Modeling method and circuit for interference injection in long-distance transmission
By using three definers for interference injection and BP neural network processing in long-term transmission, combined with Sage-Husa adaptive filtering algorithm, the problem of complex and insufficient applicability of signal interference modeling in long-term transmission in the existing technology is solved, and the problem of inadequate applicability of signal interference modeling in long-term transmission in the existing technology is realized. Accurate compensation and quality improvement of long-term transmission signals.
Patent Information
- Application Number
- CN202411050124.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2044-08-01
AI Technical Summary
The prior art is difficult to accurately model and process signal interference in long-line transmission. The traditional modeling method is complex and has a limited scope of application, making it difficult to effectively obtain and compensate interference information in long-line transmission.
Three types of definers (workload definer, fault load definer and interference load definer) are used for interference injection, combined with BP neural network to obtain mixed noise signals, and anti-interference processing is performed using Sage-Husa adaptive filtering algorithm, through Karl The Mann filter and the demodulator obtain the filtered signal to compensate for the attenuation information.
It realizes accurate acquisition and effective compensation of interference information in long-term transmission, improves the intuitiveness and effectiveness of signal processing, can effectively handle multiple noise interferences, and improves signal quality.
Smart Images

Figure CN118573230B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of signal transmission, and in particular to a modeling method and circuit for interference injection for long-line transmission. Background Art
[0002] In today's interconnected electronic devices, long-distance transmission technology has become an indispensable part of modern communication systems. However, in the process of long-distance transmission, due to factors such as long transmission distance and high transmission rate, signal interference has become an important factor affecting transmission quality. How to effectively model and process interference is one of the current research hotspots.
[0003] In the past, signal transmission modeling may use equivalent primal-dual circuit modeling and multi-harmonic distortion nonlinear model modeling. However, these traditional modeling methods have some problems in practical applications, such as:
[0004] 1) The inputs for modeling are relatively complex, and the algorithms used often have multiple outputs and are difficult to better reflect actual noise and interference;
[0005] 2) The scope of application is limited, and it is not very suitable for long-distance transmission, because the interference and various problems that may be encountered in long-distance transmission are more than those in ordinary short-distance transmission, and it is difficult to accurately obtain interference information to compensate for it. Summary of the invention
[0006] In view of the above two problems, the present invention aims to propose a modeling method and circuit for interference injection for long-line transmission, which uses three definers for interference injection, fully inputs the problems that may be encountered in long-line transmission, and then combines the BP neural network to obtain a mixed noise signal, and simplifies multiple noise mixtures into one noise, which is intuitive, effective and easy to handle; at the same time, the Sage-Husa adaptive filtering algorithm is used for anti-interference processing to obtain a filtered signal, so that in the end it can not only effectively obtain the attenuation information of long-line transmission but also perform effective filtering compensation for long-line transmission.
[0007] This is achieved through the following technical solutions:
[0008] In the first aspect, a modeling method for interference injection for long-line transmission is proposed, and the method includes the following steps: S1, setting a signal generator for generating an initial signal for long-line transmission; constructing a twisted pair transmission line model, and importing the initial signal into the transmission line model for transmission; setting an interference injection module, the interference injection module is used to perform interference injection on the initial signal of transmission, and the interference injection module includes three parallel definers, namely a workload definer, a fault load definer and an interference load definer, the workload definer sets a workload parameter for the initial signal of transmission, the fault load definer sets a fault sequence for the initial signal of transmission, and the interference load definer sets an interference sequence for the initial signal of transmission; S2, using QPSK and 16QAM to set a communication interference signal model for the initial signal in step S1, the communication interference signal model generates a harmonic interference sub-model, a noise interference sub-model and a length sub-model in the model based on the workload parameters, fault sequence and interference sequence in step S1; wherein the harmonic interference sub-model is used to simulate the electromagnetic interference signal for long-line transmission ; The noise interference sub-model is used to simulate the interference of noise on long-line transmission. The noise in the noise interference sub-model includes pink noise, pulse noise and white noise. The BP neural network is used to take pink noise, pulse noise and white noise as input for mixed calculation to obtain a mixed noise signal; the length sub-model is used to simulate the change in the length of the transmission line during long-line transmission; S3, using the communication interference signal model of step S2 to start interference injection into the twisted pair transmission line model, including adjusting the length of the transmission line and inputting the electromagnetic interference signal and the mixed noise signal; setting a Kalman filter and a demodulator, the Kalman filter uses the Sage-Husa adaptive filtering algorithm to filter the electromagnetic interference signal, the mixed noise signal and the initial signal in step S2 to obtain a filtered signal; the demodulator demodulates the filtered signal to obtain a modulated signal; continue to set a signal processing module and an analysis module, first use the signal processing module to process the modulated signal to obtain a processing result, and then use the analysis module to analyze the processing result and the length of the transmission line to obtain the attenuation data of the long-line transmission and complete the modeling.
[0009] The use of twisted pair cables for transmission can have a certain shielding effect and reduce the interference during long-line transmission. In addition, the three parallel definers of the interference injection module are used for interference injection, which can include as many problems as possible in long-line transmission, thereby improving the accuracy of the final output. At the same time, setting up multiple noise models and combining the Sage-Husa adaptive filtering algorithm with the BP neural network can simplify a variety of complex noise problems and obtain accurate noise information, thereby obtaining accurate attenuation data and compensating for the noise information.
[0010] Preferably, in step S2, the communication interference signal model also generates an impedance matching sub-model in the model, and the impedance matching sub-model is used to simulate the impedance matching between the load and the long line during long-line transmission. The impedance matching sub-model can effectively simulate the impedance matching problem, reduce the signal reflection caused by impedance mismatch, and avoid the signal reflection affecting the signal quality.
[0011] Preferably, in step S1, the twisted pair used in the twisted pair transmission line model is an STP twisted pair, an FTP twisted pair, or an SFTP twisted pair. STP twisted pair, FTP twisted pair, and SFTP twisted pair can all reduce electromagnetic interference and signal attenuation.
[0012] Preferably, in step S1, the twisted pair used in the twisted pair transmission line model includes a conductor material and an insulating material, wherein the conductor material is an aluminum-steel alloy conductor or a copper conductor, and the insulating material is a quartz glass fiber manufactured under high pressure. The quartz glass fiber manufactured under high pressure has better high-voltage resistance and heat resistance than ordinary quartz glass fibers, and the aluminum-steel alloy conductor has the characteristics of being cheap, stable, and low attenuation, and is suitable for long-distance transmission.
[0013] Preferably, in step S1, the workload parameters include the long-line transmission length, the type of the initial signal and the duration of the transmission, the fault sequence includes packet loss and garbled code, and the interference sequence includes electromagnetic interference, power supply noise, conduction interference and transmission impedance interference. The workload parameters, fault sequence and interference sequence are all transmission problems often encountered in long-line transmission. Targeted input can effectively simulate the attenuation problem caused by transmission.
[0014] Preferably, in step S2, the BP neural network adopts a 3-7-1 structure, including 3 input layers, 7 hidden layers and 1 output layer, wherein the 3 input layers are used to input noise, the 7 hidden layers are 7 computing neurons for calculation, and the 1 output layer is used to output mixed noise signals. When performing calculations, the BP neural network has strong adaptability, fault tolerance and fast calculation speed, and can also output multiple inputs in the form of a single output.
[0015] Preferably, in step S3, the steps of using the Sage-Husa adaptive filtering algorithm include: S31, initialization: determining the initial state based on the mixed signal formed by the electromagnetic interference signal, the mixed noise signal and the initial signal, and then establishing the system equation of the state covariance matrix according to the initial state; S32, prediction: performing target estimation on the system equation in step S31 without observation; S33, update: observing the mixed signal once, obtaining the observation data and correcting the result of the target estimation in step S32 according to the observation data to obtain the optimal state estimation value and the state covariance matrix; S34, Sage-Husa improvement: using a time-varying noise statistical estimator with Sage-Husa adaptive filtering function, the state estimation value and the state covariance matrix in step S33 are statistically characterized and corrected in real time to obtain the filtering result; S35, iteration stage: judging whether the filtering result in step S34 meets the convergence condition of the system equation, if so, outputting the filtered signal based on the filtering result; if not, repeating steps S32 to S34 in sequence until the convergence condition is met or the upper limit of the number of iterations is reached. The Sage-Husa adaptive filtering algorithm is an improved algorithm used in the Kalman filter, which can automatically adjust the parameters of the filter to achieve effective filtering of mixed noise signals.
[0016] In the second aspect, a circuit for interference injection for long-distance transmission is also proposed, which is connected by the above-mentioned modeling method. The circuit includes a signal generator, a twisted pair, an interference injection module, a Kalman filter, a demodulator, a signal processor and an analysis module connected in sequence; wherein the signal generator generates an initial signal; the twisted pair transmits the initial signal; the interference injection module uses three parallel workload definers, a fault load definer and an interference load definer to perform interference injection on the initial signal transmitted by the twisted pair; the Kalman filter uses the Sage-Husa adaptive filtering algorithm to perform anti-interference calculation on the initial signal after interference injection to obtain a filtered signal; the demodulator demodulates the filtered signal to obtain a modulated signal; the signal processor processes the modulated signal to obtain a processing result; the analysis module analyzes the processing result to obtain the attenuation data of the long-distance transmission. The effective effect of the circuit can refer to the above-mentioned modeling method, which will not be repeated here.
[0017] Compared with the prior art, the present invention has the following beneficial effects:
[0018] The technical solution of the present invention utilizes three definers to perform interference injection, fully inputs the problems that may be encountered in long-line transmission, and then combines the BP neural network to obtain a mixed noise signal, thereby simplifying multiple noise mixtures into one noise, which is intuitive, effective and easy to handle; at the same time, the Sage-Husa adaptive filtering algorithm is used for anti-interference processing to obtain a filtered signal, so that in the end, the attenuation information of long-line transmission can be effectively obtained and effective filtering compensation can be performed for long-line transmission. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 A flow chart of a method for modeling interference injection for long line transmission;
[0020] Figure 2 The present invention is a schematic block diagram of a circuit structure for interference injection in long-line transmission. DETAILED DESCRIPTION
[0021] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.
[0022] like Figure 1 FIG. 1 is a flow chart of a modeling method for interference injection for long-distance transmission, and the modeling method includes the following steps:
[0023] S1. Set a signal generator to generate the initial signal for long-line transmission. Then, construct a twisted pair transmission line model, such as a shielded twisted pair RLCG model, etc. The twisted pair transmission line model is used to import the initial signal into the transmission line model for transmission. Continue to set an interference injection module to inject interference into the initial signal of transmission. The interference injection module includes three parallel definers, namely, a workload definer, a fault load definer and an interference load definer. The workload definer sets the workload parameters for the initial signal of transmission, the fault load definition sets the fault sequence for the initial signal of transmission, and the interference load definer sets the interference sequence for the initial signal of transmission.
[0024] In this embodiment, in step S1, the workload parameters include the long-line transmission length, the type of initial signal and the duration of the transmission. In actual long-line transmission, the transmission line often has impedance, which will affect the quality of the transmitted signal. Adjusting the transmission length can adjust the impedance of the transmission line. Fault sequences include packet loss and garbled characters, which are common transmission problems in transmission. Interference sequences include electromagnetic interference, power supply noise, conducted interference and transmission impedance interference, which are also various problems that may be encountered in transmission. The duration of the transmission corresponds to when the signal transmission starts, and is also used to reflect the temperature changes caused by time changes. This temperature change may affect the impedance of the transmission line.
[0025] In this embodiment, in step S1, the twisted pair used in the twisted pair transmission line model is an STP twisted pair, an FTP twisted pair, or an SFTP twisted pair. The STP twisted pair is a shielded twisted pair with superior performance, supports long-distance high-bitrate transmission, and can reduce external interference that may occur during actual transmission; the FTP twisted pair is an aluminum foil shielded twisted pair, which is a shielding layer wrapped with aluminum foil outside multiple pairs of twisted pairs, and also has good anti-interference ability; the SFTP twisted pair is a double-shielded twisted pair, which is an STP twisted pair with an additional braided layer to form a better double-layer shielding structure, which is suitable for scenes with severe electromagnetic interference; the use of these three twisted pairs can reduce electromagnetic interference and signal attenuation.
[0026] In this embodiment, in step S1, the twisted pair used in the twisted pair transmission line model includes a conductor material and an insulating material, wherein the conductor material is an aluminum-steel alloy wire or a copper wire, and the insulating material is a quartz glass fiber manufactured under high pressure. The quartz glass fiber manufactured under high pressure has better high-voltage resistance and heat resistance than ordinary quartz glass fibers. The copper wire is a commonly used transmission wire, and the aluminum-steel alloy wire has the characteristics of being cheap, stable, and low attenuation, and is also suitable for long-distance transmission.
[0027] S2. Use QPSK and 16QAM to set a communication interference signal model for the initial signal in step S1. QPSK and 16QAM are two common signal modulation methods. The communication interference signal model generates a harmonic interference sub-model, a noise interference sub-model and a length sub-model in the model based on the workload parameters, fault sequence and interference sequence in step S1. Among them, the harmonic interference sub-model is used to simulate the electromagnetic interference signal to long-line transmission; the noise interference sub-model is used to simulate the interference of noise to long-line transmission. The noise in the noise interference sub-model includes pink noise, pulse noise and white noise. The BP neural network is used to mix and calculate the pink noise, pulse noise and white noise as input to obtain a mixed noise signal; the length sub-model is used to simulate the length change of the transmission line during long-line transmission.
[0028] The noise interference sub-model includes a pink noise model, an impulse noise model and a white noise model, which input pink noise, impulse noise and white noise respectively. White noise model: White noise is a type of noise whose power spectrum density is uniformly distributed in the entire frequency domain. When modeling, it can be assumed that the noise signal is a random process with unit variance and zero mean, and its power spectrum density is constant at all frequencies. The white noise model is suitable for describing many random interference sources, such as thermal noise, shot noise, etc. White noise is usually used to simulate random thermal noise or other uniformly distributed interference sources in circuits. Pink noise model: Pink noise is a type of noise whose power spectrum density gradually decreases with increasing frequency. The power spectrum density curve of pink noise shows a downward trend, which is similar to the frequency response characteristics of the human auditory system. The pink noise model is suitable for describing certain interference sources with specific frequency characteristics, such as the internal noise of certain electronic equipment. Pulse noise small model: Pulse noise is composed of a series of discrete, short-duration pulse signals. Pink noise is often used to simulate interference inside certain electronic devices, especially those with frequency characteristics. The pulse noise small model can describe the interference signals caused by certain emergencies in the electrical system, such as power supply fluctuations, electromagnetic interference, and device failures. When modeling, it can be assumed that the pulse noise signal is composed of a series of randomly distributed pulses, each of which has a specific amplitude, width, and arrival time.
[0029] In this embodiment, in step S2, the BP neural network adopts a 3-7-1 structure, including 3 input layers, 7 hidden layers and 1 output layer. The 3 input layers input pink noise, impulse noise and white noise respectively, and the 7 hidden layers are 7 computing neurons. The 7 computing neurons ensure the accuracy of the calculation without causing the calculation to be too complicated. The 1 output layer is used to output the mixed noise signal, thereby simplifying the form of multiple noises. When performing calculations, the BP neural network has strong adaptability, fault tolerance and fast calculation speed, and can also output multiple inputs in the form of a single output.
[0030] Specifically, in the BP neural network, the input layer uses j input neurons, the size of j is 3, and the output function used by the input layer is , x represents the neurons in the input layer, and the superscript (1) here represents the input layer. The hidden layer receives the output function of the input layer, and the input function of the hidden layer is , k and i are both positive integers not greater than j, k represents the neurons in the hidden layer, is the weight coefficient, and the superscript (2) indicates the hidden layer. The output function used by the hidden layer is , t is a positive integer greater than k; the output layer receives the output function of the hidden layer, and the input function of the output layer is , the superscript (3) indicates the output layer, q indicates the neurons in the output layer, and the value of q is 1, which corresponds to an output layer. The output function of the output layer is ,in, Represents a mixed noise signal. When performing calculations, the BP neural network has strong adaptability, fault tolerance and fast calculation speed, and ultimately transmits complex multiple inputs in the form of a single output.
[0031] In addition, when the corresponding function calculation process is executed in the input layer, hidden layer and output layer, the steepest descent method can be used to adjust the parameters of the corresponding function calculation process, because the steepest descent method can accelerate the convergence speed and minimize the error, thereby improving the effectiveness and accuracy.
[0032] In this embodiment, in step S2, the communication interference signal model also generates an impedance matching sub-model in the model, and the impedance matching sub-model is used to simulate the impedance matching between the load and the long line during long-line transmission. If the impedance between the signal and the load does not match, the signal energy will be reflected during the transmission process to form a reflected wave, thereby reducing the signal quality, causing signal distortion, and reducing the signal-to-noise ratio.
[0033] S3. Set a Kalman filter and a demodulator. The Kalman filter uses the Sage-Husa adaptive filtering algorithm to filter the electromagnetic interference signal, the mixed noise signal and the initial signal in step S2 to obtain a filtered signal. The demodulator demodulates the filtered signal to obtain a modulated signal. Continue to set a signal processing module and an analysis module. First, use the signal processing module to process the modulated signal to obtain a processing result. Then use the analysis module to analyze the processing result and the length of the transmission line to obtain the attenuation data of the long-line transmission and complete the modeling.
[0034] In this embodiment, in step S3, the Sage-Husa adaptive filtering algorithm is a publicly available algorithm, and the application of the Sage-Husa adaptive filtering algorithm here can provide a step for reference only, including: S31, initialization: based on the mixed signal formed by the electromagnetic interference signal, the mixed noise signal and the initial signal, determine the initial state, and then establish the system equation of the state covariance matrix according to the initial state; S32, prediction: without observation, perform target estimation on the system equation in step S31; S33, update: observe the mixed signal once, obtain observation data and correct the result of the target estimation in step S32 according to the observation data to obtain the optimal state estimation value and state covariance matrix; S34, Sage-Husa improvement: using a time-varying noise statistical estimator with Sage-Husa adaptive filtering function, statistically characterize and modify the state estimate and state covariance matrix in step S33 in real time to obtain the filtering result; S35, iteration phase: determine whether the filtering result in step S34 meets the convergence condition of the system equation. If so, output the filtered signal based on the filtering result; if not, repeat steps S32 to S34 in sequence until the convergence condition is met or the upper limit of the number of iterations is reached. When the upper limit of the number of iterations is reached, you can choose to return to step S31 for recalculation or directly output the current result. The Sage-Husa adaptive filtering algorithm is an improved algorithm used in the Kalman filter, which can automatically adjust the parameters of the filter, achieve effective filtering, and improve the accuracy of the calculation.
[0035] like Figure 2 As shown, it is a circuit structure diagram for interference injection for long-distance transmission. The circuit is connected by the above-mentioned modeling method. The circuit includes a signal generator, a twisted pair, an interference injection module, a Kalman filter, a demodulator, a signal processor and an analysis module connected in sequence. Among them, the signal generator generates an initial signal; the twisted pair transmits the initial signal; the interference injection module uses three parallel workload definers, a fault load definer and an interference load definer to perform interference injection on the initial signal transmitted by the twisted pair; the Kalman filter uses the Sage-Husa adaptive filtering algorithm to perform anti-interference calculation on the initial signal after interference injection to obtain a filtered signal; the demodulator demodulates the filtered signal to obtain a modulated signal; the signal processor processes the modulated signal to obtain a processing result; the analysis module analyzes the processing result to obtain the attenuation data of the long-distance transmission. The effective effect of the circuit can refer to the above-mentioned modeling method, which will not be repeated here.
[0036] In summary, the present invention utilizes three definers for interference injection, fully inputs the problems that may be encountered in long-line transmission, and then combines the BP neural network to obtain a mixed noise signal, simplifying multiple noise mixtures into one noise, which is intuitive, effective and easy to handle; at the same time, the Sage-Husa adaptive filtering algorithm is used for anti-interference processing to obtain a filtered signal, so that in the end, the attenuation information of long-line transmission can be effectively obtained and effective filtering compensation can be performed for long-line transmission, which is significantly advanced.
[0037] The above embodiments are only for illustrating the technical idea of the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.
Claims
1. A modeling method for interference injection for long-distance transmission, characterized in that: The steps include: S1. Set a signal generator to generate an initial signal for long-line transmission; construct a twisted pair transmission line model, and import the initial signal into the transmission line model for transmission; set an interference injection module, which is used to inject interference into the initial signal for transmission. The interference injection module includes three parallel definers, namely, a workload definer, a fault load definer, and an interference load definer. The workload definer sets workload parameters for the initial signal for transmission, the fault load definer sets a fault sequence for the initial signal for transmission, and the interference load definer sets an interference sequence for the initial signal for transmission; S2, using QPSK and 16QAM to set a communication interference signal model for the initial signal in step S1, the communication interference signal model generates a harmonic interference sub-model, a noise interference sub-model and a length sub-model in the model based on the workload parameters, fault sequence and interference sequence in step S1; Among them, the harmonic interference sub-model is used to simulate the electromagnetic interference signal of long-line transmission; the noise interference sub-model is used to simulate the interference of noise on long-line transmission. The noise in the noise interference sub-model includes pink noise, impulse noise and white noise. The BP neural network is used to mix and calculate the pink noise, impulse noise and white noise as input to obtain the mixed noise signal; the length sub-model is used to simulate the length change of the transmission line during long-line transmission; S3, using the communication interference signal model of step S2 to start interference injection into the twisted pair transmission line model, including adjusting the length of the transmission line and inputting the electromagnetic interference signal and the mixed noise signal; Set up a Kalman filter and a demodulator. The Kalman filter uses the Sage-Husa adaptive filtering algorithm to filter the electromagnetic interference signal, the mixed noise signal and the initial signal to obtain a filtered signal. The demodulator demodulates the filtered signal to obtain a modulated signal. Continue to set up a signal processing module and an analysis module. First, use the signal processing module to process the modulated signal to obtain a processing result. Then use the analysis module to analyze the processing result and the length of the transmission line to obtain the attenuation data of long-line transmission and complete the modeling.
2. A modeling method for interference injection for long-distance transmission according to claim 1, characterized in that: In step S2, the communication interference signal model also generates an impedance matching sub-model within the model, and the impedance matching sub-model is used to simulate the impedance matching between the load and the long line during long-line transmission.
3. A modeling method for interference injection for long-distance transmission according to claim 1, characterized in that: In step S1, the twisted pair used in the twisted pair transmission line model is an STP twisted pair, an FTP twisted pair, or an SFTP twisted pair.
4. A modeling method for interference injection for long-distance transmission according to claim 1, characterized in that: In step S1, the twisted pair used in the twisted pair transmission line model includes a conductor material and an insulating material. The conductor material uses an aluminum-steel alloy conductor or a copper conductor, and the insulating material uses a quartz glass fiber manufactured under high voltage.
5. A modeling method for interference injection for long-distance transmission according to claim 1, characterized in that: In step S1, the workload parameters include the long-line transmission length, the type of the initial signal and the duration of the transmission, the fault sequence includes packet loss and garbled code, and the interference sequence includes electromagnetic interference, noise, conducted interference and transmission impedance interference.
6. A modeling method for interference injection for long-distance transmission according to claim 1, characterized in that: In step S2, the BP neural network adopts a 3-7-1 structure, including 3 input layers, 7 hidden layers and 1 output layer. The 3 input layers are used to input noise, the 7 hidden layers are 7 computing neurons for calculation, and the 1 output layer is used to output mixed noise signals.
7. A modeling method for interference injection for long-distance transmission according to claim 1, characterized in that: In step S3, the steps of using the Sage-Husa adaptive filtering algorithm include: S31, initialization: determining an initial state based on a mixed signal formed by the electromagnetic interference signal, the mixed noise signal and the initial signal, and then establishing a system equation of a state covariance matrix according to the initial state; S32, prediction: without observation, target estimation is performed on the system equation in step S31; S33, update: observe the mixed signal once, obtain observation data and correct the result of target estimation in step S32 according to the observation data to obtain the optimal state estimation value and state covariance matrix; S34, Sage-Husa improvement: using a time-varying noise statistical estimator with Sage-Husa adaptive filtering function, statistically characterize and modify the state estimation value and state covariance matrix in step S33 in real time to obtain a filtering result; S35, iteration stage: determine whether the filtering result in step S34 meets the convergence condition of the system equation. If so, output the filtering signal based on the filtering result; if not, repeat steps S32 to S34 in sequence until the convergence condition is met or the upper limit of the number of iterations is reached.
8. A circuit for interference injection for long-line transmission, connected by the modeling method according to any one of claims 1 to 7, characterized in that: It includes a signal generator, a twisted pair, an interference injection module, a Kalman filter, a demodulator, a signal processor and an analysis module connected in sequence; Among them, the signal generator generates an initial signal; the twisted pair transmits the initial signal; the interference injection module uses three parallel workload definers, fault load definers and interference load definers to inject interference into the initial signal transmitted by the twisted pair; the Kalman filter uses the Sage-Husa adaptive filtering algorithm to perform anti-interference calculation on the initial signal after interference injection to obtain a filtered signal; the demodulator demodulates the filtered signal to obtain a modulated signal; the signal processor processes the modulated signal to obtain a processing result; the analysis module then analyzes the processing result to obtain the attenuation data of long-line transmission.
