Signal processing method and system based on power MOS tube

Abstract

The invention discloses a signal processing method and system based on a power MOS tube, and relates to the technical field of signal processing, and the method comprises the steps: collecting an input signal of an external environment, and monitoring an electrical characteristic parameter in real time; performing preliminary filtering processing on an input signal of an external environment to generate an original signal; the preliminary stable signal is transmitted to a multi-parameter feedback control unit, the working state of the MOS tube is adjusted by monitoring electrical characteristic parameters in real time, and an accurate signal is output; performing predictive optimization adjustment on the accurate signal by adopting an intelligent algorithm, processing the accurate signal through a last-stage power MOS tube, and outputting an optimized stable signal; according to the invention, predictive optimization adjustment of the signal is carried out by using an intelligent prediction algorithm, adaptive adjustment can be carried out according to historical data and a current signal trend, the problem of excessive or insufficient adjustment of the signal is avoided, and the signal processing capability in a complex environment is significantly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and particularly to a signal processing method and system based on a power MOS transistor. Background Art

[0002] As an important semiconductor device, the power MOS transistor is widely used in the field of power electronics, especially in applications such as high-efficiency power conversion, switching circuits, and signal processing, with high switching speed, low power loss, and excellent control characteristics. With the rapid development of industrial automation, smart grids, communication technologies, and Internet of Things devices, higher requirements are being placed on the accuracy and stability of signal processing. The application of power MOS transistors is not limited to the traditional power conversion field, and in recent years, their applications in complex signal processing, modulation and demodulation, and intelligent control systems have gradually increased. Especially in scenarios where signals need to be adjusted and optimized in real time, by leveraging the high-speed switching characteristics of power MOS transistors, efficient signal processing can be achieved. However, in the existing technologies, signal processing methods based on power MOS transistors usually only rely on simple filtering and switching control, making it difficult to handle complex and variable external environmental signals and lacking the ability to adaptively optimize signals.

[0003] In the existing related technologies, the operating state of the power MOS transistor is usually controlled by a fixed threshold voltage, lacking a dynamic response mechanism, which is a core bottleneck in improving signal processing performance; due to the inability to adaptively adjust according to the real-time changes of the input signal, the switching state of the power MOS transistor is prone to lag or mismatch when processing complex signals, resulting in poor stability of the output signal; this static control method is difficult to cope with the rapid changes of external signals, especially in applications such as industrial control and communication systems where high signal accuracy is required, and the control strategy of the fixed threshold voltage will limit the dynamic response ability of the system, ultimately affecting the signal processing effect; this deficiency makes it difficult for the existing technologies to meet the requirements for signal processing accuracy and stability in high-precision and complex environments. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a signal processing method based on a power MOS transistor to solve the problems of the lack of a dynamic response mechanism and adaptive adjustment of the power MOS transistor in the existing technologies.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a signal processing method based on a power MOS transistor, which includes: collecting an input signal of the external environment and real-time monitoring of electrical characteristic parameters;

[0008] Perform preliminary filtering on the input signal from the external environment to generate the original signal;

[0009] Transmit the original signal to the multi-layer drive architecture, dynamically adjust the switching state of the power MOS transistor, and output the preliminary stable signal;

[0010] Transmit the preliminary stable signal to the multi-parameter feedback control unit, use the real-time monitored electrical characteristic parameters to adjust the working state of the MOS transistor, and output the accurate signal;

[0011] Adopt an intelligent algorithm to perform predictive optimization adjustment on the accurate signal, and process it through the last-stage power MOS transistor to output the optimized stable signal.

[0012] As a preferred solution of the signal processing method based on power MOS transistor according to the present invention, wherein: the input signal includes: analog signal, digital signal and communication signal;

[0013] The electrical characteristic parameters include: electrical parameters, environmental parameters, state parameters and dynamic parameters.

[0014] As a preferred solution of the signal processing method based on power MOS transistor according to the present invention, wherein: the performing preliminary filtering on the input signal from the external environment to generate the original signal, the specific steps are as follows:

[0015] For the analog signal, convert it into a digital signal through an analog-to-digital converter;

[0016] For the communication signal, demodulate it into a baseband signal;

[0017] Estimate the power spectral density of the digital signal and the baseband signal through fast Fourier transform, and identify the frequency distribution of the noise;

[0018] According to the frequency distribution of the noise, apply a multi-stage filtering structure to gradually purify the signal;

[0019] For the purified signal, perform inverse Fourier transform from the frequency domain back to the time domain to generate the original signal.

[0020] As a preferred solution of the signal processing method based on power MOS transistor according to the present invention, wherein: the transmitting the original signal to the multi-layer drive architecture, dynamically adjusting the switching state of the power MOS transistor, and outputting the preliminary stable signal, the specific steps are as follows,

[0021] Use a laser source to generate a high-frequency optical signal to modulate the original signal and convert the original signal into an optical signal;

[0022] Transmit the modulated optical signal to the multi-layer drive unit through an optical fiber, and perform smoothing processing on the signal using Gaussian convolution, and the expression is: Among them, represents the optical signal output after multi-level modulation, represents the input optical signal, is the time variable, is the time delay variable, represents the Gaussian convolution kernel, represents the exponential part of the Gaussian function is the standard deviation of the Gaussian distribution, represents the differential element for integrating ;

[0023] The optical signal output after multi-level modulation is converted into an electrical signal through an optoelectronic conversion module, and the expression is:

[0024] Among them, is the electrical signal after optoelectronic conversion, is the optoelectronic responsivity, is the optoelectronic conversion efficiency;

[0025] Using an adaptive filtering model, the change of the electrical signal after optoelectronic conversion is monitored in real time, and the driving voltage of the gate is dynamically adjusted. The expression is: G (t)= 1 2 [1+ tanh (α⋅ ∫ 0 t E ( t ' )d t ' )] Among them, represents the gate voltage control signal, represents the hyperbolic tangent function, represents the gain coefficient of signal regulation, represents the electrical signal on the past time variable ; is accumulated, represents the electrical signal of the past time variable ;

[0026] The gate voltage control signal is applied to the gate of the MOS transistor to control the on and off states, and a preliminarily stable electrical signal is output.

[0027] As a preferred embodiment of the signal processing method based on power MOS transistors according to the present invention, wherein: the preliminarily stabilized signal is transmitted to the multi-parameter feedback control unit, and the working state of the MOS transistor is adjusted by real-time monitoring of electrical characteristic parameters to output an accurate signal. The specific steps are as follows: Integrate the real-time monitored electrical parameters, environmental parameters, state parameters, and dynamic parameters into a multi-dimensional data matrix; Input the multi-dimensional data matrix into a hybrid model of a convolutional neural network and a long short-term memory neural network for prediction; According to the prediction result, use an adaptive fuzzy controller to dynamically adjust the gate voltage of the MOS transistor, and at the same time, based on the recursive least squares method, real-time adjust the gain bits and gain thresholds of the controller; Output an accurate gate voltage signal based on the adjusted adaptive fuzzy controller.

[0028] As a preferred embodiment of the signal processing method based on power MOS transistors according to the present invention, wherein: based on the recursive least squares method, real-time adjust the gain bits and gain thresholds of the controller. The specific steps are as follows: Initialize the gain bits and gain thresholds as the initial values of the recursive least squares method; Collect the error signal between the control output of the power MOS transistor and the target signal; According to the error signal, use the recursive least squares method to real-time update the gain bits and gain thresholds of the controller to minimize the error.

[0029] As a preferred embodiment of the signal processing method based on power MOS transistors according to the present invention, wherein: use an intelligent algorithm to perform predictive optimization adjustment on the accurate signal, and process it through the last-stage power MOS transistor to output an optimized stable signal. The specific steps are as follows: Process the noise of the signal through a Bayesian filter; Based on the signal processed by the Bayesian filter, perform time series optimization using a long short-term memory neural network; Based on the time series optimized signal output by the long short-term memory neural network, apply a generalized autoregressive conditional heteroskedasticity model to predict the volatility of the signal; Perform global optimization on the processing results of the Bayesian filter, long short-term memory neural network time series optimization, and generalized autoregressive conditional heteroskedasticity model volatility detection through a quantum optimization algorithm to output parameter settings; According to the parameter settings output by the quantum optimization algorithm, dynamically adjust the gate voltage of the last-stage power MOS transistor to control its on and off states, and finally output an optimized stable electrical signal.

[0030] Second aspect, the present invention provides a signal processing system based on a power MOS transistor, including: an acquisition module, a filtering processing module, a dynamic adjustment module, a feedback control module, and a signal output module; The acquisition module is used to acquire an input signal of the external environment and monitor electrical characteristic parameters in real time; The filtering processing module is used to perform preliminary filtering processing on the input signal of the external environment to generate a raw signal; The dynamic adjustment module is used to transmit the raw signal to a multi-layer driving architecture, dynamically adjust the switching state of the power MOS transistor, and output a preliminary stable signal; The feedback control module is used to transmit the preliminary stable signal to a multi-parameter feedback control unit, adjust the working state of the MOS transistor by using the real-time monitored electrical characteristic parameters, and output an accurate signal; The signal output module is used to perform predictive optimization adjustment on the accurate signal by using an intelligent algorithm, and process it through the last-stage power MOS transistor to output an optimized stable signal.

[0031] Third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the signal processing method based on a power MOS transistor as described in the first aspect of the present invention is implemented.

[0032] Fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the signal processing method based on a power MOS transistor as described in the first aspect of the present invention is implemented.

[0033] The beneficial effects of the present invention are as follows: The present invention adopts a multi-layer driving architecture and an intelligent algorithm to dynamically adjust the switching state of the power MOS transistor, can monitor and feedback the electrical characteristics of the input signal in real time, thereby improving the accuracy and stability of signal processing; through a multi-stage filtering structure, high-frequency noise and transient interference are effectively suppressed, and the signal purification effect is enhanced; at the same time, using an intelligent prediction algorithm to perform predictive optimization adjustment of the signal, it can perform adaptive adjustment according to historical data and the current signal trend, avoiding the problem of over-adjustment or under-adjustment of the signal, and significantly improving the signal processing ability in a complex environment. Description of the Drawings

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0035] Figure 1 It is a flowchart of the signal processing method based on a power MOS transistor in Embodiment 1.

[0036] Figure 2 It is a module diagram of the signal processing system based on a power MOS transistor in Embodiment 1. Detailed implementation manners

[0037] To make the above objects, features, and advantages of the present invention more apparent and understandable, the following will describe the detailed implementation manners of the present invention with reference to the accompanying drawings of the specification.

[0038] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0039] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments.

[0040] Embodiment 1, referring to Figure 1 and Figure 2 , which is the first embodiment of the present invention. This embodiment provides a signal processing method based on a power MOS transistor, including the following steps:

[0041] S1. Collect the input signals of the external environment and monitor the electrical characteristic parameters in real time.

[0042] Furthermore, initialize the ADC unit for analog signals, the digital interface for digital signals, and the communication interface for communication signals;

[0043] Use the ADC to collect analog signals (such as voltage and current) in real time;

[0044] Collect digital signals (such as switch states) through the digital interface;

[0045] Receive communication signals (such as sensor data) through the communication interface;

[0046] Monitor electrical parameters, environmental parameters, state parameters, and dynamic parameters in real time through the collected analog and digital signals;

[0047] It should be noted that high-precision sensors and data processing should be used during the real-time monitoring process to ensure the accuracy and timeliness of data collection.

[0048] S2. Perform preliminary filtering on the input signal from the external environment to generate the original signal.

[0049] Furthermore, for analog signals, convert them into digital signals through an analog-to-digital converter (ADC) to facilitate subsequent digital filtering;

[0050] For communication signals, demodulate them into baseband signals to concentrate their spectra in lower frequency bands for subsequent processing;

[0051] Estimate the power spectral density (PSD) of the digital signal and the baseband signal through fast Fourier transform (FFT) to identify the frequency distribution of the noise;

[0052] According to the frequency distribution of the noise, apply a multi-stage filtering structure to gradually purify the signal as follows:

[0053] The first stage is a fixed-parameter filter, which uses a low-pass filter to remove the identified high-frequency noise and ensure the basic stability of the signal.

[0054] The second stage is an adaptive filter, which uses an LMS adaptive filter to dynamically adjust the filter coefficients to further optimize the signal and process non-linear and time-varying noise.

[0055] The third stage: a band-pass filter, which uses a band-pass filter to extract the effective signal according to the signal characteristics of a specific frequency band and further enhance the useful part of the signal.

[0056] Among them, usually first use a low-pass filter to remove high-frequency noise, and then combine an adaptive filter to dynamically adjust the filtering effect to ensure that the purified signal is more accurate.

[0057] S3. Transmit the original signal to a multi-layer drive architecture, dynamically adjust the switching state of the power MOS transistor, and output a preliminary stable signal.

[0058] Furthermore, use a laser source to generate a high-frequency optical signal to modulate the original signal and convert the original signal into an optical signal;

[0059] Transmit the modulated optical signal through an optical fiber to a multi-layer drive unit, and perform smoothing processing on the signal using Gaussian convolution. The expression is: , where represents the optical signal output after multi-layer modulation, represents the input optical signal, is the time variable, is the time delay variable, represents the Gaussian convolution kernel, which is used to perform smoothing processing on the signal, Represents the exponential part of the Gaussian function, which is used to control the shape of the convolution kernel. Is the standard deviation of the Gaussian distribution. Represents the differential element for integrating ;

[0060] The optical signal output after multi-layer modulation is converted into an electrical signal through an optoelectronic conversion module. The expression is: , Where Is the electrical signal after optoelectronic conversion. Is the optoelectronic responsivity. Is the optoelectronic conversion efficiency.

[0061] Using an adaptive filtering model, the change of the electrical signal after optoelectronic conversion is monitored in real time, and the driving voltage of the gate is dynamically adjusted. The expression is: G (t)= 1 2 [1+ tanh (α⋅ ∫ 0 t E ( t ' )d t ' )] , Where Represents the gate voltage control signal. Represents the hyperbolic tangent function. Represents the gain coefficient of signal regulation. Represents the past time variable For the electrical signal on Accumulation is performed. Represents the past time variable Of the electrical signal;

[0062] The gate voltage control signal Is applied to the gate of the MOS transistor to control the on and off states, and an initially stable electrical signal is output.

[0063] S4. Transmit the initially stable signal to the multi-parameter feedback control unit, and use the real-time monitored electrical characteristic parameters to adjust the working state of the MOS transistor, and output an accurate signal.

[0064] Furthermore, integrate the real-time monitored electrical parameters, environmental parameters, state parameters, and dynamic parameters into a multi-dimensional data matrix; It should be noted that the process of integrating real-time monitored electrical parameters, environmental parameters, state parameters, and dynamic parameters into a multi-dimensional data matrix is achieved by uniformly structuring various types of parameters according to time series or spatial dimensions: First, electrical parameters (such as voltage, current, power, etc.), environmental parameters (such as temperature, humidity, etc.), state parameters (such as the temperature of MOS transistors, electromagnetic interference, etc.), and dynamic parameters (such as the change rates of voltage and current) are all synchronously collected, with each parameter corresponding to a time point or a state value; Then, these different types of parameters are arranged in the form of rows or columns to form a matrix with multi-dimensional characteristics, where each row or column represents all the parameter values at a certain moment or a certain state. Finally, such a matrix can reflect the overall behavior at different time points or states and is used as the input for subsequent deep learning models.

[0065] Input the multi-dimensional data matrix into a hybrid model of a convolutional neural network (CNN) and a long short-term memory neural network (LSTM) for prediction; Preferably, CNN is good at extracting local spatial features from high-dimensional data, while LSTM is good at processing time series data. Therefore, the hybrid model can simultaneously capture the spatial dependence of the input data (such as the pattern of electrical parameters changing with position) and the time dynamic changes (such as the changing trends of current and voltage over time); Compared with traditional linear regression or simple neural networks, the hybrid model can understand the complex relationships of multi-dimensional data more deeply, especially having significant advantages in time-varying systems.

[0066] According to the prediction results, use an adaptive fuzzy controller to dynamically adjust the gate voltage of the MOS transistor, and at the same time, based on the recursive least squares method, adjust the gain bits and gain thresholds of the controller in real time. The specific steps are as follows: Initialize the gain bits and gain thresholds as the initial values of the recursive least squares method; It should be noted that setting the initial gain parameters and threshold parameters and using them as the initial values of the recursive least squares algorithm ensures that the recursive least squares method can have a reasonable starting point at the beginning.

[0067] Collect the error signal between the control output of the power MOS transistor and the target signal; Specifically, the output voltage and current of the power MOS transistor are monitored in real time by sensors and converted into digital signals. Then, the voltage and current signals obtained in real time are compared with the preset target signals, and the difference between the two, that is, the error signal, is calculated. This error signal is used as a feedback input and provided to the subsequent recursive least squares algorithm for parameter update and optimal control. Further, to improve accuracy, data sampling is performed regularly, and the collected error signals are smoothed to eliminate noise interference, thereby ensuring the stability and reliability of the error signals. In this way, the precise error between the output of the power MOS transistor and the target signal can be continuously obtained, providing a real-time basis for subsequent gain adjustment.

[0068] According to the error signal, the gain bits and gain thresholds of the controller are updated in real time using the recursive least squares method to minimize the error.

[0069] Based on the adjusted adaptive fuzzy controller, an accurate gate voltage signal is output.

[0070] S5. An intelligent algorithm is used to perform predictive optimization adjustment on the accurate signal, and the processed signal is output through the last-stage power MOS transistor, and an optimized and stable signal is output.

[0071] Further, the noise of the signal is processed by a Bayesian filter;

[0072] Based on the signal processed by the Bayesian filter, a long short-term memory neural network (LSTM) is used for time series optimization;

[0073] Preferably, the LSTM can capture short-term fluctuations and long-term trends in the signal, further optimize the predicted signal of the previous step, and output a signal containing time series information. This result is used to capture the periodicity and trends in the signal, enhancing the system's response ability to future signal changes.

[0074] Based on the time series optimized signal output by the long short-term memory neural network (LSTM), a generalized autoregressive conditional heteroskedasticity (GARCH) model is applied to predict the volatility of the signal;

[0075] Preferably, the GARCH model can handle the irregular fluctuations in the signal and predict the possible change range of the future signal. Through the accurate prediction of volatility, a more stable signal benchmark is provided for the subsequent optimization steps, ensuring continuous stability in a complex environment.

[0076] After the volatility prediction is completed, the quantum optimization algorithm performs global optimization on the processing results of the Bayesian filter, the long short-term memory neural network time series optimization, and the generalized autoregressive conditional heteroskedasticity (GARCH) model volatility detection;

[0077] According to the parameter settings output by the quantum optimization algorithm, dynamically adjust the gate voltage of the last-stage power MOS transistor to control its on and off states, and finally output an optimized and stable electrical signal.

[0078] This embodiment also provides a signal processing system based on a power MOS transistor, including: an acquisition module, a filtering processing module, a dynamic adjustment module, a feedback control module, and a signal output module; the acquisition module is used to acquire the input signal of the external environment and monitor the electrical characteristic parameters in real time; the filtering processing module is used to perform preliminary filtering processing on the input signal of the external environment to generate a raw signal; the dynamic adjustment module is used to transmit the raw signal to a multi-layer drive architecture, dynamically adjust the switching state of the power MOS transistor, and output a preliminary stable signal; the feedback control module is used to transmit the preliminary stable signal to a multi-parameter feedback control unit, and use the real-time monitored electrical characteristic parameters to adjust the working state of the MOS transistor to output an accurate signal; the signal output module is used to perform predictive optimization adjustment on the accurate signal using an intelligent algorithm, and process it through the last-stage power MOS transistor to output an optimized and stable signal.

[0079] This embodiment also provides a computer device applicable to the case of the signal processing method based on a power MOS transistor, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the signal processing method based on a power MOS transistor proposed in the above embodiment.

[0080] This computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0081] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the signal processing method based on a power MOS transistor as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, a magnetic disk or an optical disc.

[0082] In summary, the present invention adopts a multi-layer drive architecture and an intelligent algorithm to dynamically adjust the switching state of the power MOS transistor, can monitor and feedback the electrical characteristics of the input signal in real time, thereby improving the accuracy and stability of signal processing; through a multi-stage filtering structure, high-frequency noise and transient interference are effectively suppressed, and the signal purification effect is enhanced; at the same time, an intelligent prediction algorithm is used to perform predictive optimization adjustment of the signal, which can perform adaptive adjustment according to historical data and the current signal trend, avoiding the problem of over-adjustment or under-adjustment of the signal, and significantly improving the signal processing ability in a complex environment.

[0083] Embodiment 2, referring to Table 1, is the second embodiment of the present invention. To further verify the technical solution of the present invention, experimental simulation data of the signal processing method based on a power MOS transistor are given.

[0084] In this experiment, the signal processing method based on a power MOS transistor was compared with the traditional linear amplifier signal processing method. In the traditional method, a linear amplifier is usually used to amplify and filter the input signal, and relies on the static parameters of a fixed filter and amplifier to process the signal, lacking the ability of dynamic adjustment and unable to respond to the change of the input signal in real time, especially being more limited in a complex external environment.

[0085] First, the experiment collects two types of signals from the external environment, namely analog signals and communication signals. For analog signals, an analog-to-digital converter is used to convert them into digital signals; for communication signals, they are demodulated into baseband signals through a modem. Then, the power spectral density of the two types of signals is analyzed through fast Fourier transform (FFT) to identify the frequency distribution of the noise. On this basis, a multi-stage filtering structure is applied to gradually purify the signals, and the purified signals are restored from the frequency domain to the time domain through inverse Fourier transform to generate the original signals.

[0086] Secondly, the original signals enter a multi-layer drive architecture through an optoelectronic conversion module. First, a laser source is used to convert the original signals into high-frequency optical signals, and signal modulation is achieved through optical fiber transmission. Next, Gaussian convolution is used to smooth the signals to reduce the high-frequency noise in the signals. After the optical signals are converted into electrical signals through optoelectronic conversion, an adaptive filtering model is used to monitor them in real time, dynamically adjust the driving voltage of the gate, and output a preliminarily stable electrical signal.

[0087] Then, the preliminarily stable signals are input into a multi-parameter feedback control unit to monitor electrical parameters, environmental parameters, state parameters, and dynamic parameters in real time. These parameters are integrated into a multi-dimensional data matrix and input into a hybrid model of a convolutional neural network and a long short-term memory (LSTM) neural network for prediction. According to the prediction results, an adaptive fuzzy controller is used to dynamically adjust the gate voltage of the MOS transistor, and based on the recursive least squares method, the gain bits and gain thresholds of the controller are optimized in real time. Finally, an accurate gate voltage signal is output to generate an accurate signal.

[0088] Finally, the accurate signals are further optimized through intelligent algorithms. First, a Bayesian filter is used to process the residual noise in the signals. Then, an LSTM neural network is used to optimize the signals in terms of time series to predict the volatility of the signals. Finally, based on the generalized autoregressive conditional heteroskedasticity (GARCH) model, the signal volatility is detected, and a quantum optimization algorithm is used to globally optimize the processing results of all steps. According to the optimization results, the gate voltage of the last-stage power MOS transistor is dynamically adjusted to output a finally optimized stable electrical signal.

[0089] Specifically, it is shown in Table 1 below:

[0090] Table 1 Comparison Table of Experimental Data

[0091] It can be seen from the experimental data table that the signal processing method based on power MOS transistors shows significant improvements in multiple key performance indicators, as shown below:

[0092] Signal-to-noise level: The signal-to-noise level after processing by the traditional linear amplifier method is 10.5 dB, while the processing method based on power MOS transistors reduces the noise level to 3.2 dB, achieving a noise reduction effect of 69.52%. This significant reduction is mainly attributed to the combination of a multi-stage filtering structure and Gaussian convolution smoothing, which eliminates most of the high-frequency noise. In contrast, the traditional linear amplifier method relies on fixed filter and amplifier parameters and is difficult to dynamically adjust according to changes in environmental noise, resulting in poor noise suppression.

[0093] Signal delay: The signal processing delay of the traditional linear amplifier is 15 ms, while the processing method based on power MOS transistors reduces the delay to 6 ms, a reduction of 60%. This improvement benefits from the prediction capabilities of convolutional neural networks and LSTM neural networks, significantly enhancing the signal regulation efficiency; in the traditional method, the signal processing unit relies on a fixed processing architecture and cannot respond in real time to signal fluctuations, resulting in a longer delay.

[0094] Power consumption: The signal processing method based on power MOS transistors also shows an advantage in power consumption. The power consumption during the processing is 38 W, a 24% reduction compared to the 50 W of the traditional linear amplifier. By dynamically adjusting the gate voltage control of the MOS transistors, this method can optimize power consumption based on real-time monitored electrical parameters. In contrast, traditional linear amplifiers usually maintain a fixed power output throughout the signal processing and cannot be adjusted according to actual needs, resulting in low energy efficiency.

[0095] Signal distortion rate: The signal distortion rate is one of the key indicators to measure the quality of signal processing. The signal distortion rate of the traditional linear amplifier is 8.5%, while the processing method based on power MOS transistors reduces it to 2.1%, a reduction of 75.29%. This improvement benefits from the temporal optimization of Bayesian filtering and LSTM neural networks, ensuring the integrity of the signal during processing. In contrast, the traditional method lacks the optimization of intelligent algorithms, and the amplifier is prone to introduce additional distortion during signal processing, especially in a signal environment with high-speed changes.

[0096] Signal stability: The experiment also measured the overall stability of the signal. Under the traditional linear amplifier method, the signal stability score is 80, while the stability score of the processing method based on power MOS transistors reaches 95, an increase of 18.75%. Through the global optimization of the signal processing process by the quantum optimization algorithm, the stability of the signal in each processing stage is ensured. In contrast, the fixed filter and amplifier parameters in the traditional method are difficult to maintain a stable output when facing dynamically changing signals, especially with poor stability when processing complex signals.

[0097] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A signal processing method based on a power MOS tube, characterized in that: include: Collect input signals from the external environment and monitor electrical characteristic parameters in real time; Perform preliminary filtering on the input signal of the external environment to generate the original signal; The original signal is transmitted to the multi-layer driving architecture, the switching state of the power MOS tube is dynamically adjusted, and a preliminary stable signal is output; The preliminary stable signal is transmitted to the multi-parameter feedback control unit, and the working state of the MOS tube is adjusted by real-time monitoring of the electrical characteristic parameters to output accurate signals; An intelligent algorithm is used to predictively optimize and adjust the precise signal, and the signal is processed through the last-stage power MOS tube to output an optimized stable signal.

2. The signal processing method based on power MOS tube according to claim 1, characterized in that: The input signal includes: an analog signal, a digital signal and a communication signal; The electrical characteristic parameters include: electrical parameters, environmental parameters, state parameters and dynamic parameters.

3. The signal processing method based on power MOS tube according to claim 2, characterized in that: The input signal of the external environment is subjected to preliminary filtering processing to generate the original signal. The specific steps are as follows: For analog signals, they are converted into digital signals through analog-to-digital converters; Demodulate the communication signal into a baseband signal; The power spectral density of digital and baseband signals is estimated by fast Fourier transform, and the frequency distribution of noise is identified; According to the frequency distribution of noise, a multi-stage filtering structure is applied to gradually purify the signal; The purified signal is regressed from the frequency domain to the time domain through inverse Fourier transform to generate the original signal.

4. The signal processing method based on power MOS tube according to claim 3, characterized in that: The original signal is transmitted to the multi-layer driving architecture, the switching state of the power MOS tube is dynamically adjusted, and a preliminary stable signal is output. The specific steps are as follows: The laser source is used to generate a high-frequency optical signal to modulate the original signal and convert the original signal into an optical signal; The modulated optical signal is transmitted to the multi-layer driving unit through the optical fiber, and the signal is smoothed by Gaussian convolution, and the expression is: ； in, Represents the optical signal output after multi-layer modulation, represents the input optical signal, is the time variable, is the delay variable, represents the Gaussian convolution kernel, Represents the exponential part of the Gaussian function is the standard deviation of the Gaussian distribution, Express Differential element for integration; The optical signal output after multi-layer modulation is converted into an electrical signal through the photoelectric conversion module, and the expression is: ； in, is the electrical signal after photoelectric conversion, is the photoelectric responsivity, is the photoelectric conversion efficiency; Using the adaptive filtering model, the changes in the electrical signal after photoelectric conversion are monitored in real time, and the gate drive voltage is dynamically adjusted. The expression is: ； in, represents the gate voltage control signal, represents the hyperbolic tangent function, represents the gain factor of the signal conditioning, Represents the past time variable The electrical signal on To accumulate, Represents past time variables The electrical signal; The gate voltage control signal Applied to the gate of the MOS tube, it controls the on and off states and outputs a preliminary stable electrical signal.

5. The signal processing method based on power MOS tube according to claim 4, characterized in that: The preliminary stable signal is transmitted to the multi-parameter feedback control unit, and the working state of the MOS tube is adjusted by real-time monitoring of the electrical characteristic parameters to output an accurate signal. The specific steps are as follows: Integrate the real-time monitored electrical parameters, environmental parameters, state parameters and dynamic parameters into a multi-dimensional data matrix; Input the multidimensional data matrix into the hybrid model of convolutional neural network and long short-term memory neural network for prediction; According to the prediction results, the gate voltage of the MOS tube is dynamically adjusted using an adaptive fuzzy controller, and the gain bit and gain threshold of the controller are adjusted in real time based on the recursive least squares method. The accurate gate voltage signal is output based on the adjusted adaptive fuzzy controller.

6. The signal processing method based on power MOS tube according to claim 5, characterized in that: The recursive least square method is used to adjust the gain bit and gain threshold of the controller in real time. The specific steps are as follows: Initializing the gain bit and the gain threshold as initial values ​​for the recursive least squares method; Use quantum gate operation to make the gain bit and the threshold bit enter an entangled state; Collect the error signal between the control output of the power MOS tube and the target signal; According to the error signal, the controller gain bits and gain thresholds are updated in real time using the recursive least squares method to minimize the error.

7. The signal processing method based on power MOS tube according to claim 6, characterized in that: The intelligent algorithm is used to predictively optimize and adjust the precise signal, and the signal is processed by the last-stage power MOS tube to output the optimized stable signal. The specific steps are as follows: Process the noise of the signal through Bayesian filter; Based on the signal processed by Bayesian filtering, long short-term memory neural network is used for timing optimization; Based on the time series optimization signal output by the long short-term memory neural network, the generalized autoregressive conditional heteroskedasticity model is used to predict the volatility of the signal; The processing results of Bayesian filter, long short-term memory neural network timing optimization and generalized autoregressive conditional heteroskedasticity model volatility detection are globally optimized through quantum optimization algorithm, and the output parameter settings are set; According to the parameter settings output by the quantum optimization algorithm, the gate voltage of the last-stage power MOS tube is dynamically adjusted to control its on and off states, and finally an optimized stable electrical signal is output.

8. A signal processing system based on a power MOS tube, based on the signal processing method based on a power MOS tube according to any one of claims 1 to 7, characterized in that: include: Acquisition module, filtering processing module, dynamic adjustment module, feedback control module and signal output module; The acquisition module is used to collect input signals from the external environment and monitor electrical characteristic parameters in real time; The filtering processing module is used to perform preliminary filtering processing on the input signal of the external environment to generate an original signal; The dynamic adjustment module is used to transmit the original signal to the multi-layer driving architecture, dynamically adjust the switching state of the power MOS tube, and output a preliminary stable signal; The feedback control module is used to transmit the preliminary stable signal to the multi-parameter feedback control unit, adjust the working state of the MOS tube by real-time monitoring of the electrical characteristic parameters, and output an accurate signal; The signal output module is used to use an intelligent algorithm to perform predictive optimization and adjustment on the precise signal, and process it through the last-stage power MOS tube to output an optimized stable signal.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the signal processing method based on a power MOS tube according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the signal processing method based on a power MOS tube according to any one of claims 1 to 7 are implemented.

Images