Component high-frequency signal transmission quality prediction and optimization method and system

Abstract

The application provides a component high-frequency signal transmission quality prediction and optimization method and system, relates to the technical field of signal transmission, and comprises the following steps: based on the frequency domain characteristics of a component to-be-transmitted signal and transmission path physical characteristic parameters, extracting the attenuation change rate of each frequency band, and establishing an associated function with the frequency band boundary position; calculating the transmission path terminal predicted loss spectrum and determining the compensation gain coefficient; performing frequency-selective amplitude adjustment on the to-be-transmitted signal; collecting actual received signals and target frequency domain amplitude distribution comparison, and updating the associated function. The application realizes accurate prediction and adaptive optimization of high-frequency signal transmission quality, and improves signal transmission integrity.

Description

Technical Field

[0001] This invention relates to the field of signal transmission technology, and in particular to a method and system for predicting and optimizing the high-frequency signal transmission quality of components. Background Technology

[0002] With the increasing integration of electronic devices and the rapid development of communication technologies, the quality of high-frequency signal transmission between components has become a key factor affecting the overall system performance. High-frequency signals are affected by the physical characteristics of the transmission path during transmission, such as impedance mismatch, dielectric loss, and radiation loss, leading to signal attenuation, distortion, and phase delay, thus affecting signal integrity and reliability. Especially in applications such as high-speed data transmission, radio frequency communication, and radar systems, the quality of signal transmission directly affects the system's performance and stability.

[0003] Traditional methods for predicting the quality of high-frequency signal transmission mainly rely on empirical formulas or simplified models, making it difficult to accurately describe the attenuation characteristics of signals under complex transmission environments. Furthermore, existing signal optimization techniques often employ a uniform compensation strategy, failing to fine-tune for the differentiated attenuation characteristics of different frequency components. In addition, traditional methods are typically static compensation, lacking the ability to dynamically evaluate and adaptively optimize actual transmission performance. Summary of the Invention

[0004] The present invention provides a method and system for predicting and optimizing the high-frequency signal transmission quality of electronic components, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for predicting and optimizing the high-frequency signal transmission quality of components, comprising:

[0006] Based on the frequency domain characteristics of the signal to be transmitted from the components and the physical characteristic parameters of the transmission path, the attenuation change rate in the transmission path is extracted for each frequency band, and a correlation function between the attenuation change rate and the frequency band boundary position is established according to the physical characteristic parameters.

[0007] Based on the correlation function, the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path is calculated to obtain the predicted loss spectrum; according to the predicted loss spectrum, the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum is determined by inverse mapping operation;

[0008] Based on the compensation gain coefficient, the signal to be transmitted is frequency-selectively amplitude-adjusted at the transmission start point, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient, resulting in a compensated signal;

[0009] The compensated signal is injected into the transmission path for transmission, and the actual received signal after transmission is collected at the terminal of the transmission path.

[0010] The frequency domain amplitude distribution of the actual received signal is compared with the preset target frequency domain amplitude distribution band by band, the amplitude deviation value of each frequency band is calculated, the correction amount for the function parameters in the correlation function is generated based on the amplitude deviation value, and the correlation function is updated with the correction amount.

[0011] For each frequency band, the attenuation change rate in the transmission path is extracted, and a correlation function between the attenuation change rate and the frequency band boundary position is established based on the physical characteristic parameters, including:

[0012] Based on the distribution density of frequency components in the frequency domain features, the frequency domain features are divided into multiple continuous frequency bands, each frequency band having a start boundary frequency and an end boundary frequency;

[0013] For each frequency band, based on the dielectric loss factor and transmission path length in the physical characteristic parameters, the attenuation gradient of the signal amplitude as a function of frequency within that frequency band is calculated, and the attenuation gradient is used as the attenuation rate of that frequency band.

[0014] Using the starting and ending boundary frequencies of all frequency bands as independent variables and the attenuation change rate of the corresponding frequency bands as dependent variables, a correlation function describing the mapping relationship between the frequency band boundary positions and the attenuation change rate is established through fitting operations.

[0015] Based on the correlation function, the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path is calculated to obtain the predicted loss spectrum; according to the predicted loss spectrum, the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum is determined by inverse mapping operation, including:

[0016] Frequency values ​​of all discrete frequency components are extracted from the frequency domain characteristics of the signal to be transmitted to form a frequency component set. For each frequency value in the frequency component set, the frequency value is substituted into the correlation function, and the attenuation change rate of the frequency band to which the frequency value belongs is determined by the correlation function. The predicted attenuation amount corresponding to the frequency value is calculated based on the attenuation change rate and the physical characteristic parameters of the transmission path.

[0017] The frequency values ​​in the frequency component set are correlated and mapped with their corresponding predicted attenuation values ​​to form a predicted loss spectrum. Based on the predicted attenuation value range of each frequency component in the predicted loss spectrum, the frequency components in the predicted loss spectrum are divided into multiple attenuation levels, and each attenuation level contains frequency components whose predicted attenuation values ​​are in the same range.

[0018] For each attenuation level, based on the center value of the predicted attenuation value range corresponding to that attenuation level, the reference compensation gain coefficient of that attenuation level is determined through inverse mapping operation, and this reference compensation gain coefficient is assigned to all frequency components in that attenuation level as their respective compensation gain coefficients;

[0019] All frequency components and their corresponding compensation gain coefficients are associated and stored to form a set of compensation gain coefficients.

[0020] For each attenuation level, based on the center value of the predicted attenuation range corresponding to that attenuation level, the reference compensation gain coefficient for that attenuation level is determined through inverse mapping operations, including:

[0021] The arithmetic mean of the upper and lower bounds of the predicted attenuation value interval for each attenuation level is taken as the center value, and the difference between the upper and lower bounds is taken as the interval width; the ratio of the center value to the interval width is taken as the relative dispersion parameter.

[0022] Obtain the initial signal strength value at the beginning of the transmission path, and the target signal strength value that should be achieved after transmission compensation;

[0023] Based on the relative dispersion parameter, the compensation tolerance coefficient corresponding to the attenuation level is calculated through a monotonically decreasing mapping relationship. The compensation tolerance coefficient represents the maximum allowable proportion of the signal strength after compensation deviating from the target strength value within the attenuation level. The compensation tolerance coefficient is used as the benchmark compensation gain coefficient.

[0024] Based on the compensation gain coefficient, the signal to be transmitted is frequency-selectively amplitude-adjusted at the transmission start point, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient, resulting in a compensated signal including:

[0025] All discrete frequency components are identified from the signal to be transmitted, and the frequency value and initial amplitude value of each frequency component are extracted; for each frequency component, the corresponding compensation gain coefficient is found in the compensation gain coefficient set according to the frequency value of that frequency component;

[0026] Calculate the total energy of the signal to be transmitted in the frequency domain, determine the maximum allowable total energy threshold based on the power limitation requirements at the transmission starting point, and calculate the energy margin between the total energy and the maximum total energy threshold;

[0027] For each frequency component, the initial amplitude value of the frequency component is multiplied by its corresponding compensation gain coefficient to obtain a preliminary adjustment amplitude value. Based on the energy margin, the preliminary adjustment amplitude values ​​of all frequency components are normalized and scaled to obtain the adjusted amplitude values ​​of each frequency component that satisfy the power limit.

[0028] Keeping the frequency and phase information of each frequency component unchanged, the initial amplitude value of each frequency component is replaced with the corresponding adjusted amplitude value, and the adjusted frequency domain signal is reconstructed; the adjusted frequency domain signal is then subjected to inverse time-domain transformation to convert the adjusted frequency domain signal from frequency domain representation back to time domain representation to obtain the compensated signal.

[0029] Generating a correction amount for the function parameters in the correlation function based on the magnitude deviation value, and updating the correlation function with the correction amount includes:

[0030] Multiple validation sample points are obtained. A preset perturbation increment is applied to each function parameter in the correlation function. The change in the output value of the correlation function at all validation sample points before and after applying the perturbation increment is calculated. The ratio of the change in the output value to the perturbation increment is calculated to obtain the deviation sensitivity coefficient corresponding to the function parameter.

[0031] Based on the magnitude deviation value, the gradient vector of the output value of the correlation function relative to the function parameters is calculated. The signs of each component in the gradient vector are statistically analyzed to determine the dominant sign of the negative gradient direction. Based on the dominant sign, the correction direction indicator of the function parameters is determined.

[0032] The deviation sensitivity coefficient is multiplied by the current value of the function parameter to obtain the basic correction magnitude of the function parameter; the basic correction magnitude is then assigned a sign value according to the correction direction indicator to obtain the correction amount of the function parameter.

[0033] A second aspect of the present invention provides a system for predicting and optimizing the high-frequency signal transmission quality of components, comprising:

[0034] The first unit is used to extract the attenuation change rate of each frequency band in the transmission path based on the frequency domain characteristics of the signal to be transmitted from the acquired components and the physical characteristic parameters of the transmission path, and to establish a correlation function between the attenuation change rate and the frequency band boundary position based on the physical characteristic parameters;

[0035] The second unit is used to calculate the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path based on the correlation function, thereby obtaining a predicted loss spectrum; and to determine the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum through an inverse mapping operation based on the predicted loss spectrum.

[0036] The third unit is used to perform frequency-selective amplitude adjustment on the signal to be transmitted at the transmission start point according to the compensation gain coefficient, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient to obtain the compensated signal;

[0037] The fourth unit is used to inject the compensated signal into the transmission path for transmission, and to collect the actual received signal after the compensated signal has been transmitted at the end of the transmission path;

[0038] The fifth unit is used to compare the frequency domain amplitude distribution of the actual received signal with the preset target frequency domain amplitude distribution band by band, calculate the amplitude deviation value of each frequency band, generate a correction amount for the function parameters in the correlation function based on the amplitude deviation value, and update the correlation function with the correction amount.

[0039] A third aspect of the embodiments of the present invention,

[0040] An electronic device is provided, comprising:

[0041] processor;

[0042] Memory used to store processor-executable instructions;

[0043] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0044] Fourth aspect of the embodiments of the present invention,

[0045] A computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0046] The beneficial effects of this application are as follows:

[0047] This invention establishes a correlation function between the attenuation rate of the component transmission path and the position of the frequency band boundary, thereby achieving accurate prediction of the transmission quality of high-frequency signals and effectively solving the technical problem that traditional methods are unable to cope with complex frequency response characteristics.

[0048] This invention uses inverse mapping operation to determine the compensation gain coefficient and performs frequency-selective amplitude adjustment on the signal to be transmitted, so that the signal can obtain a more balanced spectrum distribution at the transmission terminal, which significantly improves the transmission quality and integrity of high-frequency signals.

[0049] This invention designs a closed-loop optimization mechanism that dynamically updates the correlation function parameters by comparing the actual received signal with the target frequency domain amplitude distribution, enabling the system to adaptively optimize the transmission compensation strategy, thereby enhancing the robustness and adaptability of the method and making it suitable for high-frequency signal transmission scenarios in various complex circuit environments. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating the method for predicting and optimizing the high-frequency signal transmission quality of components according to an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0053] Figure 1 This is a flowchart illustrating the method for predicting and optimizing the high-frequency signal transmission quality of components according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0054] Based on the frequency domain characteristics of the signal to be transmitted from the components and the physical characteristic parameters of the transmission path, the attenuation change rate in the transmission path is extracted for each frequency band, and a correlation function between the attenuation change rate and the frequency band boundary position is established according to the physical characteristic parameters.

[0055] Based on the correlation function, the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path is calculated to obtain the predicted loss spectrum; according to the predicted loss spectrum, the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum is determined by inverse mapping operation;

[0056] Based on the compensation gain coefficient, the signal to be transmitted is frequency-selectively amplitude-adjusted at the transmission start point, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient, resulting in a compensated signal;

[0057] The compensated signal is injected into the transmission path for transmission, and the actual received signal after transmission is collected at the terminal of the transmission path.

[0058] The frequency domain amplitude distribution of the actual received signal is compared with the preset target frequency domain amplitude distribution band by band, the amplitude deviation value of each frequency band is calculated, the correction amount for the function parameters in the correlation function is generated based on the amplitude deviation value, and the correlation function is updated with the correction amount.

[0059] In one optional implementation, extracting the attenuation change rate of each frequency band in the transmission path and establishing a correlation function between the attenuation change rate and the frequency band boundary position based on the physical characteristic parameters includes:

[0060] Based on the distribution density of frequency components in the frequency domain features, the frequency domain features are divided into multiple continuous frequency bands, each frequency band having a start boundary frequency and an end boundary frequency;

[0061] For each frequency band, based on the dielectric loss factor and transmission path length in the physical characteristic parameters, the attenuation gradient of the signal amplitude as a function of frequency within that frequency band is calculated, and the attenuation gradient is used as the attenuation rate of that frequency band.

[0062] Using the starting and ending boundary frequencies of all frequency bands as independent variables and the attenuation change rate of the corresponding frequency bands as dependent variables, a correlation function describing the mapping relationship between the frequency band boundary positions and the attenuation change rate is established through fitting operations.

[0063] The received transmitted signal undergoes preprocessing, including noise reduction, filtering, and signal enhancement. For example, for a transmitted signal with a frequency range of 100MHz to 5GHz, wavelet transform is used to remove background noise, improving the signal-to-noise ratio by at least 20dB to ensure the accuracy of subsequent analysis.

[0064] After preprocessing, spectrum analysis is performed to convert the time-domain signal into a frequency-domain representation. Specifically, a Fast Fourier Transform algorithm is used to process a signal with a sampling rate of 10 GHz and 8192 sampling points, obtaining a spectrum with a frequency resolution of 1.22 MHz. This step clearly presents the energy distribution of the signal at different frequency points, laying the foundation for subsequent frequency band allocation.

[0065] Based on the distribution density of frequency components in the frequency domain, an adaptive threshold method is used to divide the spectrum into multiple continuous frequency bands. Specifically, the energy gradient of the spectrum is calculated, and when the energy difference between adjacent frequency points exceeds a preset threshold (e.g., 3dB), it is marked as a potential frequency band boundary. To ensure the accuracy of the division, a minimum band width constraint (e.g., 50MHz) is introduced to avoid excessive frequency band fragmentation. In a typical case, the spectrum from 100MHz to 5GHz is divided into 6 frequency bands: 100-500MHz, 500-1200MHz, 1200-2100MHz, 2100-3000MHz, 3000-4200MHz, and 4200-5000MHz, each with a clearly defined start and end boundary frequency.

[0066] For each defined frequency band, the attenuation change rate is calculated based on known physical characteristic parameters. For example, for an environment with a transmission path length of 100 meters and a mixed medium, the dielectric loss factor for this specific environment is extracted from the physical parameter database. Taking the 500-1200MHz frequency band as an example, the dielectric loss factor in this environment is 0.05dB / m / MHz. By analyzing the trend of signal amplitude variation with frequency within this frequency band, the attenuation gradient is calculated to be 0.048dB / MHz, and this value is taken as the attenuation change rate for this frequency band.

[0067] After obtaining the boundary locations and corresponding attenuation change rates for all frequency bands, a correlation function is established between them. The starting and ending boundary frequencies of all frequency bands are used as independent variables, with boundary frequencies of adjacent bands repeating; the attenuation change rate of the corresponding frequency band is used as the dependent variable. A piecewise polynomial fitting method is employed to construct a correlation function describing the mapping between the frequency band boundary locations and the attenuation change rate. During the fitting process, a continuity constraint is applied at the frequency band boundaries to ensure a smooth transition of the curve. The final correlation function can predict the attenuation change rate for any given frequency.

[0068] Based on the established correlation function, the transmitted signal can be adaptively processed. For example, for multi-carrier signals that need to be transmitted in the above transmission path, the attenuation of different frequency components can be predicted according to the correlation function, and the signal can be pre-distorted to enhance those frequency components that will experience greater attenuation at the transmitting end.

[0069] Furthermore, correlation functions can be applied to channel state estimation. By periodically transmitting known probe signals, the real-time attenuation rate is calculated and compared with the predicted value of the correlation function. When the deviation exceeds a preset threshold (e.g., 0.01 dB / MHz), the channel state update process is triggered. In dynamic environment testing, channel state re-estimation can be completed within 150 milliseconds after environmental changes, demonstrating more than 50% better adaptability than traditional fixed-parameter models.

[0070] The above implementation method establishes a correlation function between the frequency band boundary position and the attenuation change rate by accurately dividing the frequency band and extracting the attenuation characteristics, which provides an effective means for signal transmission optimization and channel state estimation, and improves performance in complex transmission environments.

[0071] In one optional implementation, based on the correlation function, the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path is calculated to obtain the predicted loss spectrum; according to the predicted loss spectrum, the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum is determined by inverse mapping operation, including:

[0072] Frequency values ​​of all discrete frequency components are extracted from the frequency domain characteristics of the signal to be transmitted to form a frequency component set. For each frequency value in the frequency component set, the frequency value is substituted into the correlation function, and the attenuation change rate of the frequency band to which the frequency value belongs is determined by the correlation function. The predicted attenuation amount corresponding to the frequency value is calculated based on the attenuation change rate and the physical characteristic parameters of the transmission path.

[0073] The frequency values ​​in the frequency component set are correlated and mapped with their corresponding predicted attenuation values ​​to form a predicted loss spectrum. Based on the predicted attenuation value range of each frequency component in the predicted loss spectrum, the frequency components in the predicted loss spectrum are divided into multiple attenuation levels, and each attenuation level contains frequency components whose predicted attenuation values ​​are in the same range.

[0074] For each attenuation level, based on the center value of the predicted attenuation value range corresponding to that attenuation level, the reference compensation gain coefficient of that attenuation level is determined through inverse mapping operation, and this reference compensation gain coefficient is assigned to all frequency components in that attenuation level as their respective compensation gain coefficients;

[0075] All frequency components and their corresponding compensation gain coefficients are associated and stored to form a set of compensation gain coefficients.

[0076] After the transmitted signal is sampled, it is transformed to the frequency domain using a Fast Fourier Transform (FFT) to obtain the signal's frequency domain characteristics. From these characteristics, all discrete frequency components with significant energy are extracted; these frequency values ​​constitute a frequency component set. For example, for an electronic signal with a bandwidth of 20 kHz, an 8192-point FFT can extract multiple main frequency components ranging from 100 Hz to 19 kHz, forming a frequency component set containing hundreds of frequency values.

[0077] For each frequency value in the set of frequency components, it is substituted into a pre-established correlation function for calculation. This correlation function describes the attenuation characteristics of signals in different frequency bands as they travel through a specific transmission path. For example, for a copper transmission cable, its correlation function might indicate that in the 1kHz to 5kHz frequency band, the attenuation rate increases by 0.5dB for every 1kHz increase; in the 5kHz to 10kHz frequency band, the attenuation rate increases by 0.8dB for every 1kHz increase; and above 10kHz, the attenuation rate increases by 1.2dB for every 1kHz increase.

[0078] After determining the attenuation rate of the frequency band to which the frequency component belongs using the correlation function, the specific predicted attenuation is calculated by combining this with the physical characteristic parameters of the transmission path. These physical characteristic parameters include the transmission medium type, transmission distance, and temperature coefficient.

[0079] A one-to-one mapping relationship is established between all frequency values ​​in the frequency component set and their corresponding predicted attenuation values ​​to form a predicted loss spectrum. This loss spectrum fully describes the attenuation of the signal at each frequency point after it passes through the transmission path. For example, the predicted loss spectrum may contain the following mapping pairs: (1kHz, 0.5dB), (3kHz, 1.5dB), (7kHz, 5.6dB), (12kHz, 13.4dB), (18kHz, 22.6dB), etc.

[0080] Based on the predicted attenuation distribution of frequency components in the predicted loss spectrum, all frequency components are divided into multiple attenuation levels. This division can be based on preset numerical range boundaries, for example, into five attenuation levels: slight attenuation (0-3dB), low attenuation (3-8dB), moderate attenuation (8-15dB), high attenuation (15-25dB), and extremely high attenuation (>25dB). Thus, the frequency components in the above example will be assigned to different attenuation levels: 1kHz and 3kHz belong to the slight attenuation level, 7kHz to the low attenuation level, 12kHz to the moderate attenuation level, and 18kHz to the high attenuation level.

[0081] For each attenuation level, the center value of the predicted attenuation range corresponding to that level is determined. For example, the center value for low attenuation levels (3-8dB) is 5.5dB. The attenuation is converted into a compensation gain coefficient through an inverse mapping operation. The inverse mapping operation can be simply understood as taking the negative of the attenuation as the gain coefficient, but in actual implementation, adjustments are made to account for nonlinear characteristics. For low attenuation levels, the baseline compensation gain coefficient might be +6dB (slightly higher than the attenuation center value of 5.5dB to ensure sufficient compensation).

[0082] This baseline compensation gain factor is assigned to all frequency components within the attenuation level. For the 7kHz frequency component in the low attenuation level, although its predicted attenuation is 5.6dB, the baseline compensation gain factor for that level +6dB is uniformly applied, instead of the exact 5.6dB. This layered processing method simplifies the compensation process while ensuring the compensation effect.

[0083] Finally, all frequency components and their corresponding compensation gain coefficients are associated and stored to form a set of compensation gain coefficients. For example, this set may contain mapping pairs such as (1kHz, +2dB), (3kHz, +2dB), (7kHz, +6dB), (12kHz, +12dB), and (18kHz, +20dB). These compensation gain coefficients will be used to pre-compensate the original signal to be transmitted.

[0084] In practical applications, the set of compensation gain coefficients can be implemented as a lookup table and stored in memory. Before signal transmission, the signal to be transmitted is decomposed into its frequency components. The corresponding gain coefficient is applied to each frequency component according to the lookup table, and then the components are recombined to form the pre-compensated signal. For example, for a composite signal containing three main frequency components of 1kHz, 7kHz, and 18kHz, gain compensation of +2dB, +6dB, and +20dB will be applied to these three frequency components respectively. This ensures that the energy of each frequency component tends to be consistent after the signal passes through the transmission path, effectively overcoming the frequency-selective attenuation caused by the transmission path.

[0085] In one optional implementation, for each attenuation level, based on the center value of the predicted attenuation value range corresponding to that attenuation level, the reference compensation gain coefficient for that attenuation level is determined through an inverse mapping operation, including:

[0086] The arithmetic mean of the upper and lower bounds of the predicted attenuation value interval for each attenuation level is taken as the center value, and the difference between the upper and lower bounds is taken as the interval width; the ratio of the center value to the interval width is taken as the relative dispersion parameter.

[0087] Obtain the initial signal strength value at the beginning of the transmission path, and the target signal strength value that should be achieved after transmission compensation;

[0088] Based on the relative dispersion parameter, the compensation tolerance coefficient corresponding to the attenuation level is calculated through a monotonically decreasing mapping relationship. The compensation tolerance coefficient represents the maximum allowable proportion of the signal strength after compensation deviating from the target strength value within the attenuation level. The compensation tolerance coefficient is used as the benchmark compensation gain coefficient.

[0089] This embodiment provides a signal compensation method based on attenuation levels, applicable to adaptive compensation in various signal transmissions. For each attenuation level, this method determines the reference compensation gain coefficient through inverse mapping based on the center value of the predicted attenuation range, achieving refined compensation control.

[0090] In the specific implementation process, the attenuation levels are classified and divided. For example, the predicted attenuation can be divided into multiple attenuation levels according to a gradient of 3dB / level, such as the first level corresponding to 0-3dB, the second level corresponding to 3-6dB, the third level corresponding to 6-9dB, and so on. To calculate the reference compensation gain coefficient for each attenuation level, the following processing steps need to be performed.

[0091] For each attenuation level, calculate the center value of the predicted attenuation range for that level. Taking the second level as an example, the predicted attenuation range for this level is 3-6 dB. Calculate the arithmetic mean of the upper bound of 6 dB and the lower bound of 3 dB, resulting in a center value of 4.5 dB. Simultaneously, calculate the range width, which is the difference between the upper and lower bounds, in this example, 3 dB.

[0092] After obtaining the center value and interval width, the relative dispersion parameter is calculated, which is the ratio of the center value to the interval width. Taking the second level as an example, the relative dispersion parameter is 4.5dB divided by 3dB, resulting in 1.5. This parameter characterizes the relative concentration of predicted values ​​within this attenuation level; a larger value indicates that the predicted attenuation is more significant relative to the interval width.

[0093] It is also necessary to obtain the initial and target signal strength values ​​at the start of the transmission path. Assuming that in a certain wireless transmission, the initial transmit power is 20dBm and the target signal strength at the receiver is -75dBm, these two parameters will serve as the basis for subsequent compensation calculations.

[0094] Based on the calculated relative dispersion parameter, the compensation tolerance coefficient is calculated through a monotonically decreasing mapping relationship. This mapping relationship is designed such that the larger the relative dispersion parameter, the smaller the corresponding compensation tolerance coefficient, reflecting the principle that the more significant the predicted attenuation, the smaller the allowable compensation error should be. In specific implementations, piecewise linear mapping or exponentially decreasing mapping methods can be used.

[0095] Taking the exponentially decreasing mapping as an example, the compensation tolerance coefficient can be expressed as a negative exponential function of the relative dispersion parameter. When the relative dispersion parameter of the second level is 1.5, its compensation tolerance coefficient can be calculated as 0.223, indicating that within this attenuation level, the maximum allowable deviation of the compensated signal strength from the target strength is 22.3%.

[0096] In practical applications, the compensation tolerance coefficient is directly used as the baseline compensation gain coefficient for fine-tuning subsequent signal strength compensation. For example, for a signal with a predicted attenuation of 5dB (belonging to the second level), the baseline compensation gain coefficient is 0.223, meaning that the compensated signal strength is allowed to fluctuate within ±22.3% of the target strength value. If the target signal strength is -75dBm, then the compensated signal strength is allowed to be between -91.725dBm and -58.275dBm.

[0097] Applying the exponentially decreasing mapping relationship, the compensation tolerance coefficients (i.e., the reference compensation gain coefficients) for the five levels are calculated to be 0.607, 0.223, 0.082, 0.030, and 0.011, respectively. It can be seen that as the attenuation level increases, the reference compensation gain coefficient exhibits a clear decreasing trend, reflecting the requirement for higher compensation accuracy under high attenuation conditions.

[0098] When applying this method in signal transmission, the initial signal strength is set to 20dBm, and the target received strength is -75dBm. When the signal is detected at the third level (predicted attenuation of approximately 8dB), the reference compensation gain coefficient of 0.082 for that level is automatically applied for compensation control. Without considering other losses, the compensated signal strength will be controlled within a range of ±8.2% of -75dBm, i.e., between -81.15dBm and -68.85dBm, achieving precise compensation for a specific attenuation level.

[0099] The advantage of this method lies in its ability to dynamically adjust the compensation accuracy requirements based on the characteristics of different attenuation levels, thus avoiding the limitations of traditional fixed compensation schemes. By introducing a relative dispersion parameter and a compensation tolerance coefficient, adaptive matching between the compensation strategy and attenuation characteristics is achieved, making it particularly suitable for complex transmission environments with significant attenuation variations.

[0100] In one optional implementation, the signal to be transmitted is frequency-selectively amplitude-adjusted at the transmission start point according to the compensation gain coefficient, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient, resulting in a compensated signal including:

[0101] All discrete frequency components are identified from the signal to be transmitted, and the frequency value and initial amplitude value of each frequency component are extracted; for each frequency component, the corresponding compensation gain coefficient is found in the compensation gain coefficient set according to the frequency value of that frequency component;

[0102] Calculate the total energy of the signal to be transmitted in the frequency domain, determine the maximum allowable total energy threshold based on the power limitation requirements at the transmission starting point, and calculate the energy margin between the total energy and the maximum total energy threshold;

[0103] For each frequency component, the initial amplitude value of the frequency component is multiplied by its corresponding compensation gain coefficient to obtain a preliminary adjustment amplitude value. Based on the energy margin, the preliminary adjustment amplitude values ​​of all frequency components are normalized and scaled to obtain the adjusted amplitude values ​​of each frequency component that satisfy the power limit.

[0104] Keeping the frequency and phase information of each frequency component unchanged, the initial amplitude value of each frequency component is replaced with the corresponding adjusted amplitude value, and the adjusted frequency domain signal is reconstructed; the adjusted frequency domain signal is then subjected to inverse time-domain transformation to convert the adjusted frequency domain signal from frequency domain representation back to time domain representation to obtain the compensated signal.

[0105] The system receives the signal to be transmitted, which can be any digital or analog signal that needs to be transmitted through a specific transmission medium (such as optical fiber, coaxial cable, etc.). It performs frequency domain analysis on the received signal, converting the time-domain signal into a frequency-domain representation using methods such as Fast Fourier Transform (FFT), and identifies all discrete frequency components. For each identified frequency component, its frequency value and initial amplitude value are extracted and stored in a frequency-amplitude correspondence table. For example, when the signal to be transmitted contains components with frequencies of 200Hz, 500Hz, 1kHz, 5kHz, and 10kHz, the frequency values ​​and corresponding initial amplitude values ​​of these components are extracted respectively, assuming the extracted initial amplitude values ​​are 0.5V, 0.8V, 1.0V, 0.7V, and 0.3V.

[0106] For each frequency component, the corresponding compensation gain coefficient is searched in a pre-established set of compensation gain coefficients based on its frequency value. This set of compensation gain coefficients is predetermined based on the frequency response characteristics of the transmission path, with different compensation gain values ​​corresponding to different frequencies. For example, for the five frequency components mentioned earlier, the corresponding compensation gain coefficients found in the set are 1.1, 1.3, 1.5, 2.0, and 2.5, respectively, reflecting the characteristic that high-frequency components experience more severe attenuation during transmission.

[0107] Calculate the total energy of the signal to be transmitted in the frequency domain. The total energy is calculated by summing the energies of all frequency components, where the energy of each frequency component is equal to the square of its amplitude. For each frequency component, multiply its initial amplitude value by its corresponding compensation gain coefficient to obtain the preliminary adjusted amplitude value.

[0108] To meet power limitation requirements, the initial adjustment amplitude values ​​of all frequency components are normalized and scaled. The scaling factor is calculated by dividing the maximum total energy threshold by the square root of the initial adjusted total energy, and then multiplying the initial adjustment amplitude values ​​of all frequency components by this scaling factor to obtain the adjusted amplitude values ​​that meet the power limitation.

[0109] Keeping the frequency and phase information of each frequency component unchanged, only the initial amplitude value of each frequency component is replaced with the calculated adjusted amplitude value to reconstruct the adjusted frequency domain signal. The adjusted frequency domain signal contains all frequency components that have undergone frequency selective compensation, with high-frequency components receiving a larger compensation gain and low-frequency components receiving a smaller compensation gain.

[0110] Finally, the adjusted frequency domain signal undergoes an inverse time-domain transform to convert it back from its frequency domain representation to its time domain representation, yielding the compensated signal. The inverse time-domain transform can be implemented using methods such as the inverse fast Fourier transform (IFFT). The transformed time-domain signal is the compensated signal to be transmitted. This signal has undergone frequency-selective amplitude adjustment at the transmission start point, with high-frequency components appropriately amplified to compensate for possible frequency-selective attenuation during transmission.

[0111] After the above processing, the compensated signal is output and transmitted to the transmission medium. Since the signal has been pre-compensated according to the expected transmission characteristics, when the signal reaches the receiving end through the transmission path, each frequency component undergoes varying degrees of attenuation, resulting in a more balanced amplitude distribution and improved signal integrity and reliability. In practical applications, the compensation gain coefficient can be dynamically adjusted based on real-time monitoring of the transmission path characteristics to adapt to changing transmission environments.

[0112] In one optional implementation, generating a correction amount for the function parameters in the correlation function based on the magnitude deviation value, and updating the correlation function with the correction amount includes:

[0113] Multiple validation sample points are obtained. A preset perturbation increment is applied to each function parameter in the correlation function. The change in the output value of the correlation function at all validation sample points before and after applying the perturbation increment is calculated. The ratio of the change in the output value to the perturbation increment is calculated to obtain the deviation sensitivity coefficient corresponding to the function parameter.

[0114] Based on the magnitude deviation value, the gradient vector of the output value of the correlation function relative to the function parameters is calculated. The signs of each component in the gradient vector are statistically analyzed to determine the dominant sign of the negative gradient direction. Based on the dominant sign, the correction direction indicator of the function parameters is determined.

[0115] The deviation sensitivity coefficient is multiplied by the current value of the function parameter to obtain the basic correction magnitude of the function parameter; the basic correction magnitude is then assigned a sign value according to the correction direction indicator to obtain the correction amount of the function parameter.

[0116] This embodiment discloses a specific implementation method for generating correction values ​​for function parameters in an correlation function based on the magnitude deviation value and updating the correlation function. In practical applications, various factors may cause deviations between the output value and the expected value of the correlation function, necessitating adjustments to the function parameters to improve the accuracy of the correlation function.

[0117] In the implementation process, multiple validation sample points are first acquired. These sample points can come from actual measurement data or historical records. For example, in a predictive model of the relationship between temperature and pressure, 10 different temperature points (such as 20℃, 25℃, 30℃, etc.) can be selected as validation sample points. For each parameter in the correlation function, a preset perturbation increment is applied for sensitivity analysis. For parameter 'a', a perturbation increment of 0.01 can be applied, making it a + 0.01; for parameter 'b', a perturbation increment of 0.005 can be applied, making it b + 0.005. Then, the change in the output value of the correlation function at all validation sample points before and after applying the perturbation is calculated. For example, if the output of a certain validation point is 5.42 under the original parameter values, and the output becomes 5.48 after applying the perturbation, then the change in output is 0.06.

[0118] For each function parameter, the bias sensitivity coefficient is obtained by comparing the magnitude of the output value change with the corresponding perturbation increment. Continuing the example above, the bias sensitivity coefficient for parameter 'a' is 0.06 ÷ 0.01 = 6, meaning that when parameter 'a' changes by 1 unit, the output value changes by an average of 6 units. This calculation needs to be performed on all validation sample points, and the final average value is taken as the bias sensitivity coefficient for that parameter.

[0119] Based on the known magnitude deviation value, the gradient vector of the correlation function's output value relative to its parameters is calculated. The magnitude deviation value is the difference between the current output and the expected output of the correlation function, which can be obtained through actual measurement. During the gradient calculation process, for each validation sample point, an approximate value of the partial derivative of the output deviation with respect to the function parameters is calculated.

[0120] Perform sign statistics on each component of the gradient vector to determine the dominant sign of the negative gradient direction. Based on the dominant sign, determine the correction direction indicator for the function parameter. If the dominant sign is negative, the correction direction indicator is "+1", indicating that the parameter value needs to be increased; if the dominant sign is positive, the correction direction indicator is "-1", indicating that the parameter value needs to be decreased.

[0121] The basic correction magnitude for a function parameter is obtained by multiplying the deviation sensitivity coefficient by the current value of the parameter. For example, if the current value of parameter 'a' is 2.5 and its deviation sensitivity coefficient is 6, then the basic correction magnitude is 2.5 × 6 = 15. The correction magnitude is then assigned a sign based on the correction direction indicator to obtain the correction amount for the function parameter. If the correction direction indicator is "+1", the correction amount is +15; if it is "-1", the correction amount is -15.

[0122] In practical applications, to avoid excessively large corrections that could cause drastic changes in function parameters, a learning rate factor λ (e.g., 0.01) is typically introduced to control the correction step size. The final correction amount is the base correction magnitude multiplied by the correction direction indicator and then multiplied by the learning rate factor. For example, the final correction amount is 15 × (+1) × 0.01 = +0.15, indicating that parameter a needs to be increased by 0.15.

[0123] After calculating the corrections for all function parameters, update each parameter in the correlation function with these corrections. For example, the updated value for parameter 'a' is the original value plus the correction, i.e., 2.5 + 0.15 = 2.65. For complex correlation functions with multiple parameters, calculate the correction for each parameter using the method described above and update it accordingly.

[0124] The updated correlation function can be tested again using validation sample points to calculate a new amplitude deviation value. If the amplitude deviation value still exceeds the preset threshold (e.g., 0.1), the above correction process is repeated until the amplitude deviation value meets the accuracy requirements or reaches the maximum number of iterations (e.g., 50 times).

[0125] The above method can effectively generate correction values ​​for the function parameters in the correlation function based on the amplitude deviation value, and update the correlation function with the correction values, thereby improving the accuracy and adaptability of the correlation function. It is particularly suitable for the optimization of complex nonlinear parameters.

[0126] The present invention provides a system for predicting and optimizing the high-frequency signal transmission quality of electronic components, comprising:

[0127] The first unit is used to extract the attenuation change rate of each frequency band in the transmission path based on the frequency domain characteristics of the signal to be transmitted from the acquired components and the physical characteristic parameters of the transmission path, and to establish a correlation function between the attenuation change rate and the frequency band boundary position based on the physical characteristic parameters;

[0128] The second unit is used to calculate the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path based on the correlation function, thereby obtaining a predicted loss spectrum; and to determine the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum through an inverse mapping operation based on the predicted loss spectrum.

[0129] The third unit is used to perform frequency-selective amplitude adjustment on the signal to be transmitted at the transmission start point according to the compensation gain coefficient, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient to obtain the compensated signal;

[0130] The fourth unit is used to inject the compensated signal into the transmission path for transmission, and to collect the actual received signal after the compensated signal has been transmitted at the end of the transmission path;

[0131] The fifth unit is used to compare the frequency domain amplitude distribution of the actual received signal with the preset target frequency domain amplitude distribution band by band, calculate the amplitude deviation value of each frequency band, generate a correction amount for the function parameters in the correlation function based on the amplitude deviation value, and update the correlation function with the correction amount.

[0132] A third aspect of the embodiments of the present invention,

[0133] An electronic device is provided, comprising:

[0134] processor;

[0135] Memory used to store processor-executable instructions;

[0136] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0137] Fourth aspect of the present invention,

[0138] A computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0139] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting and optimizing the high-frequency signal transmission quality of electronic components, characterized in that, include: Based on the frequency domain characteristics of the signal to be transmitted from the components and the physical characteristic parameters of the transmission path, the attenuation change rate in the transmission path is extracted for each frequency band, and a correlation function between the attenuation change rate and the frequency band boundary position is established according to the physical characteristic parameters. Based on the correlation function, the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path is calculated to obtain the predicted loss spectrum; according to the predicted loss spectrum, the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum is determined by inverse mapping operation; Based on the compensation gain coefficient, the signal to be transmitted is frequency-selectively amplitude-adjusted at the transmission start point, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient, thereby obtaining the compensated signal; The compensated signal is injected into the transmission path for transmission, and the actual received signal after transmission is collected at the terminal of the transmission path. The frequency domain amplitude distribution of the actual received signal is compared with the preset target frequency domain amplitude distribution band by band, the amplitude deviation value of each frequency band is calculated, the correction amount for the function parameters in the correlation function is generated based on the amplitude deviation value, and the correlation function is updated with the correction amount.

2. The method according to claim 1, characterized in that, For each frequency band, the attenuation change rate in the transmission path is extracted, and a correlation function between the attenuation change rate and the frequency band boundary position is established based on the physical characteristic parameters, including: Based on the distribution density of frequency components in the frequency domain features, the frequency domain features are divided into multiple continuous frequency bands, each frequency band having a start boundary frequency and an end boundary frequency; For each frequency band, based on the dielectric loss factor and transmission path length in the physical characteristic parameters, the attenuation gradient of the signal amplitude as a function of frequency within that frequency band is calculated, and the attenuation gradient is used as the attenuation rate of that frequency band. Using the starting and ending boundary frequencies of all frequency bands as independent variables and the attenuation change rate of the corresponding frequency bands as dependent variables, a correlation function describing the mapping relationship between the frequency band boundary positions and the attenuation change rate is established through fitting operations.

3. The method according to claim 1, characterized in that, Based on the correlation function, the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path is calculated to obtain the predicted loss spectrum; according to the predicted loss spectrum, the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum is determined by inverse mapping operation, including: Frequency values ​​of all discrete frequency components are extracted from the frequency domain characteristics of the signal to be transmitted to form a frequency component set. For each frequency value in the frequency component set, the frequency value is substituted into the correlation function, and the attenuation change rate of the frequency band to which the frequency value belongs is determined by the correlation function. The predicted attenuation amount corresponding to the frequency value is calculated based on the attenuation change rate and the physical characteristic parameters of the transmission path. The frequency values ​​in the frequency component set are correlated and mapped with their corresponding predicted attenuation values ​​to form a predicted loss spectrum. Based on the predicted attenuation value range of each frequency component in the predicted loss spectrum, the frequency components in the predicted loss spectrum are divided into multiple attenuation levels, and each attenuation level contains frequency components whose predicted attenuation values ​​are in the same range. For each attenuation level, based on the center value of the predicted attenuation value range corresponding to that attenuation level, the reference compensation gain coefficient of that attenuation level is determined through inverse mapping operation, and this reference compensation gain coefficient is assigned to all frequency components in that attenuation level as their respective compensation gain coefficients; All frequency components and their corresponding compensation gain coefficients are associated and stored to form a set of compensation gain coefficients.

4. The method according to claim 3, characterized in that, For each attenuation level, based on the center value of the predicted attenuation range corresponding to that attenuation level, the reference compensation gain coefficient for that attenuation level is determined through inverse mapping operations, including: The arithmetic mean of the upper and lower bounds of the predicted attenuation value interval for each attenuation level is taken as the center value, and the difference between the upper and lower bounds is taken as the interval width; the ratio of the center value to the interval width is taken as the relative dispersion parameter. Obtain the initial signal strength value at the beginning of the transmission path, and the target signal strength value that should be achieved after transmission compensation; Based on the relative dispersion parameter, the compensation tolerance coefficient corresponding to the attenuation level is calculated through a monotonically decreasing mapping relationship. The compensation tolerance coefficient represents the maximum allowable proportion of the signal strength after compensation deviating from the target strength value within the attenuation level. The compensation tolerance coefficient is used as the benchmark compensation gain coefficient.

5. The method according to claim 4, characterized in that, Based on the compensation gain coefficient, the signal to be transmitted is frequency-selectively amplitude-adjusted at the transmission start point, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient, resulting in a compensated signal including: All discrete frequency components are identified from the signal to be transmitted, and the frequency value and initial amplitude value of each frequency component are extracted; for each frequency component, the corresponding compensation gain coefficient is found in the compensation gain coefficient set according to the frequency value of that frequency component; Calculate the total energy of the signal to be transmitted in the frequency domain, determine the maximum allowable total energy threshold based on the power limitation requirements at the transmission starting point, and calculate the energy margin between the total energy and the maximum total energy threshold; For each frequency component, the initial amplitude value of the frequency component is multiplied by its corresponding compensation gain coefficient to obtain a preliminary adjustment amplitude value. Based on the energy margin, the preliminary adjustment amplitude values ​​of all frequency components are normalized and scaled to obtain the adjusted amplitude values ​​of each frequency component that satisfy the power limit. Keeping the frequency and phase information of each frequency component unchanged, the initial amplitude value of each frequency component is replaced with the corresponding adjusted amplitude value, and the adjusted frequency domain signal is reconstructed; the adjusted frequency domain signal is then subjected to inverse time-domain transformation to convert the adjusted frequency domain signal from frequency domain representation back to time domain representation to obtain the compensated signal.

6. The method according to claim 1, characterized in that, Generating a correction amount for the function parameters in the correlation function based on the magnitude deviation value, and updating the correlation function with the correction amount includes: Multiple validation sample points are obtained. A preset perturbation increment is applied to each function parameter in the correlation function. The change in the output value of the correlation function at all validation sample points before and after applying the perturbation increment is calculated. The ratio of the change in the output value to the perturbation increment is calculated to obtain the deviation sensitivity coefficient corresponding to the function parameter. Based on the magnitude deviation value, the gradient vector of the output value of the correlation function relative to the function parameters is calculated. The signs of each component in the gradient vector are statistically analyzed to determine the dominant sign of the negative gradient direction. Based on the dominant sign, the correction direction indicator of the function parameters is determined. The deviation sensitivity coefficient is multiplied by the current value of the function parameter to obtain the basic correction magnitude of the function parameter; the basic correction magnitude is then assigned a sign value according to the correction direction indicator to obtain the correction amount of the function parameter.

7. A system for predicting and optimizing the high-frequency signal transmission quality of electronic components, used to implement the method as described in any one of claims 1-6, characterized in that, include: The first unit is used to extract the attenuation change rate of each frequency band in the transmission path based on the frequency domain characteristics of the signal to be transmitted from the acquired components and the physical characteristic parameters of the transmission path, and to establish a correlation function between the attenuation change rate and the frequency band boundary position based on the physical characteristic parameters; The second unit is used to calculate the predicted attenuation of each frequency component of the signal to be transmitted at the end of the transmission path based on the correlation function, thereby obtaining a predicted loss spectrum; and to determine the compensation gain coefficient corresponding to the attenuation of each frequency component in the predicted loss spectrum through an inverse mapping operation based on the predicted loss spectrum. The third unit is used to perform frequency-selective amplitude adjustment on the signal to be transmitted at the transmission start point according to the compensation gain coefficient, so that each frequency component in the signal to be transmitted is differentially amplified according to the compensation gain coefficient to obtain the compensated signal; The fourth unit is used to inject the compensated signal into the transmission path for transmission, and to collect the actual received signal after the compensated signal has been transmitted at the end of the transmission path; The fifth unit is used to compare the frequency domain amplitude distribution of the actual received signal with the preset target frequency domain amplitude distribution band by band, calculate the amplitude deviation value of each frequency band, generate a correction amount for the function parameters in the correlation function based on the amplitude deviation value, and update the correlation function with the correction amount.

8. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.

Images