Signal decoding method based on PSI5 interface

Abstract

The invention provides a signal decoding method based on a PSI5 interface, and relates to the technical field of signal decoding, and the decoding method comprises the following steps: processing and adjusting a received PSI5 signal to obtain an accurate identification signal; and extracting a clock signal from the PSI5 signal, designing a clock compensation algorithm according to the quantum neural network, and dynamically adjusting the clock signal to obtain clock synchronization data. And determining a data frame boundary, and decoding the accurate identification signal to obtain original data. And performing comparative analysis on the original data and the PSI5 signal to obtain comparative difference data, judging whether the comparative difference data accords with a preset error type or not, and performing corresponding processing. According to the invention, fractional calculus processing is carried out on the received signal, then signal distortion is corrected by using a channel equalization method, and the quantum neural network is constructed to dynamically adjust the clock signal, so that clock synchronization is realized, a reliable time reference is provided for accurate processing of the signal, and the accuracy of signal decoding is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of signal decoding, in particular to a signal decoding method based on a PSI5 interface. BACKGROUND

[0002] In the field of modern electronic communication, especially in application scenarios such as automotive electronics, PSI5 interface is widely used in signal transmission between sensors and electronic control units due to its high efficiency and reliability. PSI5 interface can realize fast and accurate transmission of sensor data, which is of great significance to improve the performance and reliability of the system.

[0003] However, in actual application, signal transmission based on PSI5 interface faces many challenges. First of all, signal distortion problem is more prominent. Due to various interference factors in the transmission process, such as electromagnetic interference, channel noise and transmission line loss, PSI5 signal will be distorted to varying degrees. Traditional signal processing methods often cannot effectively cope with these complex distortion conditions, making it difficult for the receiving end to accurately identify and decode the signal, thereby affecting the normal operation of the system.

[0004] Secondly, clock synchronization is also a key problem. Accurate clock synchronization is the basis for correct signal decoding and processing. In the PSI5 signal transmission system, due to the influence of external environmental changes, device characteristics differences and other factors, the clock signal will deviate, causing the time reference of data sampling and processing to be inconsistent. This will cause errors in signal decoding, seriously affecting the performance and reliability of the system.

[0005] In addition, the lack of error detection and processing capability is also a defect in the existing technology. In the transmission and decoding process of PSI5 signal, various errors such as data errors and transmission errors will inevitably occur. However, the existing signal decoding method often lacks effective error detection mechanism and cannot detect these errors in time. Even if the error can be detected, there is no perfect processing measure to accurately correct the error, which affects the reliability and stability of the system.

[0006] In summary, the existing signal decoding technology based on PSI5 interface has obvious shortcomings in dealing with signal distortion, clock synchronization and error detection and processing, and an effective signal decoding method is needed to solve these problems to improve the performance and reliability of the PSI5 signal transmission system. SUMMARY

[0007] The present application provides a signal decoding method based on a PSI5 interface to solve the defects of signal distortion, clock synchronization and insufficient error detection in the prior art.

[0008] The application provides a signal decoding method based on a PSI5 interface, comprising the following steps: The received PSI5 signal is subjected to fractional calculus processing to obtain a processing signal, and a channel equalization method is used to adjust the processing signal to obtain an accurately recognized signal.

[0009] A clock signal is extracted from the PSI5 signal, a clock compensation algorithm is designed according to a quantum neural network, and the clock signal is dynamically adjusted to obtain clock synchronization data.

[0010] The data frame starting mark and the node number mark in the processing signal are used to determine the data frame boundary, and the accurately recognized signal is decoded according to the data frame boundary to obtain original data.

[0011] The original data is compared with the PSI5 signal according to the clock synchronization data to obtain comparison difference data, and it is judged whether the comparison difference data conforms to a preset error type; if yes, the PSI5 signal is decoded according to the clock compensation algorithm; otherwise, the signal is received again.

[0012] According to the signal decoding method based on the PSI5 interface, the step of processing to obtain the processing signal comprises the following steps: A level conversion circuit is used to convert the level of the PSI5 signal into a preset level range, and the transmission line of the PSI5 signal is subjected to impedance matching.

[0013] The details and edge features of the converted PSI5 signal are enhanced according to fractional differentiation, sampling is performed according to the Nyquist sampling theorem, and the signal is converted into a discrete-time signal.

[0014] In a preset fractional order range, a plurality of different values are selected to process the discrete-time signal to obtain corresponding signal performance indexes, and the optimal order is selected as the minimum mean square error.

[0015] The discrete-time signal corresponding to the optimal order is used as the processing signal.

[0016] According to the signal decoding method based on the PSI5 interface, the step of adjusting to obtain the accurately recognized signal comprises the following steps: The characteristics of the processing signal are determined from the frequency range, noise characteristics and signal amplitude.

[0017] The training sequence is extracted from the processing signal, the discrete-time signal is compared with the training sequence to obtain signal distortion data.

[0018] The equalizer type is selected according to the signal distortion data and the characteristics, the parameters are determined, the training sequence is input, the coefficients of the equalizer are optimized using the recursive least square algorithm to obtain the equalizer optimized coefficients.

[0019] The processing signal is input into the equalizer, and the equalizer optimization coefficient is used for processing to obtain accurate identification signal data.

[0020] According to the signal decoding method based on the PSI5 interface provided by the application, the step of optimizing the equalizer optimization coefficient comprises: Initialize the selected equalizer coefficient vector and covariance matrix.

[0021] Iteratively calculate the gain vector, error, coefficient vector and covariance matrix of each time sample point of the training sequence.

[0022] When the updated error and coefficient vector change is less than a preset threshold, the obtained coefficient vector is the equalizer optimization coefficient.

[0023] According to the signal decoding method based on the PSI5 interface provided by the application, the step of extracting the clock signal comprises: The PSI5 signal is sampled, converted into a data signal, and normalized to obtain a preprocessed signal.

[0024] According to the encoding characteristics of the PSI5 signal, the clock characteristics are detected from the preprocessed signal.

[0025] Based on the clock characteristics, the PSI5 signal is decoded using a Manchester decoding circuit to obtain a clock pulse sequence.

[0026] According to the digital phase-locked loop algorithm, the clock pulse sequence is processed to obtain the clock signal.

[0027] According to the signal decoding method based on the PSI5 interface provided by the application, the step of designing the clock compensation algorithm comprises: Define the requirements and goals of the clock compensation problem, collect the original clock signal, reference clock signal and environment data, and perform feature extraction and uniformization to the same scale range to obtain training data.

[0028] According to the clock characteristics, the number of layers and the number of quantum bits of the quantum neural network are determined, and a full connection structure is used as the connection mode between quantum bits.

[0029] Initialize the weights of the quantum bits and the full connection structure, and define the mean absolute error as the loss function according to the requirements and goals.

[0030] The quantum gradient descent algorithm is used to update the parameters of the quantum neural network, and the corresponding parameters are selected as the building parameters to construct the quantum neural network.

[0031] Use the training data to determine whether the performance of the quantum neural network meets the preset standard, and if so, stop updating, otherwise continue updating.

[0032] The prediction deviation is obtained by using a quantum neural network to predict the deviation of the clock signal, the compensation amount is calculated according to the prediction deviation, and the clock compensation algorithm is obtained by applying the compensation amount to the clock signal.

[0033] According to the signal decoding method based on the PSI5 interface provided by the application, the step of dynamically adjusting to obtain clock synchronization data comprises: The clock adjustment signal is obtained by adjusting the clock signal according to the clock compensation algorithm, and the adjustment feature is obtained by feature extraction.

[0034] The adjustment feature is compared with the clock feature, the synchronization error is calculated, and it is judged whether the synchronization error meets the preset synchronization requirement, if yes, the clock synchronization data is generated, otherwise the compensation amount is recalculated.

[0035] According to the signal decoding method based on the PSI5 interface provided by the application, the step of determining the data frame boundary comprises: According to the format of the processing signal, the section number mark is found from the starting position of the processing signal according to the one-by-one checking strategy.

[0036] According to the value of the section number mark and the fixed length of each section, the end position of the data frame is calculated, so as to determine the data frame boundary.

[0037] According to the signal decoding method based on the PSI5 interface provided by the application, the step of decoding to obtain the original data comprises: According to the data frame boundary, the starting position and the end position of each data frame in the accurate identification signal are found, and the frame header information is obtained by analyzing the frame header part of each data frame according to the length of the frame and the checking mode.

[0038] According to the length of the frame, the effective data is extracted from the accurate identification signal, and the decoding data is obtained by decoding according to the encoding mode of the accurate identification signal.

[0039] According to the checking mode of the frame, the decoding data is checked, and the original data is obtained by recombining according to the preset order.

[0040] According to the signal decoding method based on the PSI5 interface provided by the application, the step of analyzing to obtain the comparison difference data comprises: According to the clock synchronization data, the original data and the PSI5 signal are aligned on the time axis.

[0041] From the starting position after alignment, the data in the original data and the PSI5 signal are compared point by point, and for each corresponding time point, the difference between the original data and the PSI5 signal is calculated according to the type and property of the data to obtain a plurality of groups of difference data.

[0042] The plurality of sets of difference data are integrated, and the position information of the data is recorded to obtain contrast difference data.

[0043] The signal decoding method based on the PSI5 interface provided by the application enhances the detail edge features of the signal by performing fractional calculus processing on the received PSI5 signal, so that the signal is more easily identified. Then, a channel equalization method is used to select a suitable equalizer type according to the signal distortion data and optimize the coefficients thereof, so as to adjust the processed signal to correct the signal distortion, effectively improve the signal quality, reduce the influence of signal distortion on decoding, and make it possible to accurately identify the signal, thereby improving the accuracy of signal decoding. Then, a clock compensation algorithm is designed according to the quantum neural network, the problem requirements are defined, the quantum neural network is constructed, the model is trained, and the compensation amount is calculated according to the prediction deviation, so as to dynamically adjust the clock signal, realize clock synchronization, provide a reliable time reference for accurate processing of the signal, and avoid decoding errors caused by clock asynchronization. The method for determining the data frame boundary by using the data frame start flag and the node number flag can accurately divide the data frame, provide a correct data range for subsequent decoding operations, and ensure the accuracy of the original data extraction. In addition, the contrast difference data are obtained by comparing the original data and the PSI5 signal, and the corresponding processing is performed according to the preset error type, so that errors in the decoding process can be found and corrected in time, the reliability and stability of the signal decoding are improved, and the fault tolerance of the system is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0045] Figure 1 is one of the flowcharts of the signal decoding method based on the PSI5 interface provided by the embodiments of the application; Figure 2 is another flowchart of the signal decoding method based on the PSI5 interface provided by the embodiments of the application; Figure 3 is a third flowchart of the signal decoding method based on the PSI5 interface provided by the embodiments of the application; Figure 4 is a fourth flowchart of the signal decoding method based on the PSI5 interface provided by the embodiments of the application. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0047] The following is combined Figures 1-4 This invention describes a signal decoding method based on the PSI5 interface.

[0048] like Figure 1 As shown, an embodiment of the present invention provides a signal decoding method based on a PSI5 interface, comprising: The received PSI5 signal is processed by fractional calculus to obtain a processed signal, and the channel equalization method is used to adjust the processed signal to obtain an accurate identification signal.

[0049] like Figure 2 As shown, the steps for processing the signal include: A level conversion circuit is used to convert the level of the PSI5 signal to a preset level range and to perform impedance matching on the transmission line of the PSI5 signal.

[0050] Based on the detailed edge features of the PSI5 signal after fractional-order differential enhancement, sampling is performed according to the Nyquist sampling theorem, and the signal is converted into a discrete-time signal, expressed by the following formula:

[0051] In the formula, It is a discrete-time signal, where T is the sampling period and n is... N is the number of sampling points.

[0052] Within a preset fractional order range, multiple different values ​​are selected to process discrete-time signals, yielding corresponding signal performance metrics. The order with the smallest mean square error is then chosen as the optimal order. Performance metrics include signal-to-noise ratio, edge sharpness, and computational complexity (measured by the number of multiplications). Each value has a different impact on the discrete-time signal. By analyzing the signal's spectrum, time-domain characteristics, and other aspects, corresponding signal performance metrics are obtained. Mean square error, as a crucial indicator of signal processing effectiveness, directly reflects the degree of difference between the processed and original signals. Therefore, selecting the order with the smallest mean square error as the optimal order maximizes the preservation of the original signal's characteristics.

[0053] The discrete-time signal corresponding to the optimal order is used as the processing signal.

[0054] As Figure 3 shown, the step of adjusting to obtain the accurate identification signal includes: Determine the characteristics of the processing signal from the frequency range, noise characteristics and signal amplitude. By analyzing the frequency range of the processing signal, we can understand the distribution of different frequency components in the signal, and thus judge the bandwidth and main frequency components of the signal. The analysis of noise characteristics helps to identify various noise types in the signal, such as Gaussian noise, white noise, etc., so as to take appropriate noise reduction measures. The determination of signal amplitude can help us understand the strength and dynamic range of the signal, providing an important basis for subsequent processing.

[0055] Extract the training sequence from the processing signal, and compare and analyze the discrete-time signal with the training sequence to obtain signal distortion data. Signal distortion data can include amplitude attenuation, phase shift, frequency response change, etc. The training sequence is a representative part of the signal selected from the processing signal. By comparing the discrete-time signal with the training sequence, we can accurately find the distortion of the signal in the transmission and processing process. The attenuation of amplitude may be due to the loss of signal in the transmission process or the insufficient gain of the amplifier, the phase shift may be related to the transmission delay or phase modulation of the signal, and the change of frequency response will affect the spectral characteristics of the signal.

[0056] According to the signal distortion data and characteristics, select the type of equalizer and determine the parameters, input the training sequence, and use the recursive least squares algorithm to optimize the coefficients of the equalizer to obtain the optimized coefficients of the equalizer.

[0057] The step of optimizing the equalizer optimization coefficient includes: Initialize the coefficient vector and covariance matrix of the selected equalizer. In the initialization process, the initial values of the coefficient vector and covariance matrix need to be set according to the type of equalizer and the actual application scenario, to ensure that the recursive least squares algorithm can quickly converge and obtain accurate results.

[0058] Iteratively calculate the gain vector, error, coefficient vector and covariance matrix of each time sample point of the training sequence, which is expressed as:

[0059] In the formula, is the gain vector, is the exponential weighting factor, is the training sequence, is the expected output signal, is the error, is the coefficient vector, is the covariance matrix, is the transpose of the training sequence, is the covariance matrix of the last time, is is the transpose of the coefficient vector of the time, is the unit matrix.

[0060] When the updated error and the change of the coefficient vector are less than the preset threshold, the obtained coefficient vector is the equalizer optimization coefficient. By continuously comparing the output of the equalizer and the discrete-time signal, and updating the coefficient according to the error, the performance of the equalizer is gradually optimized. The preset threshold is set to determine whether the recursive least square algorithm converges. When the error and the change of the coefficient vector are less than the preset threshold, it means that the performance of the equalizer has reached a relatively stable state, and the obtained coefficient vector is the optimization coefficient of the equalizer.

[0061] The processing signal is input into the equalizer, and the equalizer optimization coefficient is used for processing to obtain accurate recognition signal data. The equalizer output signal is further suppressed by a limiter (such as a hard limiter or a soft limiter), and the limiter threshold is dynamically adjusted according to the average power of the signal.

[0062] The clock signal is extracted from the PSI5 signal, and a clock compensation algorithm is designed according to the quantum neural network to dynamically adjust the clock signal to obtain clock synchronization data.

[0063] The step of extracting the clock signal includes: The PSI5 signal is sampled to convert it into a data signal, and normalized to obtain a preprocessed signal.

[0064] According to the encoding characteristics of the PSI5 signal, clock features are detected from the preprocessed signal. The clock features can include level transitions of Manchester encoding, rising and falling edges of the signal, frequency and phase of the clock, periodicity of the clock signal, etc.

[0065] Based on the clock features, a Manchester decoding circuit is used to decode the PSI5 signal to obtain a clock pulse sequence. The Manchester decoding circuit can decode the clock information in the PSI5 signal according to the rules of Manchester encoding to obtain a clock pulse sequence.

[0066] The clock pulse sequence is processed according to a digital phase-locked loop algorithm to obtain a clock signal. The digital phase-locked loop algorithm tracks the clock frequency and phase of the PSI5 signal by adjusting internal parameters, and calculates the adjustment amount according to the difference between the clock change point and the internal clock, and then adjusts the internal clock to synchronize with the clock of the input signal, thereby obtaining an accurate clock signal.

[0067] As shown in Figure 4 The steps of designing the clock compensation algorithm include: The requirements and goals of the clock compensation problem are defined, the original clock signal, reference clock signal and environmental data are collected, and the training data is obtained by feature extraction and unification to the same scale range. The original clock signal, reference clock signal and environmental data are collected to fully understand the changes of the clock signal and the influencing factors. Feature extraction is to extract features related to clock compensation from these data, such as changes in clock frequency and phase drift.

[0068] The number of layers and the number of qubits of the quantum neural network are determined according to the clock characteristics, and a fully connected structure is used as the connection method between qubits. The frequency and phase changes of the clock are influenced by a variety of factors, and a multi-layer quantum neural network is needed to capture these complex relationships. When the number of clock characteristics is greater, the number of qubits required is greater.

[0069] The weights of the qubits and the fully connected structure are initialized, and the mean absolute error is defined as the loss function according to the requirements and goals, and the formula is expressed as:

[0070] In the formula, L is the mean absolute error, M is the total number of qubits, i is the index variable of summation, is the true value of the i-th qubit, is the predicted value of the i-th qubit.

[0071] The quantum gradient descent algorithm is used to update the parameters of the quantum neural network, and the corresponding parameters are selected as the building parameters to construct the quantum neural network, and the formula is expressed as:

[0072] In the formula, is the weight of the qubit and the fully connected structure, is the gradient of the loss function at the parameter , is the learning rate, is the updated parameter, is the parameter before updating.

[0073] The performance of the quantum neural network is judged using the training data whether it reaches the preset standard, yes, stop updating, otherwise continue updating. The setting of the preset standard needs to be reasonably selected according to the actual application requirements and the performance characteristics of the quantum neural network. Through continuous training and evaluation of the quantum neural network using training data, the performance of the quantum neural network can be gradually improved until the preset standard is reached.

[0074] The prediction deviation is obtained by predicting the deviation of the clock signal using a quantum neural network, the compensation amount is calculated according to the prediction deviation, and the clock compensation algorithm is obtained by applying the compensation amount to the clock signal, and the formula is expressed as:

[0075] In the formula, is the prediction deviation, is the compensation amount, is the compensation coefficient.

[0076] The step of dynamically adjusting to obtain the clock synchronization data includes: According to the clock compensation algorithm, the clock signal is adjusted to obtain the clock adjustment signal, and the feature extraction is performed to obtain the adjustment feature.

[0077] The adjustment feature is compared with the clock feature, the synchronization error is calculated, and it is judged whether the synchronization error meets the preset synchronization requirement. If yes, the clock synchronization data is generated, otherwise the compensation amount is recalculated. The synchronization error is obtained by comparing the difference between the adjustment feature and the clock feature, and the preset synchronization requirement is an error range set according to the actual application requirement. When the synchronization error meets the preset synchronization requirement, it means that the clock signal has realized synchronization, and at this time the clock synchronization data can be generated.

[0078] According to the data frame start flag and the section number flag in the processing signal, the data frame boundary is determined, and the original data is obtained by decoding the accurately identified signal according to the data frame boundary.

[0079] The step of determining the data frame boundary includes: Starting from the starting position of the processing signal, the data frame start flag is found according to the one-by-one checking strategy, and the section number flag is found according to the format of the processing signal. The data frame start flag is a specific pattern or sequence used to identify the start of the data frame. This flag can be a fixed binary sequence (such as a specific byte combination representing the start flag in some communication protocols), or a signal pattern with specific characteristics (such as a specific level jump combination). When the signal part matching the preset start flag is found, the position is recorded, which is the starting position of the data frame.

[0080] According to the value of the section number flag and the fixed length of each section, the end position of the data frame is calculated, so as to determine the data frame boundary. The section number flag is used to indicate the number of sections contained in the data frame. The section number flag can also be a specific numerical code, such as a byte or several bytes to represent the number of sections. The value of the section number flag is read and parsed to determine the total number of sections in the data frame.

[0081] The step of decoding to obtain the original data includes: According to the data frame boundary, the starting position and the ending position of each data frame are found in the accurate identification signal, and the header part of each data frame is parsed according to the length and the checking mode of the frame to obtain the header information.

[0082] According to the length of the frame, the effective data is extracted from the accurate identification signal, and the decoding data is obtained by decoding according to the encoding mode of the accurate identification signal. The extraction of the effective data needs to be carried out according to the length and the boundary information of the data frame, so as to ensure that the extracted data is complete and accurate. The decoding process needs to be carried out according to the encoding mode of the accurate identification signal, so as to restore the original data from the encoded data.

[0083] According to the checking mode of the frame, the decoding data is checked, and the original data is obtained by recombining according to the preset order. The checking process ensures the accuracy and integrity of the decoding data. By checking the decoding data, errors or losses that may exist in the data can be found. The recombination according to the preset order is to combine the decoding data of each data frame into the original data, so as to restore the original structure and order of the data.

[0084] According to the clock synchronization data, the original data is compared with the PSI5 signal to obtain comparison difference data.

[0085] The step of analyzing the comparison difference data includes: According to the clock synchronization data, the original data and the PSI5 signal are aligned on the time axis.

[0086] Starting from the aligned starting position, the data in the original data and the PSI5 signal are compared point by point, and for each corresponding time point, the difference between the original data and the PSI5 signal is calculated according to the type and property of the data to obtain a plurality of sets of difference data.

[0087] The plurality of sets of difference data are integrated, and the position information of the data is recorded to obtain the comparison difference data. The comparison difference data can include frame synchronization difference, encoding mode difference, data error or tampering, etc.

[0088] It is judged whether the comparison difference data conforms to the preset error type. If yes, the PSI5 signal is decoded according to the clock compensation algorithm, otherwise the signal is received again.

[0089] The signal decoding method based on the PSI5 interface provided in the embodiment enhances the detail edge features of the signal by performing fractional calculus processing on the received PSI5 signal, so that the signal is more easily identified. Then, using a channel equalization method, the appropriate equalizer type is selected according to the signal distortion data, and the coefficients are optimized, and the processed signal is adjusted to solve the distortion problem in the signal transmission process, thereby improving the quality and recognizability of the signal. The clock signal is obtained through steps such as sampling, normalization, clock feature detection, Manchester decoding and digital phase-locked loop algorithm, and then the clock compensation algorithm is designed according to the quantum neural network, the problem requirements are defined, the quantum neural network is constructed, the model is trained, and the compensation amount is calculated according to the prediction deviation, and the clock signal is dynamically adjusted to solve the clock deviation problem, realize clock synchronization, and provide a reliable time reference for correct processing and transmission of the signal. According to the clock synchronization data, the original data and the PSI5 signal are compared and analyzed, the comparison difference data is obtained through time axis alignment, point-by-point comparison and integration of difference data, and then it is judged whether the difference conforms to the preset error type. If it conforms, the PSI5 signal is decoded according to the clock compensation algorithm to solve the error judgment and correction problems that may occur in the signal transmission and processing process, thereby improving the reliability and stability of the entire system.

[0090] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and the necessary general hardware platform, and of course, it can also be realized by hardware. Based on such understanding, the above technical solutions or the essential part of the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium such as ROM / RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the method described in each embodiment or some part of the embodiment.

[0091] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A signal decoding method based on a PSI5 interface, characterized in that, include: The received PSI5 signal is processed by fractional calculus to obtain a processed signal, and the processed signal is adjusted using a channel equalization method to obtain an accurate identification signal. The clock signal is extracted from the PSI5 signal, and a clock compensation algorithm is designed based on a quantum neural network to dynamically adjust the clock signal to obtain clock synchronization data. Based on the data frame start flag and segment number flag in the processed signal, the data frame boundary is determined, and the precise identification signal is decoded according to the data frame boundary to obtain the original data. The original data is compared and analyzed with the PSI5 signal based on the clock synchronization data to obtain comparison difference data. It is then determined whether the comparison difference data conforms to a preset error type. If it does, the PSI5 signal is decoded according to the clock compensation algorithm; otherwise, the signal is received again.

2. The signal decoding method based on the PSI5 interface according to claim 1, characterized in that, The steps for obtaining the processed signal include: A level conversion circuit is used to convert the level of the PSI5 signal to a preset level range, and impedance matching is performed on the transmission line of the PSI5 signal. Based on the detailed edge features of the PSI5 signal after fractional derivative enhancement, sampling is performed according to the Nyquist sampling theorem, and the signal is converted into a discrete-time signal. Within a preset fractional order range, multiple different values ​​are selected to process the discrete-time signal to obtain corresponding signal performance indicators, and the one with the smallest mean square error is selected as the optimal order. The discrete-time signal corresponding to the optimal order is used as the processing signal.

3. The signal decoding method based on the PSI5 interface according to claim 2, characterized in that, The steps for adjusting to obtain the accurate identification signal include: The characteristics of the processed signal are determined from the frequency range, noise characteristics, and signal amplitude. Training sequences are extracted from the processed signal, and the discrete-time signal is compared and analyzed with the training sequences to obtain signal distortion data. Based on the signal distortion data and the characteristics, select the equalizer type, determine the parameters, input the training sequence, and use the recursive least squares algorithm to optimize the equalizer coefficients to obtain the equalizer optimization coefficients; The processed signal is input into the equalizer, and processed according to the equalizer optimization coefficients to obtain the accurate identification signal data.

4. The signal decoding method based on the PSI5 interface according to claim 3, characterized in that, The steps for obtaining the equalizer optimization coefficients include: Initialize the coefficient vector and covariance matrix of the selected equalizer; Iteratively calculate the gain vector, error, coefficient vector, and covariance matrix for each time sample point of the training sequence; When the updated error and coefficient vector change are less than a preset threshold, the resulting coefficient vector is the equalizer optimization coefficient.

5. The signal decoding method based on the PSI5 interface according to claim 1, characterized in that, The steps for extracting the clock signal include: The PSI5 signal is sampled, converted into a data signal, and normalized to obtain a preprocessed signal. Based on the encoding characteristics of the PSI5 signal, clock features are obtained by detecting clock-related features from the preprocessed signal; Based on the clock characteristics, the PSI5 signal is decoded using a Manchester decoding circuit to obtain a clock pulse sequence; The clock signal is obtained by processing the clock pulse sequence according to the digital phase-locked loop algorithm.

6. The signal decoding method based on the PSI5 interface according to claim 5, characterized in that, The steps for designing the clock compensation algorithm include: Define the requirements and objectives of the clock compensation problem, collect raw clock signals, reference clock signals and environmental data, and perform feature extraction and unify them to the same scale range to obtain training data; The number of layers and the number of qubits in the quantum neural network are determined based on the clock characteristics, and a fully connected structure is used as the connection method between the qubits. The weights of the qubits and the fully connected structure are initialized, and the mean absolute error is defined as the loss function according to the requirements and objectives. The parameters of the quantum neural network are updated using the quantum gradient descent algorithm, and the corresponding parameters are selected as the construction parameters to build the quantum neural network. The training data is used to determine whether the performance of the quantum neural network has reached a preset standard. If it has, the update is stopped; otherwise, the update continues. The clock signal deviation is predicted using the quantum neural network to obtain the predicted deviation, and a compensation amount is calculated based on the predicted deviation. The compensation amount is then applied to the clock signal to obtain the clock compensation algorithm.

7. The signal decoding method based on the PSI5 interface according to claim 5, characterized in that, The steps for dynamically adjusting the clock synchronization data include: The clock signal is adjusted according to the clock compensation algorithm to obtain a clock adjustment signal, and the adjustment features are extracted to obtain the adjustment features. The adjustment feature is compared with the clock feature to calculate the synchronization error, and it is determined whether the synchronization error meets the preset synchronization requirements. If yes, the clock synchronization data is generated; otherwise, the compensation amount is recalculated.

8. The signal decoding method based on the PSI5 interface according to claim 7, characterized in that, The steps for determining the data frame boundaries include: Starting from the beginning position of the processing signal, the data frame start flag is found according to the check-by-check strategy, and the segment number flag is found according to the format of the processing signal. Based on the value of the segment number flag and the fixed length of each segment, the end position of the data frame is calculated, thereby determining the boundary of the data frame.

9. The signal decoding method based on the PSI5 interface according to claim 1, characterized in that, The steps for decoding to obtain the original data include: Based on the data frame boundaries, the start and end positions of each data frame are found in the precise identification signal, and the frame header of each data frame is parsed according to the frame length and verification method to obtain the frame header information; Valid data is extracted from the precise identification signal according to the length of the frame, and decoded according to the encoding method of the precise identification signal to obtain decoded data; The decoded data is verified according to the frame verification method, and then reassembled in a preset order to obtain the original data.

10. A signal decoding method based on a PSI5 interface according to claim 1, characterized in that, The steps for analyzing and obtaining the comparative difference data include: Based on the clock synchronization data, align the original data and the PSI5 signal on the time axis; Starting from the aligned starting position, the original data and the data in the PSI5 signal are compared point by point. For each corresponding time point, the difference between the original data and the PSI5 signal is calculated according to the data type and nature to obtain multiple sets of difference data. The comparative difference data is obtained by integrating multiple sets of difference data and recording the location information of the data.

Images