Communication line noise filtering system and method

Abstract

The present invention relates to the field of noise filtering technology, specifically a communication line noise filtering system and method, the system comprising: a temperature coupling module, a voltage response module, a parameter joint adjustment module, and a link verification module. In the present invention, by collecting the temperature data of the inductor and capacitor nodes and inputting them into a piecewise linear regression model, the temperature deviation gradient of the resonant frequency is calculated, the noise suppression bandwidth threshold is dynamically matched in combination with the reference table, the frequency stability under temperature fluctuations is improved, the Kalman filter processes the voltage sampling to generate a fluctuation prediction value, the interval index is extracted based on the parameter mapping table, the least squares method is used to fit and generate the real-time inductor and capacitor parameters, and the voltage transient fluctuation is adaptively realized. The noise power spectrum density is calculated by fast Fourier transform, and the parameter rollback is triggered by comparing the 3dB attenuation point with the threshold difference, thereby forming a closed-loop control to ensure that the suppression bandwidth is continuously matched with the noise frequency band, reduce the bandwidth offset to within 1.2%, and optimize the response time to shorten it to 200ms.

Description

Technical Field

[0001] The present invention relates to the technical field of noise filtering, and in particular to a communication line noise filtering system and method. Background Art

[0002] The field of noise filtering technology includes technical methods for reducing or suppressing unnecessary electrical noise interference in various communication and electronic systems. The core content of this technical field is to filter out interference signals in communication lines by selecting different types of filtering circuits to ensure the integrity and stability of communication signals. Noise filtering is usually based on filter structures formed by capacitors, inductors and their combinations, and is widely used in wired communications, wireless communications, power transmission and electronic equipment. In terms of systemic applications, noise filtering technology covers various aspects such as source noise suppression, path noise blocking and receiving end noise elimination, and selects adaptive filtering strategies based on the frequency characteristics and intensity of the noise, such as using circuit structures such as low-pass filters, high-pass filters, and band-pass filters to effectively process noise in different frequency bands.

[0003] Among them, the communication line noise filtering system refers to a hardware system or circuit structure used in a wired communication path to reduce external electromagnetic interference or self-signal crosstalk. The subject of this patent is mainly aimed at the noise problem caused by the complex electromagnetic interference environment in the communication line, covering the use of passive filters, active noise reduction circuits, power line filters and signal line filters. Specifically, by setting a low-pass filter combined with an LC filter or a π-type filter in series between the communication line and the signal processing unit, the impedance matching technology is used to weaken the interference caused by the transmission of external electromagnetic waves on the communication line. At the same time, the grounding device is reasonably arranged to reduce the interference conduction between the lines, thereby completing the suppression and management of the communication path noise.

[0004] Existing fixed-structure filters fail to automatically adjust their inductor and capacitor parameters with temperature fluctuations. For every 10°C increase in ambient temperature, the inductor drifts by 5%-8%, causing a resonant frequency shift and a 3-5MHz suppression blind spot. Traditional impedance matching strategies fail to account for transient voltage fluctuations. When voltage changes exceed ±7%, the filter's insertion loss curve exhibits nonlinear distortion, resulting in a 2.4-3.6dB reduction in effective suppression bandwidth. Manual parameter adjustment relies on offline spectrum test data. Each calibration requires interrupting the communication link and takes over 30 minutes. It also fails to capture transient interference signals shorter than 500ms. Existing power spectrum detection uses a timed sampling mechanism, with a capture probability of less than 65% for intermittent pulse noise and a 15%-22% miss rate in sudden interference scenarios. Grounding layout strategies fail to integrate dynamic parameters, resulting in a 40%-55% decrease in inter-line interference conduction suppression efficiency with increasing temperature. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a communication line noise filtering system and method.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A communication line noise filtering system includes:

[0007] A temperature coupling module is configured to collect real-time temperature values ​​of the inductor node and the capacitor node through dual-channel thermistors, input the real-time temperature values ​​of the inductor node and the capacitor node into a piecewise linear regression model to calculate the deviation gradient of the temperature to the resonant frequency, generate a temperature deviation gradient value, and transmit the temperature deviation gradient value and a preset resonant frequency reference table to the voltage response module;

[0008] a voltage response module, configured to receive the temperature deviation gradient value and the resonant frequency reference table, match the noise suppression bandwidth threshold corresponding to the current temperature based on the resonant frequency reference table, detect the input voltage amplitude fluctuation rate, invoke a Kalman filter algorithm to process the voltage sampling sequence to generate a voltage fluctuation prediction value, and pass the result to the parameter joint adjustment module;

[0009] A parameter coordination module is configured to receive the voltage fluctuation prediction value, extract an interval index from a preset inductance-capacitance parameter mapping table based on the voltage fluctuation prediction value, perform least squares linear fitting on the inductance value to generate an instantaneous inductance parameter, perform the same operation on the capacitance value to generate an instantaneous capacitance parameter, and pass the instantaneous inductance parameter and the instantaneous capacitance parameter to a link verification module.

[0010] As a further solution of the present invention, the temperature deviation gradient value includes a temperature measurement error correction factor, a resonant frequency change compensation coefficient, and a node thermal response characteristic parameter; the voltage fluctuation prediction value includes a voltage stability index, a voltage fluctuation trend parameter, and a voltage noise prediction value; the real-time inductance parameter includes a dynamic inductance adjustment coefficient, a real-time temperature drift calibration amount, and an inductance tolerance correction amount; the real-time capacitance parameter includes a dynamic capacitance adjustment coefficient, a real-time frequency drift calibration amount, and a capacitance tolerance correction amount.

[0011] As a further solution of the present invention, the piecewise linear regression model dynamically divides the segment nodes based on the gradient change rate of the inductance and capacitance temperature intervals, and sets the weight coefficients of the linear regression in multiple intervals;

[0012] The Kalman filter algorithm constructs a state equation by using the ambient temperature interference correction factor, the voltage attenuation dynamic coefficient and the voltage sequence dispersion index;

[0013] The inductance-capacitance parameter mapping table divides the fluctuation range into upper and lower limits based on 1.2-1.5 times the average value of the voltage fluctuation rate.

[0014] As a further solution of the present invention, the temperature coupling module includes:

[0015] The temperature acquisition submodule detects the temperature sensor signals of the inductor and capacitor nodes, uses a dual-channel signal synchronization acquisition circuit, sets the signal sampling interval and noise filtering threshold, performs amplitude calibration and timing alignment on the original signal, and records the inductor temperature value and capacitor temperature value separately, establishing two sets of real-time temperature time series data with synchronized timestamps to generate dual-channel temperature values.

[0016] The regression processing submodule divides the inductor and capacitor temperature intervals into segmented nodes based on the dual temperature values, sets weight coefficients for linear regression within multiple intervals, inputs the inductor temperature value as an independent variable and the capacitor temperature value as a covariate into the model, calculates the slope change between the temperature and the resonant frequency within the multiple segmented intervals, extracts the slope parameter set of the segmented intervals, and generates a regression coefficient set;

[0017] The segment nodes are dynamically adjusted when the gradient change rate of adjacent temperature sampling points exceeds 0.5°C / s;

[0018] The gradient calculation submodule calls the regression coefficient set, selects the corresponding slope parameter in combination with the segmented interval to which the current dual-channel temperature value belongs, and multiplies the difference between the dual-channel temperature value and the reference frequency value by the slope parameter according to the temperature-frequency mapping relationship in the resonant frequency reference table. The module outputs the frequency offset caused by the unit temperature change to generate a temperature deviation gradient value.

[0019] As a further solution of the present invention, the voltage response module includes:

[0020] The threshold matching submodule calls the temperature-frequency mapping relationship in the resonant frequency reference table based on the temperature deviation gradient value, performs cubic spline interpolation calculation on the current temperature value, matches the nearest neighbor reference frequency point, extracts the upper and lower limit calibration values ​​of the noise suppression bandwidth corresponding to the frequency point, and generates a bandwidth threshold matching value;

[0021] The cubic spline interpolation satisfies the requirement that the first-order derivative at the node is continuous and the second-order derivative error is less than 0.1 Hz / °C;

[0022] The fluctuation detection submodule detects the peak-to-valley difference of the input voltage signal amplitude, intercepts the sampling sequence with a fixed time window, calculates the absolute value of the amplitude difference of adjacent sampling points within the window, and divides the accumulated sum by the product of the time window length and the reference amplitude to generate the voltage fluctuation rate;

[0023] The filter prediction submodule calls the bandwidth threshold matching value and the voltage fluctuation rate using the formula:

[0024] ;

[0025] The voltage fluctuation prediction component is obtained by calculation, and the prediction covariance matrix in the Kalman filter state equation is superimposed to generate the voltage fluctuation prediction value;

[0026] in, represents the voltage fluctuation prediction component, Represents the temperature deviation gradient value, Represents the resonant frequency reference value, represents the noise suppression bandwidth threshold, represents the voltage fluctuation rate, Represents the ambient temperature interference correction factor, which is obtained by experimentally calibrating the contribution of temperature interference to voltage fluctuation. Represents the voltage attenuation dynamic coefficient, which is fitted through regression analysis of historical voltage data. Represents the voltage sequence dispersion index, which is the ratio of the standard deviation to the mean of the voltage sampling sequence.

[0027] As a further solution of the present invention, the parameter joint adjustment module includes:

[0028] The fluctuation interval index submodule collects the voltage fluctuation prediction value, compares the prediction value with the upper and lower limits of the fluctuation interval in the mapping table according to the interval division rules defined in the preset inductance-capacitance parameter mapping table, determines the index code of the interval to which the prediction value belongs, and generates a fluctuation interval index;

[0029] The inductance parameter generation submodule calls the fluctuation interval index, extracts the inductance value set corresponding to the index in the mapping table, constructs a linear equation system with the index sequence as the independent variable and the inductance value as the dependent variable, calculates the minimum value of the residual sum of squares of the equation system through the least squares method, solves the slope and intercept parameters of the inductance value, and generates the real-time inductance parameter;

[0030] The least squares method iteration termination condition is that the residual sum of squares is less than 0.01 or the number of iterations exceeds 50;

[0031] The capacitance parameter generation submodule calls the fluctuation interval index, extracts the capacitance value set corresponding to the index in the mapping table, and based on the linear relationship between the index sequence and the capacitance value, applies the same least squares iterative process as the inductance parameter generation submodule to calculate the slope and intercept parameters of the capacitance value to generate the real-time capacitance parameter.

[0032] As a further embodiment of the present invention, the system further comprises:

[0033] The link verification module is used to receive the real-time inductance parameter and the real-time capacitance parameter, calculate the noise power spectrum density at the filter output end through fast Fourier transform, compare the difference between the 3dB attenuation point of the power spectrum density and the noise suppression bandwidth threshold, and trigger a parameter rollback instruction if the difference exceeds the threshold. The instruction is fed back to the parameter joint adjustment module to regenerate the parameters, output the effective inductance parameter and the effective capacitance parameter, and load them into the filter control unit.

[0034] As a further solution of the present invention, the effective inductance parameters and the effective capacitance parameters include a noise suppression accuracy index, a filter performance verification result, and a link signal integrity evaluation value;

[0035] The difference threshold is calibrated to ±10% of the noise suppression bandwidth threshold through spectrum analysis experiments.

[0036] As a further solution of the present invention, the link verification module includes:

[0037] The noise spectrum calculation submodule calls the instantaneous inductance parameter and the instantaneous capacitance parameter to obtain a noise time domain signal sequence at the filter output end, applies a fast Fourier transform to convert the time domain signal into a frequency domain component, calculates the ratio of the square of the amplitude of the frequency domain component to the frequency bandwidth, and generates a noise power spectrum density;

[0038] The threshold verification submodule locates the frequency point in the power spectrum where the amplitude attenuates to 70.7% of the maximum amplitude based on the noise power spectrum density as the 3dB attenuation point, extracts the preset noise suppression bandwidth threshold, calculates the absolute difference between the attenuation point and the threshold, compares the difference with the threshold allowable range, determines whether the difference exceeds the upper limit of the range, and generates a difference verification result;

[0039] The threshold range is calibrated to [85%, 115%] through spectrum analysis experiments of 500 groups of noise samples;

[0040] The parameter validation submodule calls the difference verification result and executes branch logic according to the over-limit status. If the limit is exceeded, the current parameter is cleared and the rollback instruction is triggered. If the limit is not exceeded, the parameter validation status is marked and the effective inductance parameter and the effective capacitance parameter are generated.

[0041] A communication line noise filtering method is provided, wherein the communication line noise filtering method is performed based on the above-mentioned communication line noise filtering system and comprises the following steps:

[0042] S1: Use dual temperature sensors to collect real-time temperature values ​​of the inductor node and capacitor node respectively. Input the two sets of temperature values ​​into the piecewise linear regression model to calculate the temperature deviation gradient of the resonant frequency and generate the temperature deviation gradient value.

[0043] S2: Based on the temperature deviation gradient value, a preset resonant frequency reference table is searched to match the noise suppression bandwidth threshold corresponding to the current temperature. At the same time, the input voltage amplitude fluctuation rate is detected, and a Kalman filter algorithm is used to perform prediction processing on the voltage sampling sequence to generate a voltage fluctuation prediction value.

[0044] S3: Locating an interval index in the inductance-capacitance parameter mapping table according to the voltage fluctuation prediction value, performing a least squares linear fitting operation on the inductance value to generate an instantaneous inductance parameter, and simultaneously performing the same operation on the capacitance value to generate an instantaneous capacitance parameter;

[0045] S4: For the instantaneous inductance parameters and the instantaneous capacitance parameters, the noise power spectrum density at the output end of the filter is calculated by fast Fourier transform, and the difference between the 3dB attenuation point of the power spectrum density and the noise suppression bandwidth threshold is compared. If the difference exceeds the preset range, the parameter rollback instruction is triggered to regenerate the parameters; otherwise, the effective inductance parameters and the effective capacitance parameters are output to the filter control unit.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are:

[0047] In the present invention, by collecting the temperature data of the inductor and capacitor nodes and inputting them into the piecewise linear regression model, the deviation gradient of the temperature to the resonant frequency is calculated, and the noise suppression bandwidth threshold is dynamically matched in combination with the resonant frequency reference table to improve the frequency stability of the filter in the temperature fluctuation scenario. The voltage sampling sequence is processed by the Kalman filter algorithm to generate the fluctuation prediction value, and the interval index is extracted based on the inductor-capacitor parameter mapping table. The real-time inductor and capacitor parameters are generated by least squares fitting to achieve parameter adaptability during transient voltage fluctuations. The output noise power spectrum density is calculated using fast Fourier transform, and the difference between the 3dB attenuation point and the threshold is compared to trigger parameter rollback, forming a closed-loop control structure to ensure that the suppression bandwidth and the noise frequency band are continuously matched. This solution couples the temperature gradient, voltage fluctuation and frequency domain characteristics for analysis, so that the noise suppression bandwidth offset in a dynamic environment is reduced to less than 1.2%, and the parameter optimization response time is shortened to the 200ms level. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a system flow chart of the present invention;

[0049] Figure 2 This is a flow chart of the temperature coupling module of the present invention;

[0050] Figure 3 This is a flow chart of the voltage response module of the present invention;

[0051] Figure 4 This is a flow chart of the parameter joint adjustment module of the present invention;

[0052] Figure 5This is a flow chart of the link verification module of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0055] Example 1

[0056] See also Figure 1 The present invention provides a technical solution: a communication line noise filtering system comprising:

[0057] The temperature coupling module is used to collect the real-time temperature values ​​of the inductor node and the capacitor node through dual-channel thermistors, input the two into a piecewise linear regression model to calculate the deviation gradient of temperature to resonant frequency, generate a temperature deviation gradient value, and transmit the temperature deviation gradient value and a preset resonant frequency reference table to the voltage response module;

[0058] The voltage response module is used to receive the temperature deviation gradient value and the resonant frequency reference table, match the noise suppression bandwidth threshold corresponding to the current temperature based on the resonant frequency reference table, detect the input voltage amplitude fluctuation rate, call the Kalman filter algorithm to process the voltage sampling sequence to generate a voltage fluctuation prediction value, and pass it to the parameter joint debugging module;

[0059] A parameter joint adjustment module is used to receive the voltage fluctuation prediction value, extract the interval index from the preset inductance-capacitance parameter mapping table based on the voltage fluctuation prediction value, perform the least squares linear fitting on the inductance value to generate the real-time inductance parameter, perform the same operation on the capacitance value to generate the real-time capacitance parameter, and pass the real-time inductance parameter and the real-time capacitance parameter to the link verification module;

[0060] The link verification module is used to receive the real-time inductance parameters and real-time capacitance parameters, calculate the noise power spectrum density at the filter output end through fast Fourier transform, and compare the difference between the 3dB attenuation point of the power spectrum density and the noise suppression bandwidth threshold. If the difference exceeds the threshold, the parameter rollback instruction is triggered and fed back to the parameter joint adjustment module to regenerate the parameters. The effective inductance parameters and effective capacitance parameters are output and loaded into the filter control unit.

[0061] The temperature deviation gradient value includes the temperature measurement error correction factor, the resonant frequency change compensation coefficient, and the node thermal response characteristic parameters. The voltage fluctuation prediction value includes the voltage stability index, the voltage fluctuation trend parameter, and the voltage noise estimation value. The real-time inductance parameters include the dynamic inductance adjustment coefficient, the real-time temperature drift calibration amount, and the inductance tolerance correction amount. The real-time capacitance parameters include the dynamic capacitance adjustment coefficient, the real-time frequency drift calibration amount, and the capacitance tolerance correction amount. The effective inductance parameters and effective capacitance parameters include the noise suppression accuracy index, the filter performance verification results, and the link signal integrity assessment value.

[0062] The piecewise linear regression model dynamically divides segment nodes based on the gradient change rate of the inductor and capacitor temperature intervals, and sets the weight coefficients of the linear regression in multiple intervals;

[0063] The Kalman filter algorithm constructs the state equation through the ambient temperature interference correction factor, voltage attenuation dynamic coefficient and voltage sequence dispersion index;

[0064] The inductor-capacitor parameter mapping table divides the fluctuation range into upper and lower limits based on 1.2-1.5 times the average voltage fluctuation rate;

[0065] The difference threshold is calibrated to ±10% of the noise suppression bandwidth threshold through spectrum analysis experiments.

[0066] See also Figure 2 , the temperature coupling module includes:

[0067] The temperature acquisition submodule detects the temperature sensor signals of the inductor and capacitor nodes, uses a dual-channel signal synchronization acquisition circuit, sets the signal sampling interval and noise filtering threshold, performs amplitude calibration and timing alignment on the original signal, and records the inductor temperature value and capacitor temperature value separately, establishing two sets of real-time temperature time series data with synchronized timestamps to generate dual-channel temperature values.

[0068] The temperature acquisition submodule accurately and synchronously acquires the operating temperatures of the inductors and capacitors. First, high-precision temperature sensors, such as PT1000 platinum resistance thermometers, are deployed in close proximity to the critical heat dissipation paths of the inductors and capacitors. These sensors continuously monitor the temperature of the target node and output a corresponding resistance or voltage signal. This triggers a dual-channel synchronous signal acquisition circuit, which uses a shared high-frequency clock signal (set to 100 kHz) to ensure that the sampling instants of the two sensor signal channels are strictly aligned, avoiding errors introduced by sampling time differences. The signal sampling interval is set to 10 milliseconds, meaning the system takes 100 readings per second from each sensor.

[0069] The collected raw signal is first processed by noise filtering. The noise filtering threshold is set to 0.1℃. This threshold is determined based on the analysis of the inherent noise characteristics of the sensor and the normal operating temperature change rate. It is intended to filter out rapid, random fluctuations with an amplitude of less than 0.1℃ caused by electromagnetic interference or the sensor itself. Compared with the reading of the previous sampling point Compare them. If the absolute difference between the two is less than 0.1℃, that is, , then determine is noise, and its value is corrected to This process is applied independently to the temperature signals of the inductor and capacitor.

[0070] After noise filtering is completed, the signal enters the amplitude calibration phase. The filtered readings are calibrated using the sensor calibration data table stored in the non-volatile memory (the table is generated based on factory calibration and periodic on-site multi-point temperature calibration, and contains the mapping relationship between the sensor's precise output value and the corresponding temperature at standard temperatures of 0°C, 25°C, 50°C, 75°C, and 100°C). Processing. Using piecewise linear interpolation, the precise Celsius temperature value is calculated based on the specific interval in the calibration data table where the reading falls. Taking the inductor temperature as an example, if the filtered reading is 1.185V, the calibration table shows that the output is 1.150V at 50℃ and 1.200V at 60℃, then the calibrated temperature Calculated as . Capacitor temperature Perform the same calibration process.

[0071] Finally, the timing alignment is confirmed. Although the hardware design ensures synchronous acquisition, the software level will recheck the timestamps of the two data channels. Based on the timestamp of the inductor temperature channel, fine-tune (usually within microseconds) the timestamp of the capacitor temperature data point to ensure that the timestamp recorded in the database is accurate. and Strictly correspond to the same physical moment. Calibrated and aligned inductor temperature values and capacitance temperature value , along with a precise timestamp , are recorded in a structured manner to form a continuous time series data stream, such as [(t1,57.0,54.2),(t2,57.1,54.3),(t3,57.1,54.4),…]. This submodule finally outputs the dual-channel temperature value at the current moment, i.e., the inductor temperature and capacitor temperature .

[0072] The regression processing submodule divides the inductor and capacitor temperature intervals into segmented nodes based on the dual-path temperature values, sets the weight coefficients for linear regression within multiple intervals, inputs the inductor temperature value as the independent variable and the capacitor temperature value as the covariate into the model, calculates the slope change between temperature and resonant frequency within multiple segmented intervals, extracts the slope parameter set of the segmented interval, and generates a regression coefficient set;

[0073] Segment nodes are dynamically adjusted when the gradient change rate of adjacent temperature sampling points exceeds 0.5°C / s;

[0074] The regression processing submodule receives the dual-channel temperature time-series data continuously output by the temperature acquisition submodule. Its core task is to establish a mathematical model of the dynamic relationship between temperature and resonant frequency. First, it is necessary to determine the interval demarcation points, or segment nodes, of the piecewise linear model describing this relationship. Initial nodes are set based on the device specification and historical operating data analysis, covering the expected operating temperature range: [20°C, 30°C, 40°C, 50°C, 60°C, 70°C, 80°C]. These nodes are not fixed but are dynamically adjusted based on real-time data.

[0075] The basis for dynamic adjustment is the rate of temperature change. Calculate the temperature gradient change rate within the last N sampling points (set N=5, corresponding to a 50ms time window). and capacitance temperature series Calculate adjacent sampling points (time interval ) temperature change rate and The changes in these rates of change are further calculated, namely the second-order derivative or the difference in the rates of change. A dynamic adjustment threshold is set: when the average gradient change rate of any temperature channel at M consecutive points (set as M=3) exceeds 0.5°C / s, the system determines that the temperature change has entered a rapid dynamic period. The gradient change rate threshold of 0.5°C / s is set based on statistical analysis of the system's thermal inertia and the temperature rise rate under normal operating conditions. Exceeding this value indicates the possibility of entering a nonlinear response region or rapid environmental changes. At this time, the system inserts a new segment node near the current temperature point or adjusts the position of existing nodes to more accurately capture the characteristics of the rapidly changing region. For example, if the inductor temperature rises from 48.8°C to 49.0°C over 30ms (3 sampling points), the average gradient is (49.0-48.8) / 0.03=6.67°C / s, which far exceeds 0.5°C / s. Therefore, the node near 49°C is adjusted or added. Assume that after adjustment, the current valid nodes are [20,30,40,49,58,70,80]℃.

[0076] Next, a linear regression weight coefficient is assigned to each temperature interval determined by the node The setting of weight coefficient reflects the confidence or data quality of model fitting in different temperature ranges. The historical data points in a certain interval are dense and A value close to 1 indicates that the interval relationship is stable and is given a higher weight; on the contrary, the data is sparse or The interval with lower value is given lower weight. Specific setting process: Analyze historical data and calculate the interval [40℃, 49℃) The value is 0.98, which is set as the baseline weight , interval [49℃,58℃) The value is 0.92, then its weight is set to . Set the weights for all intervals .

[0077] Then, the inductor temperature value collected in real time is As an independent variable, the capacitance temperature value As a covariate, and the corresponding resonant frequency value measured by other modules of the system , are jointly input into the piecewise weighted linear regression model. temperature ranges (defined by dynamic nodes, such as [49°C, 58°C)), the model is expressed as . Using all valid historical and real-time data points within this interval {( )}, and combined with the weight coefficient of the interval , the weighted least squares method is used to solve the model parameters. The goal of the weighted least squares method is to minimize the weighted residual sum of squares By solving this optimization problem, the slope of the resonant frequency to the inductor temperature in this range is obtained. and the slope of the capacitance versus temperature Repeat this calculation for all N segment intervals to obtain a complete set of slope parameters {( ),( ),…,( )}. Assume that for the interval [49℃,58℃), the calculated slope is and Finally, this submodule generates a set of regression coefficients containing the slope parameters of all segmented intervals.

[0078] The gradient calculation submodule calls the regression coefficient set, combines the segmented interval to which the current dual-channel temperature value belongs, selects the corresponding slope parameter, and multiplies the difference between the dual-channel temperature value and the reference frequency value by the slope parameter according to the temperature-frequency mapping relationship in the resonant frequency reference table. It outputs the frequency offset caused by the unit temperature change and generates the temperature deviation gradient value.

[0079] When the gradient calculation submodule is started, it first calls the regression coefficient set {( ),…,( )}. At the same time, obtain the latest dual-channel temperature value, that is, the temperature acquisition submodule and . Assume the current inductor temperature is , the capacitor temperature is .

[0080] The system determines the segment interval to which the current temperature value belongs based on the dynamically adjusted segment node list [20,30,40,49,58,70,80]℃. and , both temperature values ​​fall within the third valid interval [49°C, 58°C) (the index starts counting from 0, so it is the third interval, or the fourth interval according to the list order). Therefore, the system accurately selects the slope parameter pair corresponding to this interval from the regression coefficient set ( ). According to the calculation results of the previous submodule, this pair of parameters is and .

[0081] Next, the resonant frequency reference at the current temperature needs to be calculated. The system consults an internally stored resonant frequency reference table. This table, obtained by calibrating a precision network analyzer in a high-precision temperature-controlled chamber, records the ideal resonant frequency at a specific reference temperature point.

[0082] Table 1: Example of a temperature-frequency mapping benchmark table

[0083]

[0084] As shown in Table 1, the table lists the resonant frequency reference values ​​corresponding to some discrete reference temperature points.

[0085] The core task of this submodule is to calculate a gradient value representing the frequency offset caused by a unit temperature change in the current state. This gradient value reflects the overall sensitivity of the system frequency to temperature changes. The calculation method is based on the selected slope parameter ( ). Considering that the inductor is usually the main component affecting the frequency stability and the absolute value of its temperature coefficient Greater than the capacitance , this embodiment selects the absolute value of the slope of the inductor temperature as the dominant gradient. Therefore, the frequency offset caused by a unit temperature change of the output is determined to be This value is named as the temperature deviation gradient value Finally, this submodule generates the temperature deviation gradient value , for use by subsequent modules.

[0086] See also Figure 3 , the voltage response module includes:

[0087] The threshold matching submodule uses the temperature deviation gradient value to call the temperature-frequency mapping relationship in the resonant frequency reference table, performs cubic spline interpolation calculation on the current temperature value, matches the nearest neighbor reference frequency point, extracts the upper and lower limit calibration values ​​of the noise suppression bandwidth corresponding to the frequency point, and generates the bandwidth threshold matching value;

[0088] The cubic spline interpolation satisfies the requirement that the first-order derivative at the nodes is continuous and the second-order derivative error is less than 0.1 Hz / °C;

[0089] The threshold matching submodule receives the temperature deviation gradient value output by the gradient calculation submodule It combines the current real-time temperature and calls the stored resonant frequency reference table (Table 1) for calculation. First, calculate the average temperature of the current working state .

[0090] According to the discrete (temperature, frequency) data points in the benchmark table (Table 1), The reference resonant frequency at the location is accurately interpolated. The cubic spline interpolation method is used, which ensures that the interpolation curve is not only continuous at all original data nodes, but also the first-order derivative (slope) is continuous. At the same time, the selected cubic spline interpolation implementation must meet additional smoothness constraints: the rate of change of the second-order derivative (representing the degree of curve curvature) of the interpolation polynomial segment between any two data nodes with temperature must be less than 0.1Hz / ℃². This constraint is based on the recognition that the frequency-temperature curve of the physical system usually shows a smooth change. The upper limit of 0.1Hz / ℃² is set to suppress the sharp curvature that may be introduced by interpolation and does not conform to physical laws. Use the reference table The surrounding data points (40℃, 13.5615MHz), (50℃, 13.5600MHz), (60℃, 13.5580MHz), (70℃, 13.5570MHz) are interpolated by cubic spline using the standard numerical calculation library to obtain Corresponding interpolation reference frequency The calculation result is .

[0091] After obtaining the precise current reference frequency, match it with the discrete frequency values ​​in the reference table (Table 1) to find the reference point that is closest in value. The absolute value of the difference from each frequency in the table: The difference from 13.5600MHz (50℃) is , and the difference from 13.5580MHz (60℃) is By comparison, we can see that 13.5600MHz is the nearest neighbor reference frequency point. .

[0092] Finally, the system accesses an internal "frequency-bandwidth" calibration database. This database stores the filter bandwidth calibration values ​​required to achieve the target noise suppression performance for different reference center frequency points. These calibration values ​​are determined through offline experiments: the center frequency is tuned to Multiple filter prototypes are used to apply standard noise signals and repeatedly adjust their Parameters (i.e. adjust bandwidth) and measure output noise power until the noise suppression index (such as out-of-band suppression) is achieved. ) meets the standard, record the corresponding 3dB bandwidth at this time. , the corresponding calibrated noise suppression bandwidth is This submodule finally generates the bandwidth threshold matching value This value will be used as a reference for the target bandwidth in subsequent calculations.

[0093] The fluctuation detection submodule detects the peak-to-valley difference of the input voltage signal amplitude, intercepts the sampling sequence with a fixed time window, calculates the absolute value of the amplitude difference of adjacent sampling points within the window, and divides the accumulated sum by the product of the time window length and the reference amplitude to generate the voltage fluctuation rate;

[0094] The fluctuation detection submodule continuously monitors the voltage signal of the input system It intercepts and analyzes the signal in a fixed time window to quantify its fluctuation. Set the time window length The selection of the window length is based on the balance between the system dynamic response time and the frequency characteristics of the typical interference signal. 100ms can capture most power frequency interference and harmonics, while not being too long for the control loop response. The sampling rate of the voltage signal is set to , which means that within the 100ms window, a total of discrete voltage sample points.

[0095] The obtained voltage sampling sequence is recorded as A specific sampling sequence segment can be: 3.05V, 3.08V, 3.02V, 3.06V, 3.10V, 3.07V, 3.03V, ..., 3.09V. The submodule then calculates the sum of the absolute values ​​of the amplitude differences of all adjacent sampling points in this window. The calculation process is . Calculate the first few differences of the above sequence fragments: After traversing all 99 differences in the window and adding them up, we get the total .

[0096] At the same time, the arithmetic mean of all 100 voltage samples in the time window is calculated as the reference amplitude during this window. The calculation formula is Sum all samples in the window and divide by 100 to get .

[0097] Finally, the voltage fluctuation rate is calculated according to the formula The formula is defined as the accumulated sum of absolute differences divided by the product of the time window length (in seconds) and the reference amplitude: . Substituting the values: This value The unit is , which reflects the cumulative relative amplitude of voltage jump per unit time. This submodule finally generates the voltage fluctuation rate .

[0098] The filter prediction submodule calls the bandwidth threshold matching value and the voltage fluctuation rate using the formula:

[0099] ;

[0100] The voltage fluctuation prediction component is obtained by calculation, and the prediction covariance matrix in the Kalman filter state equation is superimposed to generate the voltage fluctuation prediction value;

[0101] in, represents the voltage fluctuation prediction component, Represents the temperature deviation gradient value, Represents the resonant frequency reference value, represents the noise suppression bandwidth threshold, represents the voltage fluctuation rate, Represents the ambient temperature interference correction factor, which is obtained by experimentally calibrating the contribution of temperature interference to voltage fluctuation. Represents the voltage attenuation dynamic coefficient, which is fitted through regression analysis of historical voltage data. Represents the voltage sequence dispersion index, which is the ratio of the standard deviation to the mean of the voltage sampling sequence.

[0102] The filter prediction submodule integrates the information from the previous steps and calculates a comprehensive voltage fluctuation prediction component , which is designed to quantify the overall disturbance level faced by the system and is used to guide subsequent Kalman filter adjustments. It calls the bandwidth threshold matching value and voltage fluctuation rate , and apply the following formula:

[0103] This formula combines the effects of temperature drift and inherent voltage fluctuation. Explanation and value of each parameter in the formula:

[0104] : Temperature deviation gradient value, provided by the gradient calculation submodule, reflects the sensitivity of frequency to temperature. .

[0105] : The current average temperature, calculated by the threshold matching submodule, .

[0106] : The nominal or design reference operating temperature of the system is a fixed design parameter, set to .

[0107] : The noise suppression bandwidth threshold, provided by the threshold matching submodule, is the target bandwidth of the filter. .

[0108] : Voltage attenuation dynamic coefficient is a unitless parameter, which is calculated by analyzing the historical voltage fluctuation rate. The coefficient reflecting the persistence of fluctuations is obtained by fitting the time series data of the autoregressive model (AR(1)). The specific operation is to collect the data calculated every second in the past few hours. Value sequence, fitting , the coefficients obtained This is what you need. .

[0109] : Relative voltage fluctuation rate is a unitless indicator used to represent voltage fluctuation in formulas. It is calculated using data from the fluctuation detection submodule: .

[0110] : Ambient temperature interference correction factor, unitless, quantifies the degree of influence of ambient temperature on relative voltage fluctuation. Calibrated through controlled experiments: at different stable ambient temperatures Run the system for a long time under the following conditions (from 20℃ to 70℃, step 10℃), and record the voltage sequence dispersion index at each temperature. The average value of . Plot ( ) scatter plot, perform linear fitting The slope obtained is This is what we are looking for. The experimental calibration results are .

[0111] : Voltage sequence dispersion index, unitless, defined as the standard deviation of voltage samples in the current time window and its mean For the voltage window data in paragraph 5, calculate its standard deviation as , the mean is ,but .

[0112] Now, substitute these values ​​into the formula to calculate. Note that the first term (temperature contribution) ): Calculate the absolute frequency deviation caused by temperature .bandwidth To calculate the ratio, the units need to be unified. Conversion rules: To convert from kHz to Hz, multiply by the conversion factor 1000 Hz / kHz. .but . This is a unitless ratio.

[0113] The second term (voltage contribution ): all parameters All are unitless. . This is also a unitless value.

[0114] Total voltage fluctuation prediction component .

[0115] Calculated is a comprehensive unitless disturbance indicator. This value is then used to adjust the prediction covariance matrix of the Kalman filter Specifically, the update process noise covariance , so that it is consistent with Related, e.g. ,in is the base process noise setting, is a regulating factor. As the value increases, the process noise This increase makes the Kalman filter more trustworthy of the current measurement, more adaptable, and able to track changes in system state more quickly. After this adjustment, the Kalman filter outputs an optimized estimate and prediction of the system state (such as voltage and frequency). This output is the voltage fluctuation prediction value.

[0116] Should Calculation results is relatively small (refer to the interval definition in Table 2 below), indicating that the overall disturbance level of the current system is low. The value itself is not the final output voltage prediction, but is passed as a key input to the subsequent parameter joint debugging module to select the appropriate L / C parameter range.

[0117] See also Figure 4 , the parameter joint debugging module includes:

[0118] The fluctuation interval index submodule collects the voltage fluctuation prediction value, compares the prediction value with the upper and lower limits of the fluctuation interval in the mapping table according to the interval division rules defined in the preset inductance-capacitance parameter mapping table, determines the index code of the interval to which the prediction value belongs, and generates the fluctuation interval index;

[0119] The fluctuation interval index submodule receives the voltage fluctuation prediction value calculated by the filter prediction submodule Its task is to convert this continuous The value is mapped to a discrete interval index corresponding to the preset inductor-capacitor parameter adjustment strategy. This mapping is performed according to the fluctuation interval division rules defined in the inductor-capacitor parameter mapping table stored in the system (as shown in Table 2 below).

[0120] Table 2: Example of inductor-capacitor parameter mapping table

[0121]

[0122] As shown in Table 2, this table defines different fluctuation prediction values ​​based on system design and simulation experiments. interval, the corresponding fluctuation level, and the recommended value range of inductance and capacitance parameters.

[0123] The system will value Compare the values ​​of the boundaries of each interval in Table 2 in sequence. First, determine .because , this condition is met. Therefore, the current voltage fluctuation prediction value It is determined to belong to the "low" fluctuation range with index code 1. The submodule then generates the result: the fluctuation range index is 1.

[0124] The inductor parameter generation submodule calls the fluctuation interval index, extracts the set of inductance values ​​corresponding to the index in the mapping table, constructs a linear equation system with the index sequence as the independent variable and the inductance value as the dependent variable, calculates the minimum value of the residual sum of squares of the equation system through the least squares method, solves the slope and intercept parameters of the inductance value, and generates the real-time inductance parameters;

[0125] The least squares method iteration termination condition is that the sum of squared residuals is less than 0.01 or the number of iterations exceeds 50;

[0126] The inductor parameter generation submodule receives the fluctuation range index 1. Based on this index, it first extracts the corresponding recommended inductance value set (95nH, 105nH) from the inductor-capacitor parameter mapping table (Table 2). To determine an accurate instantaneous inductance value within this range, the submodule utilizes stored historical operating data. This data consists of a sequence of pairs (index, actual effective inductance value / nH)), recording the inductance values ​​used to maintain stable operation within different fluctuation ranges over a period of time.

[0127] Assume that the most recent seven sets of historical data are: X (index) = [2, 1, 1, 3, 2, 1, 4], Y (inductance value / nH) = [108, 102, 100, 118, 111, 103, 128]. The system builds a linear model based on these historical data. , trying to find the linear trend between the index and the optimal inductance value. The least squares method is used to calculate the model parameters (slope) and (Intercept). The principle of least squares method is to find the parameter Make the predicted value of the historical data point ( ) and the actual value The sum of squared errors between Minimum. Solved by standard analytical solution or numerical optimization method, we get nH / index unit and nH. The solution process sets the iterative termination conditions: it stops when the change in the sum of squared residuals is less than 0.01nH² or the number of iterations exceeds 50. These conditions are set based on the trade-off between convergence speed and computing resources.

[0128] After obtaining the model parameters, use the current fluctuation range index 1 to substitute into the model for prediction: Check the predicted values Whether it falls within the recommended range [95nH, 105nH] corresponding to index 1. The result is within the range. Therefore, the system uses the predicted value as the instant inductance parameter. This submodule finally generates the instant inductance parameter .

[0129] The capacitor parameter generation submodule calls the fluctuation interval index, extracts the capacitance value set corresponding to the index in the mapping table, and applies the same least squares iterative process as the inductor parameter generation submodule based on the linear relationship between the index sequence and the capacitance value to calculate the slope and intercept parameters of the capacitance value and generate the real-time capacitance parameters.

[0130] The capacitor parameter generation submodule also receives the fluctuation interval index 1. It extracts the recommended capacitor value set corresponding to index 1, namely [130pF, 140pF], from the inductor-capacitor parameter mapping table (Table 2). Similar to the inductor parameter generation submodule, this submodule also uses historical data to accurately determine the capacitor value.

[0131] Use the same historical index sequence X = [2, 1, 1, 3, 2, 1, 4] and the corresponding historical effective capacitance value sequence Y (capacitance / pF) = [126, 138, 135, 115, 122, 136, 108] to build a linear model. , and apply the same least squares procedure (including the same termination conditions) to calculate the parameters and , the goal is to minimize Calculated pF / index unit and pF.

[0132] Use the current index 1 to substitute into the capacitance model for prediction:

[0133] Check the predicted values Is it within the recommended range [130pF, 140pF] corresponding to index 1? The result is within the range. The system uses this predicted value as the instantaneous capacitance parameter. This submodule finally generates the instantaneous capacitance parameter. .

[0134] See also Figure 5 , the link verification module includes:

[0135] The noise spectrum calculation submodule uses the real-time inductance and capacitance parameters to obtain the noise time-domain signal sequence at the filter output. It then applies the fast Fourier transform to convert the time-domain signal into a frequency-domain component. It then calculates the ratio of the square of the amplitude of the frequency-domain component to the frequency bandwidth to generate the noise power spectrum density.

[0136] The noise spectrum calculation submodule receives the instantaneous inductance parameters calculated by the inductance and capacitance parameter generation submodule and instantaneous capacitance parameters The system first loads these two parameter values ​​into the physical tunable filter circuit to make it work under the new L / C configuration.

[0137] Then, the noise signal at the filter output is immediately started. Time domain acquisition. Set acquisition duration , using high sampling rate The 20ms sampling rate is chosen to obtain enough signal cycles to ensure frequency resolution, while the 10MHz sampling rate is chosen to be able to analyze noise components up to 5MHz (Nyquist theorem), covering the broadband noise that the system may be concerned about. discrete noise voltage sample points .

[0138] After obtaining the complete time domain noise sequence, the Fast Fourier Transform (FFT) algorithm is applied to it. FFT efficiently converts these 200,000 time domain sample points into frequency domain representation. Since the input is a real signal, the output of FFT is a signal containing A sequence of complex frequency domain coefficients Each coefficient Corresponding to a specific frequency Therefore, the frequency domain analysis covers the period from 0Hz to The frequency resolution (i.e. the interval between adjacent frequency points) is .

[0139] Next, calculate each frequency domain coefficient The square of the magnitude This value represents the center frequency of , width is In order to obtain the standardized noise power spectral density (PSD), the power of each frequency point is Divide by the frequency resolution The calculation formula is Calculation results The unit is By all Performing this calculation from 0 to N / 2, we can obtain the noise power spectrum density values ​​at a series of discrete frequency points. This submodule finally generates this Noise power spectral density curve of (frequency, PSD) data points , for subsequent analysis.

[0140] The threshold verification submodule locates the frequency point where the amplitude in the power spectrum attenuates to 70.7% of the maximum amplitude based on the noise power spectrum density as the 3dB attenuation point, extracts the preset noise suppression bandwidth threshold, calculates the absolute difference between the attenuation point and the threshold, compares the difference with the threshold allowable range, determines whether the difference exceeds the upper limit of the range, and generates the difference verification result;

[0141] The threshold allowable range is calibrated to [85%, 115%] through spectrum analysis experiments of 500 sets of noise samples;

[0142] The threshold check submodule receives the noise power spectrum density data generated by the noise spectrum calculation submodule Its task is to evaluate whether the actual bandwidth of the filter under the current L / C parameter configuration meets the expectations. First, Search and locate the maximum value of the power spectral density on the curve This peak usually occurs at the resonant frequency of the filter. nearby.

[0143] Find the peak and its corresponding center frequency Afterwards, from Start searching in the low-frequency and high-frequency directions respectively, and find the first time that the power spectrum density value drops to (i.e. the power attenuates by 3dB). The 3dB point on the low frequency side is , the 3dB point on the high frequency side is By calculating Search the data and locate and .

[0144] Calculate the difference between these two 3dB frequency points to get the actual measured bandwidth of the filter . .

[0145] Next, the system calls the preset target noise suppression bandwidth threshold , which is provided by the bandwidth threshold matching submodule (or set directly). In this example, Then, the bandwidth will be measured and target bandwidth Compare them. Calculate the consistency between the two, usually measured as a ratio: .

[0146] The calculated ratio (i.e. 100%) is compared with the preset bandwidth allowable range. The allowable range is [85%, 115%], which is determined after statistical analysis experiments on the noise spectrum of 500 groups of samples with different working conditions and different devices. Statistical results show that when the bandwidth fluctuates within this range, the key performance indicators of the filter (such as selectivity and out-of-band suppression) can still meet the system requirements. Setting process: Calculate the distribution of the ratio of the measured bandwidth to the target bandwidth of 500 samples, take the 7.5th percentile and the 92.5th percentile as the lower and upper limits of the interval (approximately covering the middle 85% of the samples), and get [85%, 115%]. The judgment conditions are: .because , the condition is met. Therefore, the measurement bandwidth is determined to be within the allowable deviation range. This submodule ultimately generates the difference verification result: "Not exceeded".

[0147] The parameter validation submodule calls the difference verification result and executes the branch logic according to the over-limit status. If the limit is exceeded, the current parameter is cleared and the rollback instruction is triggered. If the limit is not exceeded, the parameter validation status is marked and the effective inductance parameter and effective capacitance parameter are generated.

[0148] The parameter validation submodule receives the difference calibration result from the threshold calibration submodule, and the current result is "within the limit." It then executes subsequent operations based on the internal control logic.

[0149] The logic states that if the received result is "out of limit" (meaning the measurement bandwidth Falls on [85%,115%] The system will determine the instantaneous inductance parameters generated by the current calculation. and instantaneous capacitance parameters If the filter fails to achieve the expected performance, the system will discard this set of parameters, not apply them to the hardware, and perform a rollback operation: reverting to the last successful and still-in-use L / C parameter pair. This may also trigger the parameter optimization algorithm to recalculate or issue an alarm.

[0150] However, since the current calibration result is "within the limit", the system executes the normal validation process. and instantaneous capacitance parameters Marked as "Verification Passed" status. Subsequently, the system issues a command to write these two values ​​into the register of the tunable filter hardware through the control interface, so that it can physically adjust the internal inductance and capacitance components (or equivalent circuits), so that the filter can be adjusted according to the new parameters. This submodule ultimately generates the inductance and capacitance parameters that are verified to be effective and represent the actual operation of the current system: and effective capacitance parameters These parameters will remain in effect until the next successful parameter update process is completed.

[0151] A communication line noise filtering method is provided, which is performed based on the above-mentioned communication line noise filtering system and includes the following steps:

[0152] S1: Use dual temperature sensors to collect real-time temperature values ​​of the inductor node and capacitor node respectively. Input the two sets of temperature values ​​into the piecewise linear regression model to calculate the temperature deviation gradient of the resonant frequency and generate the temperature deviation gradient value.

[0153] S2: Based on the temperature deviation gradient value, the preset resonant frequency reference table is searched to match the noise suppression bandwidth threshold corresponding to the current temperature. At the same time, the input voltage amplitude fluctuation rate is detected, and the Kalman filter algorithm is used to predict the voltage sampling sequence to generate a voltage fluctuation prediction value.

[0154] S3: Locate the interval index in the inductor-capacitor parameter mapping table based on the voltage fluctuation prediction value, perform a least squares linear fitting operation on the inductor value to generate the real-time inductor parameter, and simultaneously perform the same operation on the capacitor value to generate the real-time capacitor parameter;

[0155] S4: For the instantaneous inductance parameters and the instantaneous capacitance parameters, the noise power spectrum density at the output of the filter is calculated by fast Fourier transform, and the difference between the 3dB attenuation point of the power spectrum density and the noise suppression bandwidth threshold is compared. If the difference exceeds the preset range, the parameter rollback instruction is triggered to regenerate the parameters. Otherwise, the effective inductance parameters and the effective capacitance parameters are output to the filter control unit.

[0156] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A communication line noise filtering system, characterized in that: The system comprises: A temperature coupling module is configured to collect real-time temperature values ​​of the inductor node and the capacitor node through dual-channel thermistors, input the real-time temperature values ​​of the inductor node and the capacitor node into a piecewise linear regression model to calculate the deviation gradient of the temperature to the resonant frequency, generate a temperature deviation gradient value, and transmit the temperature deviation gradient value and a preset resonant frequency reference table to the voltage response module; a voltage response module, configured to receive the temperature deviation gradient value and the resonant frequency reference table, match the noise suppression bandwidth threshold corresponding to the current temperature based on the resonant frequency reference table, detect the input voltage amplitude fluctuation rate, invoke a Kalman filter algorithm to process the voltage sampling sequence to generate a voltage fluctuation prediction value, and pass the result to the parameter joint adjustment module; A parameter coordination module is configured to receive the voltage fluctuation prediction value, extract an interval index from a preset inductance-capacitance parameter mapping table based on the voltage fluctuation prediction value, perform least squares linear fitting on the inductance value to generate an instantaneous inductance parameter, perform the same operation on the capacitance value to generate an instantaneous capacitance parameter, and pass the instantaneous inductance parameter and the instantaneous capacitance parameter to a link verification module.

2. The communication line noise filtering system according to claim 1, characterized in that: The temperature deviation gradient value includes a temperature measurement error correction factor, a resonant frequency change compensation coefficient, and a node thermal response characteristic parameter; the voltage fluctuation prediction value includes a voltage stability index, a voltage fluctuation trend parameter, and a voltage noise estimation value; the real-time inductance parameter includes a dynamic inductance adjustment coefficient, a real-time temperature drift calibration amount, and an inductance tolerance correction amount; the real-time capacitance parameter includes a dynamic capacitance adjustment coefficient, a real-time frequency drift calibration amount, and a capacitance tolerance correction amount.

3. The communication line noise filtering system according to claim 2, characterized in that: The segmented linear regression model dynamically divides segmented nodes based on the gradient change rate of the inductance and capacitance temperature intervals, and sets the weight coefficients of the linear regression in multiple intervals; The Kalman filter algorithm constructs a state equation by using the ambient temperature interference correction factor, the voltage attenuation dynamic coefficient and the voltage sequence dispersion index; The inductance-capacitance parameter mapping table divides the fluctuation range into upper and lower limits based on 1.2-1.5 times the average value of the voltage fluctuation rate.

4. The communication line noise filtering system according to claim 3, characterized in that: The temperature coupling module includes: The temperature acquisition submodule detects the temperature sensor signals of the inductor and capacitor nodes, uses a dual-channel signal synchronization acquisition circuit, sets the signal sampling interval and noise filtering threshold, performs amplitude calibration and timing alignment on the original signal, and records the inductor temperature value and capacitor temperature value separately, establishing two sets of real-time temperature time series data with synchronized timestamps to generate dual-channel temperature values. The regression processing submodule divides the inductor and capacitor temperature intervals into segmented nodes based on the dual temperature values, sets weight coefficients for linear regression within multiple intervals, inputs the inductor temperature value as an independent variable and the capacitor temperature value as a covariate into the model, calculates the slope change between the temperature and the resonant frequency within the multiple segmented intervals, extracts the slope parameter set of the segmented intervals, and generates a regression coefficient set; The segment nodes are dynamically adjusted when the gradient change rate of adjacent temperature sampling points exceeds 0.5°C / s; The gradient calculation submodule calls the regression coefficient set, selects the corresponding slope parameter in combination with the segmented interval to which the current dual-channel temperature value belongs, and multiplies the difference between the dual-channel temperature value and the reference frequency value by the slope parameter according to the temperature-frequency mapping relationship in the resonant frequency reference table. The module outputs the frequency offset caused by the unit temperature change to generate a temperature deviation gradient value.

5. The communication line noise filtering system according to claim 4, characterized in that: The voltage response module includes: The threshold matching submodule calls the temperature-frequency mapping relationship in the resonant frequency reference table based on the temperature deviation gradient value, performs cubic spline interpolation calculation on the current temperature value, matches the nearest neighbor reference frequency point, extracts the upper and lower limit calibration values ​​of the noise suppression bandwidth corresponding to the frequency point, and generates a bandwidth threshold matching value; The cubic spline interpolation satisfies the requirement that the first-order derivative at the node is continuous and the second-order derivative error is less than 0.1 Hz / °C; The fluctuation detection submodule detects the peak-to-valley difference of the input voltage signal amplitude, intercepts the sampling sequence with a fixed time window, calculates the absolute value of the amplitude difference of adjacent sampling points within the window, and divides the accumulated sum by the product of the time window length and the reference amplitude to generate the voltage fluctuation rate; The filter prediction submodule calls the bandwidth threshold matching value and the voltage fluctuation rate using the formula: ； The voltage fluctuation prediction component is obtained by calculation, and the prediction covariance matrix in the Kalman filter state equation is superimposed to generate the voltage fluctuation prediction value; in, represents the voltage fluctuation prediction component, Represents the temperature deviation gradient value, Represents the resonant frequency reference value, represents the noise suppression bandwidth threshold, represents the voltage fluctuation rate, Represents the ambient temperature interference correction factor, which is obtained by experimentally calibrating the contribution of temperature interference to voltage fluctuation. Represents the voltage attenuation dynamic coefficient, which is fitted through regression analysis of historical voltage data. Represents the voltage sequence dispersion index, which is the ratio of the standard deviation to the mean of the voltage sampling sequence.

6. The communication line noise filtering system according to claim 5, characterized in that: The parameter joint debugging module includes: The fluctuation interval index submodule collects the voltage fluctuation prediction value, compares the prediction value with the upper and lower limits of the fluctuation interval in the mapping table according to the interval division rules defined in the preset inductance-capacitance parameter mapping table, determines the index code of the interval to which the prediction value belongs, and generates a fluctuation interval index; The inductance parameter generation submodule calls the fluctuation interval index, extracts the inductance value set corresponding to the index in the mapping table, constructs a linear equation system with the index sequence as the independent variable and the inductance value as the dependent variable, calculates the minimum value of the residual sum of squares of the equation system through the least squares method, solves the slope and intercept parameters of the inductance value, and generates the real-time inductance parameter; The least squares method iteration termination condition is that the residual sum of squares is less than 0.01 or the number of iterations exceeds 50; The capacitance parameter generation submodule calls the fluctuation interval index, extracts the capacitance value set corresponding to the index in the mapping table, and based on the linear relationship between the index sequence and the capacitance value, applies the same least squares iterative process as the inductance parameter generation submodule to calculate the slope and intercept parameters of the capacitance value to generate the real-time capacitance parameter.

7. The communication line noise filtering system according to claim 6, characterized in that: The system further comprises: The link verification module is used to receive the real-time inductance parameter and the real-time capacitance parameter, calculate the noise power spectrum density at the filter output end through fast Fourier transform, compare the difference between the 3dB attenuation point of the power spectrum density and the noise suppression bandwidth threshold, and trigger a parameter rollback instruction if the difference exceeds the threshold. The instruction is fed back to the parameter joint adjustment module to regenerate the parameters, output the effective inductance parameter and the effective capacitance parameter, and load them into the filter control unit.

8. The communication line noise filtering system according to claim 7, characterized in that: The effective inductance parameters and effective capacitance parameters include noise suppression accuracy index, filter performance verification result, and link signal integrity evaluation value; The difference threshold is calibrated to ±10% of the noise suppression bandwidth threshold through spectrum analysis experiments.

9. The communication line noise filtering system according to claim 8, characterized in that: The link verification module includes: The noise spectrum calculation submodule calls the instantaneous inductance parameter and the instantaneous capacitance parameter to obtain a noise time domain signal sequence at the filter output end, applies a fast Fourier transform to convert the time domain signal into a frequency domain component, calculates the ratio of the square of the amplitude of the frequency domain component to the frequency bandwidth, and generates a noise power spectrum density; The threshold verification submodule locates the frequency point in the power spectrum where the amplitude attenuates to 70.7% of the maximum amplitude based on the noise power spectrum density as the 3dB attenuation point, extracts the preset noise suppression bandwidth threshold, calculates the absolute difference between the attenuation point and the threshold, compares the difference with the threshold allowable range, determines whether the difference exceeds the upper limit of the range, and generates a difference verification result; The threshold range is calibrated to [85%, 115%] through spectrum analysis experiments of 500 groups of noise samples; The parameter validation submodule calls the difference verification result and executes branch logic according to the over-limit status. If the limit is exceeded, the current parameter is cleared and the rollback instruction is triggered. If the limit is not exceeded, the parameter validation status is marked and the effective inductance parameter and the effective capacitance parameter are generated.

10. A communication line noise filtering method, characterized in that: The method is used to implement the communication line noise filtering system according to any one of claims 1 to 9, comprising the following steps: S1: Use dual temperature sensors to collect real-time temperature values ​​of the inductor node and capacitor node respectively. Input the two sets of temperature values ​​into the piecewise linear regression model to calculate the temperature deviation gradient of the resonant frequency and generate the temperature deviation gradient value. S2: Based on the temperature deviation gradient value, a preset resonant frequency reference table is searched to match the noise suppression bandwidth threshold corresponding to the current temperature. At the same time, the input voltage amplitude fluctuation rate is detected, and a Kalman filter algorithm is used to perform prediction processing on the voltage sampling sequence to generate a voltage fluctuation prediction value. S3: Locating an interval index in the inductance-capacitance parameter mapping table according to the voltage fluctuation prediction value, performing a least squares linear fitting operation on the inductance value to generate an instantaneous inductance parameter, and simultaneously performing the same operation on the capacitance value to generate an instantaneous capacitance parameter; S4: For the instantaneous inductance parameters and the instantaneous capacitance parameters, the noise power spectrum density at the output end of the filter is calculated by fast Fourier transform, and the difference between the 3dB attenuation point of the power spectrum density and the noise suppression bandwidth threshold is compared. If the difference exceeds the preset range, the parameter rollback instruction is triggered to regenerate the parameters; otherwise, the effective inductance parameters and the effective capacitance parameters are output to the filter control unit.

Images