in the embodiment of the present application is shown in FIG. Figure 2 ;

[0051] Figure 3 A schematic diagram of the structure of a signal integrity analysis device in an embodiment of the present application is shown;

[0052] Figure 4 A schematic diagram of the structure of an electronic device in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0053] In order to make the purpose, features, and advantages of the present disclosure more obvious and easy to understand, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0054] A signal integrity analysis method, device and electronic device provided by the present application are described below in conjunction with the accompanying drawings.

[0055] like Figure 1 As shown, an embodiment of the present application provides a signal integrity analysis method, the method comprising:

[0056] S101, collecting signal data, and converting the signal data into corresponding quantum state data;

[0057] It is understandable that the signal data in this application can be a notebook high-speed serial computer expansion bus standard (Peripheral Component Interconnect Express, PCIE) 4.0 signal, which exists in the form of a slot on the computer motherboard and is used to plug in various devices that need to communicate with the computer motherboard at high speed, such as graphics cards, hard drives, SSDs, wireless network cards, sound cards, etc. Currently, most electronic products are using PCIE 4.0, and its data transmission speed per Lan is 16GT / s, which is sufficient for current consumer electronic products.

[0058] In some optional embodiments, the collected signal data includes:

[0059] Using an appropriate network analyzer to collect signal data of signal parameters;

[0060] The signal parameters include: at least one of signal strength, noise level, signal loss, bit error rate, impedance value and operating temperature.

[0061] It is understandable that the signal data in this application is measured by a vector network analyzer for PCIe4.0 signals, and the frequency response, phase, group delay and other parameters of the signal are monitored in real time under different operating conditions (such as temperature, voltage, power, etc.). The signal data obtained from each measurement is recorded. This process may require long-term and continuous measurements to ensure that the amount of data collected is large and covers various possible working conditions.

[0062] It is understandable that when collecting PCIe4.0 signal data, it is usually necessary to pay attention to and collect the following types of data:

[0063] Signal strength: The strength of the signal will directly affect the signal transmission quality and the recognition effect of the receiving end, so the signal strength information is very important.

[0064] Noise level: Noise can also seriously affect the quality of the signal, so it is necessary to pay attention to changes in the noise level.

[0065] Channel loss: The degree to which a channel attenuates a signal is critical to signal integrity, so data related to channel loss needs to be collected.

[0066] Bit Error Rate (BER): Bit error rate is an important indicator for measuring signal integrity and needs to be collected.

[0067] Impedance value: Impedance value affects the transmission and reception of signals and needs to be taken into consideration.

[0068] Operating Temperature: The temperature conditions of the operating environment can affect device performance and signal integrity.

[0069] These data need to be measured and recorded in real time under different operating conditions by a vector network analyzer with corresponding functions. Using these data as input for training models can improve the prediction accuracy and performance of the models.

[0070] Then, the present application converts the signal data into corresponding quantum state data through quantum processing technology. It can be understood that each data point is regarded as a quantum bit, and the quantum superposition state is used to allow the data point to exist in multiple states at the same time. Quantum state definition and preprocessing is to convert traditional data into a form that can be processed by a quantum computer, that is, a quantum state. For an n-qubit (quantum bit) quantum system, it is usually represented in a 2n-dimensional complex vector space.

[0071] S102, extracting key features from the quantum state data using a linear discriminant analysis algorithm;

[0072] S103, iteratively training the quantum machine learning model using the key eigenvalues ​​to obtain a quantum channel model; the quantum channel model is used to characterize the physical limitations and evolution laws of the signal from sending to receiving;

[0073] S104, inputting the signal to be analyzed into the quantum channel model to obtain a signal integrity analysis result.

[0074] The signal integrity analysis method provided in the present application collects signals and converts the signals into quantum state data, extracts key features in the quantum state data through a linear discriminant analysis algorithm, and then uses the key features to iteratively train a quantum machine learning model to obtain a quantum channel model, thereby using the quantum channel model to analyze signal integrity.

[0075] In some optional embodiments, the signal integrity analysis method provided by the present application further includes:

[0076] Based on the signal integrity analysis result, the signal to be analyzed is optimized to obtain an optimized signal.

[0077] In the present application, after obtaining the signal integrity analysis result, the signal to be analyzed can be optimized according to the analysis result, thereby obtaining an optimized signal. The model can be used to predict the PCIe4.0 signal class and improve the signal integrity. Since the model can predict the integrity of the PCIe4.0 signal under different working conditions, the integrity of the PCIe4.0 signal can be improved by changing the working conditions (such as adjusting the power supply voltage, changing the operating temperature, etc.).

[0078] In some optional embodiments, converting the signal data into corresponding quantum state data comprises:

[0079] Standardizing the signal data using a matrix equation calculation method to obtain standardized data;

[0080] The standardized data is mapped onto a quantum state to obtain a quantum state of at least one quantum bit.

[0081] As a specific implementation, in PCIe4.0 signal analysis, quantum state processing can be performed according to the following steps:

[0082] First, the signal data is standardized, specifically, the collected signal data is standardized. With the help of matrix calculation methods, various signal parameters, such as signal strength, noise level, channel loss, bit error rate, impedance value, etc., are mapped between 0 and 1. This allows the data to adapt to the needs of quantum computing, because quantum states are usually operated in a unit spherical space (Bloch sphere), which requires the input data to be normalized.

[0083] Then the quantum state is assigned, specifically for each standardized data point, it is mapped to a quantum state. In quantum computing, the state of a single quantum bit can be represented by a complex number, the Pearson relation:

[0084] |ψ〉=α|0〉+β|1〉

[0085] Among them, α and β are complex numbers and need to satisfy

[0086] |α| 2 +|β| 2 =1

[0087] Finally, quantum superposition state generation, specifically, after defining the quantum state, can use the quantum superposition property to generate and process multiple possible signal states at the same time. If there are n quantum bits, 2n states that exist at the same time can be generated. This will greatly increase the efficiency of signal data processing and analysis.

[0088] like Figure 2As shown, in some optional embodiments, the linear discriminant analysis algorithm is used to extract key features in the quantum state data, including:

[0089] S201, classifying the quantum state data to obtain multiple categories of quantum state data;

[0090] S202, calculating a first mean vector of samples in each class, and calculating a first deviation of each sample in the class to the first mean vector, and performing transposition processing on the first deviation to obtain an intra-class scatter matrix;

[0091] S203, calculating the second mean vector of all samples, and calculating the second deviation from the first mean vector to the second mean vector in each category, and transposing the second deviation to obtain an inter-class scatter matrix;

[0092] S204, determining a target projection vector based on the intra-class scatter matrix and the inter-class scatter matrix; the target projection vector is used to represent the intra-class target and the inter-class target;

[0093] S205, projecting the sample based on the target projection vector to obtain low-dimensional key features.

[0094] Specifically, because the signal data of the present application contains factors such as signal strength, noise level, channel loss, impedance, etc. With these characteristics, a feature vector can be constructed and LDA (Linear Discriminant Analysis) can be used to extract feature values. The main goal of Linear Discriminant Analysis (LDA) is to make the samples within a class as close as possible and the samples between classes as far apart as possible.

[0095] After classifying the quantum state data, the linear discriminant analysis algorithm is used to extract the key features of the two-day data, as follows:

[0096] Calculate the intra-class scatter matrix: For each sample of category i, calculate its mean vector m i Then calculate the deviation of each sample to its mean vector and multiply it by its transpose to get the scatter matrix of each sample. Add the scatter matrices of all samples in a category to get the intra-class scatter matrix S w .

[0097] S w =Σ i Σ {x∈Xi} (xm i )(xm i ) T

[0098] Among them, X iRepresents all samples of category i, and x is one of them.

[0099] Calculate the inter-class scatter matrix: First calculate the mean vector m of all samples. Then for each class, calculate the deviation of its mean vector to the mean vector of all samples, multiply it by its transpose, and get a scatter matrix. Add the scatter matrices of all classes to get the inter-class scatter matrix S b :

[0100] S b =Σ i |X i |(m i -m)(m i -m) T

[0101] Among them, |X i | represents the number of samples of category i.

[0102] Solve the optimization problem: The goal is to find a projection vector w so that the projected samples can meet the goal of compactness within the class and dispersion between classes as much as possible. Therefore, the following optimization problem needs to be solved:

[0103]

[0104] The solution to this optimization problem is The eigenvectors corresponding to the largest eigenvalues ​​of S are W*, which is the optimal projection vector to be found, so that the projected samples can meet the goal of compactness within the class and dispersion between classes as much as possible; S b is the inter-class scatter matrix, which is used to represent the overall scatter between classes; S w is the intra-class scatter matrix, which is used to represent the scatter within each class. The number of them is determined by the dimension to be reduced, arg max W Compute the standard function for the quantity that maximizes the score.

[0105] Finally, W* is used to project the sample to obtain new features.

[0106] The linear discriminant analysis algorithm can be used to extract the most discriminative low-dimensional features from the original high-dimensional features for subsequent model training.

[0107] In some optional embodiments, the quantum machine learning model includes:

[0108] A quantum error correction mechanism, used to encode the quantum state data using an error correction code to obtain error-corrected quantum state data;

[0109] Quantum control mechanisms for managing quantum states.

[0110] It is understandable that the present application utilizes superquantum quantum error correction codes and quantum control to ensure signal integrity during model training, data storage and reading. During the training process, due to the non-idealization of the system and external disturbances, problems such as bit flipping or loss of coherence may occur. In response to these problems, it is necessary to introduce quantum error correction and quantum control mechanisms into the system. For quantum error correction, the main approach is to encode each bit using an error correction code. Commonly used quantum error correction codes include quantum bit flip codes and quantum phase flip codes. These codes can correct errors such as bit flips and phase flips.

[0111] Quantum error correction and quantum control processes are key steps to ensure information fidelity during training and actual operation. In the process of converting signal data into quantum state data, after the data is defined as quantum state and quantum superposition state, these quantum states can be used in the subsequent quantum computing process. At this stage, error correction coding can be introduced to prevent quantum bit errors caused by noise and other environmental factors. In the process of building a quantum channel model, quantum characteristics are taken into consideration in the model structure, and appropriate quantum control operations are introduced to simulate and adapt to actual channel conditions, such as channel attenuation, noise and other factors.

[0112] The purpose of quantum error correction is to correct the flipped bits, while quantum control is to properly adjust and control the interference in the channel so that information can be correctly transmitted in the channel, thereby ensuring the integrity and fidelity of the data during the entire quantum computing process. The following is a detailed implementation process:

[0113] Quantum error correction: In order to protect quantum information in a complex environment, its encoding is called a special state of quantum error correction code. Among them, the most basic are 3-bit quantum bit flip code and quantum phase flip code. The following is a detailed implementation process:

[0114] 3-bit quantum bit flip code:

[0115] Original quantum state: |ψ〉=α|0〉+β|1〉

[0116] Encoded quantum state: |ψ〉=α|000〉+β|111〉

[0117] If the original quantum state encounters a bit flip during transmission (for example, |0〉 flips to |1〉), the erroneous bit can be easily found and corrected during the decoding process (generally through the majority decision method).

[0118] Quantum phase-flip codes:

[0119] Original quantum state: |ψ〉=α|0〉+β|1〉

[0120] Encoded quantum state: |ψ〉=α|+++〉+β|---〉

[0121] This is a transposition operation of the bit-flip code, where bit flipping corresponds to the Z-gate operation and phase flipping corresponds to the X-gate operation. In this way, the quantum state can better resist bit errors caused by environmental noise to ensure the fidelity of information.

[0122] Quantum control: used to control and manage the state of quantum systems to achieve precise quantum operations, including state preparation, quantum gate operations, and measurements.

[0123] In some embodiments, the iterative training of the quantum machine learning model using the key eigenvalue to obtain the quantum channel model includes:

[0124] Inputting the key feature value into the quantum machine learning model to obtain the prediction accuracy;

[0125] Determine a loss value of a loss function of the quantum machine learning model based on the prediction accuracy;

[0126] Determining a target update gradient based on a loss value of the loss function;

[0127] Based on the target update gradient, the model parameters of the quantum machine learning model are updated, and the training is performed to a preset convergence condition;

[0128] The quantum machine learning model trained to the preset convergence condition is used as the output quantum channel model.

[0129] Specifically, in this application, the extracted features are used to train the quantum channel model using a quantum machine learning algorithm, specifically:

[0130] The extracted features are used to train a quantum channel model that can predict the integrity of PCIe4.0 signals under specific working conditions. The basic idea of ​​training is to operate a series of known quantum states, observe the final output results, and then train based on the feedback signal to optimize the process. A model based on quantum neural network (QNN) is used. The details are as follows:

[0131] (1) Quantum neural network construction: Regarding quantum neural networks, their main structure consists of quantum gates, quantum bits (qubits), and parameterized rotation gates. The structure (for example, how many layers to use, which gates to use, how to connect, etc.) and the required parameters need to be determined in advance. During the construction process, a loss function also needs to be selected, and the mean square error loss function is used to measure the gap between the predicted value and the label.

[0132] (2) Quantum state preparation: First, the initial state needs to be prepared based on the result of feature extraction. Usually, a vector is used to represent a quantum state, and the preparation of this state can be completed through some revolving gates and Bloch spheres.

[0133] (3) Parameterized revolving gate operation: After the quantum state is prepared, the quantum state is optimized by rotating the newly added parameters of each qubit. Each revolving gate has a parameter, which is continuously updated during the learning process.

[0134] (4) Measurement and prediction: After a series of revolving gate operations, a predicted value is obtained through measurement. A Z-basis measurement is performed on a specific qubit, and the loss is calculated using the measurement result (i.e., the predicted value).

[0135] (5) Parameter update: Through the feedback of measurement results and loss function, the gradient descent optimization algorithm can be used to gradually reduce the loss value and update the revolving door parameters. In quantum computers, this step can be completed by parameter gradient calculation methods, such as parameter shift methods, which obtain gradient estimates by changing a small amount of parameters.

[0136] (6) Repeated iteration: The above process is repeated until a satisfactory prediction effect is obtained or the preset number of iterations is reached.

[0137] The quantum channel model simulates the transmission and evolution of signals in the channel.

[0138] In the process of building a quantum channel model, the following factors that have an important impact on PCIe4.0 signal transmission are usually considered:

[0139] Weak coherence: Due to factors such as noise and signal instability, the coherence of the signal will be weakened during transmission. This can be simulated by introducing corresponding coherence operations.

[0140] Attenuation: Signals may attenuate due to various reasons during transmission, which requires the introduction of an attenuation operator to simulate.

[0141] Noise: Noise is unavoidable in actual transmission and can be simulated by introducing a noise operator.

[0142] In quantum theory, the state of a quantum system is generally represented by a quantum state matrix ρ. A quantum channel E can be regarded as a mapping acting on ρ and is defined by the following formula:

[0143]

[0144] Among them, E k is the Kraus operator, satisfying the normalization condition I is the unit matrix. E(ρ) represents the state of a quantum signal after passing through a quantum channel; ρ is the initial state of a quantum channel, usually represented by a density matrix, Represents the Kraus operator E k The conjugate transpose of , k: refers to the index of the different paths leading through the channel.

[0145] In some optional embodiments, the step of optimizing the signal to be analyzed based on the signal integrity analysis result to obtain an optimized signal includes:

[0146] When there is signal loss in the signal to be analyzed under the target working condition, the signal to be analyzed is optimized according to the corresponding processing rules under the target working condition until the optimization target is reached, and then the optimized signal is output.

[0147] In this application, after the quantum channel model training is completed, the quantum channel model can be used to predict the PCIe4.0 signal class and improve the signal integrity. Since the quantum channel model can predict the integrity of the PCIe4.0 signal under different working conditions, the integrity of the PCIe4.0 signal can be improved by changing the working conditions (such as adjusting the power supply voltage, changing the operating temperature, etc.). The main process of signal integrity improvement includes problem prediction, problem identification and optimization improvement, as follows:

[0148] Problem prediction: Using the trained quantum machine learning model, the integrity of the PCIe4.0 signal under various working conditions is predicted. For example, the degree of loss of the PCIe4.0 signal under a certain power supply voltage or operating temperature can be predicted. In the quantum machine learning model, the output will be a "prediction value" that can give the possible degree of loss of the PCIe4.0 signal.

[0149] Problem identification: Identify possible problems by comparing the predicted results with the ideal state. For example, if it is predicted that under certain working conditions, the signal loss exceeds the tolerance range, then it is necessary to find the cause of this problem. The specific way to find the cause is through the prediction results of the quantum machine learning model. For example, if the model predicts that under certain working conditions, the signal loss exceeds the tolerance range, then it can be determined that there is a problem. The sources of specific problems are multifaceted, such as poor working conditions or hardware problems.

[0150] To determine whether the problem is caused by poor working conditions, the model can be used to predict whether the problem is caused by poor working conditions by changing the working conditions (such as adjusting the power supply voltage or operating temperature). If these changes can significantly improve signal integrity, then it can be considered that the problem may be caused by the working conditions.

[0151] To determine whether the problem is caused by hardware problems, you can check the status of the hardware devices, such as whether the impedance of the transmission line is matched, whether there are signal reflection problems, etc. If hardware problems are found, and after these problems are fixed, the signal integrity predicted by the model is improved, then it can be considered that the problem may be caused by the hardware.

[0152] Optimization and improvement: Based on the identified problems, the signal transmission process is improved by changing the working conditions or optimizing the hardware to improve the signal integrity. A signal integrity standard or threshold is set first. If the predicted loss exceeds this threshold, it is considered that there is a signal integrity problem. The basis for identifying the problem is mainly based on the characteristics of the PCIe4.0 signal and the real-time conditions in the working environment, such as at least one of the signal strength, noise level, channel loss, bit error rate, impedance value and operating temperature.

[0153] Here are the problem identification methods:

[0154] If the predicted signal loss is mainly caused by the power supply voltage, then the problem is that the power supply voltage is unstable or not within a suitable range, and the power supply voltage needs to be adjusted.

[0155] If the predicted signal loss is mainly caused by the operating temperature, then the problem is that the operating environment is too hot or too cold, and it is necessary to improve the heat dissipation system or adjust the temperature of the operating environment.

[0156] If the predicted signal loss is mainly caused by channel loss, then the problem is impedance mismatch of the transmission line or there is a signal reflection problem, and the design of the transmission line needs to be optimized to improve impedance matching.

[0157] If the model predicts a high bit error rate, the possible problem is that the signal is subject to significant interference during transmission or there are performance issues with the hardware equipment. Improvements can be made by increasing the channel protection level or optimizing the hardware equipment.

[0158] If the predicted signal loss is mainly caused by the noise level, then the problem is that the noise suppression capability of the device is insufficient, and noise reduction strategies can be adopted, such as using better noise suppression technology, hardware device isolation, etc.

[0159] Through the above problem identification, further targeted optimization and improvement can be carried out. For example, if the problem lies in the power supply voltage, then it is necessary to adjust the power supply voltage or replace a more stable power supply device; if the problem lies in the operating temperature, then it is necessary to improve the cooling system or change the working environment of the equipment; if the problem lies in the channel loss, then it is necessary to optimize the line design; if the problem lies in the noise level, then it is necessary to adopt a better noise suppression strategy, etc.

[0160] For example, the working environment can be improved by adjusting the power supply voltage, optimizing the cooling system, etc., or improving the hardware design, such as optimizing the design of the transmission line, improving impedance matching, etc. The prediction function is denoted as F(V,T), where V is the power supply voltage and T is the operating temperature. V and T can be changed to

[0161] min V,TF(V,T)

[0162] Find the V and T that minimize F to obtain the optimal working environment.

[0163] After the improvement, the quantum machine learning model can be used again to predict the signal integrity to verify the effect of the optimization improvement. Subsequently, the problem prediction and identification can be continued, and the optimization and improvement can be repeated again, and the cycle can be repeated until the set signal integrity target is achieved or no further improvement can be achieved.

[0164] The signal integrity analysis method provided in this application collects PCIE4.0 signal data through a vector network analyzer, and preprocesses the data into quantum states for analysis. Based on these data, a quantum channel model of the signal is established to describe the physical limitations and evolution laws of the signal from sending to receiving, and extract key features from it. Superquantum quantum error correction codes and quantum control are used to ensure signal integrity during model training, data storage, and reading. On this basis, a quantum machine learning algorithm is used to train the model, and the model prediction results are used to find possible problems. Then, based on this information, the signal transmission process is optimized and improved to improve signal integrity.

[0165] This application uses quantum computing technology to perform accurate quantum state calculation and analysis on a large amount of high-speed, high-frequency signal data, effectively solving the impact of noise, signal attenuation, and reduced coherence on signal transmission. By learning the characteristics and laws of processing signals through machine learning technology, the signal integrity can be accurately and quickly predicted and optimized, thereby improving the integrity of the signal. Not only does it improve the stability of the signal, it also greatly reduces errors and losses, and enhances the real-time adjustment and optimization capabilities of the equipment, thereby significantly improving the user experience. The technical solution provided in this application performs AI digital transformation in signal integrity analysis, and can also be applied to other high-speed serial protocols, with strong versatility and practical value.

[0166] like Figure 3 As shown, an embodiment of the present application provides a signal integrity analysis device, the device comprising:

[0167] The acquisition module 301 is used to acquire signal data and convert the signal data into corresponding quantum state data;

[0168] An extraction module 302 is used to extract key features from the quantum state data using a linear discriminant analysis algorithm;

[0169] A training module 303 is used to iteratively train the quantum machine learning model using the key eigenvalues ​​to obtain a quantum channel model; the quantum channel model is used to characterize the physical limitations and evolution laws of the signal from sending to receiving;

[0170] The analysis module 304 is used to input the signal to be analyzed into the quantum channel model to obtain a signal integrity analysis result.

[0171] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.

[0172] Figure 4 A schematic block diagram of an example electronic device 400 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0173] like Figure 4 As shown, the device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0174] A number of components in the device 400 are connected to the I / O interface 405, including: an input unit 406, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the device 400 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0175] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the signal integrity analysis method. For example, in some embodiments, the signal integrity analysis method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the signal integrity analysis method described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the signal integrity analysis method in any other appropriate manner (e.g., by means of firmware).

[0176] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0177] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0178] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0179] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0180] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0181] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0182] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0183] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0184] The above is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present disclosure, which should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A signal integrity analysis method, characterized in that: The method comprises: Collecting signal data, and converting the signal data into corresponding quantum state data; Using a linear discriminant analysis algorithm to extract key features from the quantum state data; Iteratively training the quantum machine learning model using the key eigenvalues ​​to obtain a quantum channel model; the quantum channel model is used to characterize the physical limitations and evolution laws of the signal from sending to receiving; The signal to be analyzed is input into the quantum channel model to obtain a signal integrity analysis result.

2. The method according to claim 1, further comprising: Based on the signal integrity analysis result, the signal to be analyzed is optimized to obtain an optimized signal.

3. The method according to claim 1, wherein converting the signal data into corresponding quantum state data comprises: Standardizing the signal data using a matrix equation calculation method to obtain standardized data; The standardized data is mapped onto a quantum state to obtain a quantum state of at least one quantum bit.

4. The method according to claim 1, wherein the step of extracting key features from the quantum state data using a linear discriminant analysis algorithm comprises: Classifying the quantum state data to obtain multiple categories of quantum state data; Calculate the first mean vector of samples in each class, calculate the first deviation of each sample in the class to the first mean vector, and transpose the first deviation to obtain an intra-class scatter matrix; Calculate the second mean vector of all samples, calculate the second deviation from the first mean vector to the second mean vector in each category, and transpose the second deviation to obtain an inter-class scatter matrix; Determining a target projection vector based on the intra-class scatter matrix and the inter-class scatter matrix; The target projection vector is used to represent the intra-class target and the inter-class target; The sample is projected based on the target projection vector to obtain low-dimensional key features.

5. The method according to claim 1, wherein the quantum machine learning model comprises: A quantum error correction mechanism, used to encode the quantum state data using an error correction code to obtain error-corrected quantum state data; Quantum control mechanisms for managing quantum states.

6. The method according to claim 1, wherein the key eigenvalues ​​are used to iteratively train the quantum machine learning model to obtain the quantum channel model, comprising: Inputting the key feature value into the quantum machine learning model to obtain the prediction accuracy; Determine a loss value of a loss function of the quantum machine learning model based on the prediction accuracy; Determining a target update gradient based on a loss value of the loss function; Based on the target update gradient, the model parameters of the quantum machine learning model are updated, and the training is performed to a preset convergence condition; The quantum machine learning model trained to the preset convergence condition is used as the output quantum channel model.

7. The method according to claim 2, wherein collecting signal data comprises: Using an appropriate network analyzer to collect signal data of signal parameters; The signal parameters include: at least one of signal strength, noise level, signal loss, bit error rate, impedance value and operating temperature.

8. The method according to claim 2, wherein optimizing the signal to be analyzed based on the signal integrity analysis result to obtain an optimized signal comprises: When there is signal loss in the signal to be analyzed under the target working condition, the signal to be analyzed is optimized according to the corresponding processing rules under the target working condition until the optimization target is reached, and then the optimized signal is output.

9. A signal integrity analysis device, the device comprising: An acquisition module, used to acquire signal data and convert the signal data into corresponding quantum state data; An extraction module, used to extract key features from the quantum state data using a linear discriminant analysis algorithm; A training module, used to iteratively train the quantum machine learning model using the key eigenvalues ​​to obtain a quantum channel model; The quantum channel model is used to characterize the physical limitations and evolution laws of signals from transmission to reception; The analysis module is used to input the signal to be analyzed into the quantum channel model to obtain a signal integrity analysis result.

10. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

Images