Infrared Narrow Beam Emission Communication Control System and Method

Abstract

The present invention belongs to the technical field of signal processing, and discloses an infrared narrow beam emission communication control system and method. The method includes: extracting features from the operating parameters of the emitting device collected in real time per unit time to obtain the operating feature data of the emitting device; extracting features from the environmental parameters of the emitting device collected in real time per unit time to obtain the environmental feature data of the emitting device; inputting the operating feature data of the emitting device and the environmental feature data of the emitting device into an optical characteristic curve prediction model to obtain a set of optical characteristic curves of the emitting device in the next unit time; inputting the set of optical characteristic curves into a signal distortion diagnosis model to obtain a set of signal distortion diagnosis results in the next unit time; based on the operating feature data of the emitting device, the set of optical characteristic curves in the next unit time, and the set of signal distortion diagnosis results, performing digital pre-distortion intelligent adaptive adjustment on the emitting device to dynamically compensate for signal distortion; and improving the signal quality.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and more specifically, to an infrared narrow beam emission communication control system and method. Background Art

[0002] With the wide application of infrared narrow beam communication technology in high-speed and high-reliability scenarios, digital pre-distortion technology has become the core means to solve the non-linear distortion of transmitting devices. Existing digital pre-distortion methods usually rely on establishing a non-linear mathematical model of the transmitting device, and calculating pre-distortion compensation parameters based on the non-linear mathematical model to preprocess the baseband signal, so as to offset the amplitude-phase distortion introduced by hardware such as power amplifiers and laser drivers, so that the signal after passing through the non-linear system can be as close as possible to the ideal signal. However, such methods usually assume that the non-linear characteristics of the transmitting device are stable in a short period of time. Therefore, the compensation strategy depends on the fixed parameters measured in advance or is adjusted based on the feedback from the receiving end, which to a certain extent limits its adaptability to dynamic environments.

[0003] However, the actual application of infrared narrow beam communication is usually in a dynamic environment. The non-linear distortion characteristics of the transmitting device present fast time-varying characteristics. The non-linear distortion characteristics are not fixed, but change dynamically with time. There are scenarios where the change rate of the non-linear distortion characteristics exceeds the response ability of the traditional feedback mechanism. As a result, in the scenario of channel congestion, the feedback link delay causes the pre-distortion parameters to mismatch with the current hardware state, resulting in temporary signal distortion and feedback hysteresis, and then leading to the failure of digital pre-distortion compensation. In addition, the existing technology only performs reverse correction based on the current or historical distortion data, and cannot predict the evolution trend of the non-linear characteristics of the transmitting device, resulting in the disconnection between the digital pre-distortion technology and the state of future transmitting devices.

[0004] Therefore, the existing technology cannot compensate for the possible future distortion in advance before signal transmission, and there is still a problem of deteriorating communication quality, that is, the existing technology cannot adaptively optimize digital pre-distortion, and it is difficult to meet the real-time requirements of high-speed communication, which seriously restricts the reliability and adaptability of the infrared narrow beam communication system. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art and achieve the above object, the present invention provides the following technical solutions: an infrared narrow beam emission communication control method, including:

[0006] Performing feature extraction based on the operation parameters of the transmitting device collected in real time per unit time to obtain the operation feature data of the transmitting device;

[0007] Performing feature extraction based on the environmental parameters of the transmitting device collected in real time per unit time to obtain the environmental feature data of the transmitting device;

[0008] Input the operating characteristic data and environmental characteristic data of the transmitting device into the optical characteristic curve prediction model to obtain the set of optical characteristic curves of the transmitting device in the future unit time;

[0009] Input the set of optical characteristic curves into the signal distortion diagnosis model to obtain the set of signal distortion diagnosis results in the future unit time;

[0010] Based on the operating characteristic data of the transmitting device, the set of optical characteristic curves in the future unit time, and the set of signal distortion diagnosis results, perform digital pre-distortion intelligent adaptive adjustment on the transmitting device to dynamically compensate for signal distortion.

[0011] Furthermore, the method for performing digital pre-distortion intelligent adaptive adjustment on the transmitting device includes:

[0012] S700: Denote the number of subsets of signal distortion diagnosis results in the set of signal distortion diagnosis results as NUM, set the initial value of the counting variable num to 1, and the value range of num is from 1 to NUM;

[0013] S701: Obtain the num-th subset of signal distortion diagnosis results from the set of signal distortion diagnosis results;

[0014] If the signal distortion type in the num-th subset of signal distortion diagnosis results is amplitude distortion, perform digital pre-distortion intelligent adaptive adjustment for amplitude distortion;

[0015] If the signal distortion type in the num-th subset of signal distortion diagnosis results is phase distortion, perform digital pre-distortion intelligent adaptive adjustment for phase distortion;

[0016] If the signal distortion type in the num-th subset of signal distortion diagnosis results is memory effect, perform digital pre-distortion intelligent adaptive adjustment for memory effect;

[0017] S702: Let num = num + 1. If num is less than or equal to NUM, continue to execute S701. If num is greater than NUM, complete the digital pre-distortion intelligent adaptive adjustment and end the current process.

[0018] Furthermore, the method for performing digital pre-distortion intelligent adaptive adjustment for amplitude distortion includes:

[0019] Numerically label the amplitude distortion to obtain the amplitude distortion value;

[0020] Input the amplitude distortion value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves in the future unit time into the amplitude distortion adjustment model to obtain the amplitude distortion gain coefficient;

[0021] The output signal after digital predistortion corresponding to amplitude distortion is calculated based on the amplitude distortion gain coefficient, denoted as the amplitude distortion output signal;

[0022] The signal distortion serial number interval corresponding to amplitude distortion is obtained from the signal distortion diagnosis result subset. When the signal output moment of the transmitting device matches the signal distortion serial number interval of amplitude distortion, the amplitude distortion output signal is used as the output signal of the transmitting device to dynamically compensate for amplitude distortion, and the digital predistortion intelligent adaptive adjustment for amplitude distortion is completed.

[0023] Furthermore, the method for digital predistortion intelligent adaptive adjustment for phase distortion includes:

[0024] The phase distortion is numerically labeled to obtain the phase distortion value;

[0025] The phase distortion value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves in the future unit time are input into the phase distortion adjustment model to obtain the phase distortion compensation angle;

[0026] The output signal after digital predistortion corresponding to phase distortion is calculated based on the phase distortion compensation angle, denoted as the phase distortion output signal;

[0027] The signal distortion serial number interval corresponding to phase distortion is obtained from the signal distortion diagnosis result subset. When the signal output moment of the transmitting device matches the signal distortion serial number interval of phase distortion, the phase distortion output signal is used as the output signal of the transmitting device to dynamically compensate for phase distortion, and the digital predistortion intelligent adaptive adjustment for phase distortion is completed.

[0028] Furthermore, the method for digital predistortion intelligent adaptive adjustment for memory effect includes:

[0029] The memory effect is numerically labeled to obtain the memory effect value;

[0030] The memory effect value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves in the future unit time are input into the memory effect adjustment model to obtain the memory depth and the memory effect weight;

[0031] The output signal after digital predistortion corresponding to the memory effect is calculated based on the memory depth and the memory effect weight, denoted as the memory effect output signal;

[0032] The signal distortion serial number interval corresponding to the memory effect is obtained from the signal distortion diagnosis result subset. When the signal output moment of the transmitting device matches the signal distortion serial number interval of the memory effect, the memory effect output signal is used as the output signal of the transmitting device to dynamically compensate for the memory effect, and the digital predistortion intelligent adaptive adjustment for the memory effect is completed.

[0033] Furthermore, the operating parameters of the emitting device include the PN junction temperature, drive current, bias voltage, output optical power, and beam pointing angle;

[0034] The method for obtaining the operating characteristic data of the emitting device includes:

[0035] Divide the unit time into T time points, construct the PN junction temperature set from the PN junction temperatures at the T time points, construct the drive current set from the drive currents at the T time points, construct the bias voltage set from the bias voltages at the T time points, construct the output optical power set from the output optical powers at the T time points, and construct the beam pointing angle set from the beam pointing angles at the T time points;

[0036] Based on the output optical power set and the drive current set, calculate the optical power - current efficiency to obtain the optical power - current curve;

[0037] Based on the output optical power set and the bias voltage set, calculate the half - wave voltage to obtain the half - wave voltage curve;

[0038] Based on the PN junction temperature set, the drive current set, and the beam pointing angle set, perform thermo - electro - optical coupling distortion analysis to obtain the thermo - electro - optical coupling distortion curve;

[0039] Construct the PN junction temperature set, the drive current set, the bias voltage set, the output optical power set, the beam pointing angle set, the optical power - current curve, the half - wave voltage curve, and the thermo - electro - optical coupling distortion curve into the operating characteristic data of the emitting device.

[0040] Furthermore, the method for obtaining the optical power - current curve includes:

[0041] S100: Let the initial value of the time - step variable t be 1, and the value range of t is from 1 to T - 1; Let the initial value of the counting variable bl be 1, and the value range of bl is from 1 to T - 1;

[0042] S101: Obtain the t - th and (t + 1) - th output optical powers from the output optical power set, subtract the t - th output optical power from the (t + 1) - th output optical power to get the bl - th output optical power change difference; Obtain the t - th and (t + 1) - th drive currents from the drive current set, subtract the t - th drive current from the (t + 1) - th drive current to get the bl - th drive current change difference;

[0043] S102: Divide the bl - th output optical power change difference by the drive current change difference to obtain the optical power - current efficiency;

[0044] S103: Add the optical power - current efficiency to the optical power - current efficiency set;

[0045] S104: Let \(t = t + 1\). If \(t\) is less than or equal to \(T - 1\), then let \(bl = bl + 1\) and continue to execute S101 to S103; if \(t\) is greater than \(T - 1\), then execute S105;

[0046] S105: Using the \(bl\) counting variables as the horizontal axis data and the \(bl\) optical power - current efficiencies in the optical power - current efficiency set as the vertical axis data, plot the optical power - current curve to end the current process.

[0047] Further, the method for obtaining the half - wave voltage curve includes:

[0048] S200: Preset the initial value of the time - step variable \(bc\) to 1, and the value range of \(bc\) is from 1 to \(T\);

[0049] S201: Obtain the \(bc\) - th bias voltage from the bias voltage set and the \(bc\) - th output optical power from the output optical power set;

[0050] S202: Calculate the corresponding half - wave voltage based on the \(bc\) - th bias voltage and the output optical power;

[0051] S203: Add the \(bc\) - th half - wave voltage to the half - wave voltage set;

[0052] S204: Let \(bc = bc + 1\). If \(bc\) is less than or equal to \(T\), then continue to execute S201 to S203; if \(bc\) is greater than \(T\), then execute S205;

[0053] S205: Using the \(bc\) time - step variables as the horizontal axis data and the \(bc\) half - wave voltages in the half - wave voltage set as the vertical axis data, plot the half - wave voltage curve to end the current process.

[0054] Further, the method for obtaining the thermal - electro - optical coupling distortion curve includes:

[0055] S300: Let the initial value of the time - step variable \(t\) be 1, and the value range of \(t\) is from 1 to \(T - 1\); preset the initial value of the counting variable \(bd\) to 1, and the value range of \(bd\) is from 1 to \(T - 1\);

[0056] S301: Obtain the t-th and (t + 1)-th PN junction temperatures from the PN junction temperature set, subtract the t-th PN junction temperature from the (t + 1)-th PN junction temperature to obtain the bd-th PN junction temperature difference; obtain the t-th and (t + 1)-th drive currents from the drive current set, calculate the average value of the t-th and (t + 1)-th drive currents, denoted as the drive current average value, to obtain the bd-th drive current average value; obtain the t-th and (t + 1)-th beam pointing angles from the beam pointing angle set, subtract the t-th beam pointing angle from the (t + 1)-th beam pointing angle to obtain the bd-th beam pointing angle difference; the time points of the (t + 1)-th PN junction temperature, the (t + 1)-th drive current, and the (t + 1)-th beam pointing angle are the same;

[0057] S302: Calculate the bd-th thermo-electro-optical coupling distortion degree based on the bd-th PN junction temperature difference, the drive current average value, the beam pointing angle difference, and the time interval;

[0058] S303: Add the bd-th thermo-electro-optical coupling distortion degree to the thermo-electro-optical coupling distortion set;

[0059] S304: Let t = t + 1. If t is less than or equal to T - 1, then let bd = bd + 1, and continue to execute S301 to S303; if t is greater than T - 1, then execute S305;

[0060] S305: Use the bd counting variables as the horizontal axis data, and use the bd thermo-electro-optical coupling distortion degrees in the thermo-electro-optical coupling distortion set as the vertical axis data to plot the thermo-electro-optical coupling distortion curve, and end the current process.

[0061] Furthermore, the environmental parameters of the emitting device include environmental temperature, environmental humidity, air flow velocity, electric field strength, magnetic field strength, and interference power;

[0062] The method for obtaining the environmental characteristic data of the emitting device includes:

[0063] Divide the unit time into T time points, construct the environmental temperature set from the environmental temperatures of the T time points, construct the environmental humidity set from the environmental humidities of the T time points, construct the air flow velocity set from the air flow velocities of the T time points, construct the electric field strength set from the electric field strengths of the T time points, construct the magnetic field strength set from the magnetic field strengths of the T time points, and construct the interference power set from the interference powers of the T time points;

[0064] Calculate the environmental temperature average value and environmental temperature variance based on the environmental temperature set; calculate the environmental humidity average value and environmental humidity variance based on the environmental humidity set; calculate the magnetic field strength average value and magnetic field strength variance based on the magnetic field strength set;

[0065] Calculate the air velocity difference set and the air velocity change rate set based on the air velocity set;

[0066] Calculate the electric field intensity difference set and the electric field intensity change rate set based on the electric field intensity set;

[0067] Calculate the interference power difference set and the interference power change rate set based on the interference power set;

[0068] Construct the environmental characteristic data of the transmitting device by using the mean value of the environmental temperature, the variance of the environmental temperature, the mean value of the environmental humidity, the variance of the environmental humidity, the mean value of the magnetic field intensity, the variance of the magnetic field intensity, the air velocity difference set, the air velocity change rate set, the electric field intensity difference set, the electric field intensity change rate set, the interference power difference set and the interference power change rate set.

[0069] An infrared narrow-beam emission communication control system for implementing the infrared narrow-beam emission communication control method, comprising:

[0070] A first processing module for extracting features from the operating parameters of the transmitting device collected in real time per unit time to obtain the operating characteristic data of the transmitting device;

[0071] A second processing module for extracting features from the environmental parameters of the transmitting device collected in real time per unit time to obtain the environmental characteristic data of the transmitting device;

[0072] A curve prediction module for inputting the operating characteristic data of the transmitting device and the environmental characteristic data of the transmitting device into an optical characteristic curve prediction model to obtain a set of optical characteristic curves of the transmitting device in the future per unit time;

[0073] A distortion diagnosis module for inputting the set of optical characteristic curves into a signal distortion diagnosis model to obtain a set of signal distortion diagnosis results in the future per unit time;

[0074] An intelligent optimization module, based on the operating characteristic data of the transmitting device, the set of optical characteristic curves in the future per unit time, and the set of signal distortion diagnosis results, performs digital pre-distortion intelligent adaptive adjustment on the transmitting device to dynamically compensate for signal distortion.

[0075] Compared with the prior art, the technical effects and advantages of the infrared narrow-beam emission communication control system and method of the present invention:

[0076] First of all, this solution breaks through the lag of traditional digital pre-distortion technology. By predicting the non-linear change trend of the transmitting device in the future per unit time through a machine learning model, it realizes feed-forward compensation, can pre-adjust the compensation parameters before signal transmission, and avoids the problem of short-term distortion accumulation caused by feedback lag, improving the real-time performance and robustness of the system.

[0077] Secondly, this solution enhances the adaptability to dynamic environments. By real-time collection of the environmental parameters of the transmitting device and constructing environmental characteristic data, the system can perceive the impact of external factors such as ambient temperature changes, electromagnetic interference, and airflow disturbances on the operating status of the transmitting device. In this way, the evolution trend of the optical characteristic curve can still be accurately predicted in a dynamic environment, ensuring that the digital pre-distortion compensation strategy always matches the current hardware status, thereby improving the stability of signal transmission in complex environments.

[0078] Furthermore, this solution implements adaptive optimization strategies for different distortion types. It constructs independent adjustment models for amplitude distortion, phase distortion, and memory effect, respectively. Based on the distortion diagnosis results, the signal is adjusted specifically to ensure that amplitude nonlinearity, phase offset, and memory effect distortion are effectively compensated.

[0079] In summary, the present invention performs digital pre-distortion intelligent adaptive adjustment on the transmitting device, dynamically compensates for signal distortion, and improves the signal quality and stability of the system, thereby effectively improving the signal quality, transmission reliability and environmental adaptability of the infrared narrow-beam communication system. It overcomes the problem of insufficient nonlinear distortion compensation capability of the existing technology in complex dynamic environments, and has broad application value in high-speed, high-stability optical communication scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 Schematic diagram of an infrared narrow-beam transmission communication control system according to embodiment 1 of the present invention;

[0081] Figure 2 This is a schematic diagram of an infrared narrow-beam transmission communication control system according to embodiment 2 of the present invention;

[0082] Figure 3 A flow chart of a method for performing intelligent adaptive adjustment of digital predistortion on a transmitting device;

[0083] Figure 4 A flow chart of a method for performing intelligent adaptive adjustment of digital predistortion for amplitude distortion;

[0084] Figure 5 A flow chart of a method for performing intelligent adaptive adjustment of digital predistortion for phase distortion;

[0085] Figure 6 A flow chart of a method for intelligent adaptive adjustment of digital predistortion for memory effect;

[0086] Figure 7 The system mind map corresponding to the first acquisition module;

[0087] Figure 8 This is the system mind map corresponding to the second acquisition module. DETAILED DESCRIPTION

[0088] The technical solutions in the embodiments of the present invention will be described in detail, clearly, and completely in conjunction with the accompanying drawings in the embodiments of the present invention. It should be specifically noted that the following specific embodiments are only used to better illustrate and explain the technical solutions of the present invention, aiming to enable those skilled in the art to better understand and implement the present invention, and should not be construed as limiting the protection scope of the present invention. Without departing from the spirit and essence of the present invention, those skilled in the art can modify, adjust, or make equivalent replacements according to the content disclosed in the present invention, and these should all be regarded as within the protection scope of the present invention.

[0089] Embodiment 1

[0090] Please refer to Figure 1 As shown, this embodiment discloses an infrared narrow-beam emission communication control system, including a first acquisition module, a first processing module, a second acquisition module, a second processing module, a curve prediction module, a distortion diagnosis module, and an intelligent optimization module. Each module is connected by wire and / or wirelessly to achieve data transmission.

[0091] The first acquisition module is used to collect the operating parameters of the emitting device in real time per unit time. The operating parameters of the emitting device include the PN junction temperature, drive current, bias voltage, output optical power, and beam pointing angle.

[0092] It should be noted that the system brain map corresponding to the first acquisition module is as Figure 7 shown. The emitting device refers to a hardware set that converts an electrical signal into an optical signal and emits it, usually including a light source (laser diode / LED), a modulator (electro-optic modulator, such as MZM, EAM), a drive circuit (current driver, bias controller), and optical components (lens, MEMS beam control mirror), etc.

[0093] The PN junction is the core structure of a semiconductor device, which is formed by combining a P-type semiconductor and an N-type semiconductor on the same semiconductor material; when the P-type and N-type semiconductors come into contact, holes and electrons diffuse and recombine with each other to form a charge distribution region, called the PN junction. A micro-thermocouple (such as a K-type) is integrated inside the laser diode package, and the voltage signal is read through an ADC (analog-to-digital converter) and converted into a temperature value to obtain the PN junction temperature. The bias voltage refers to the DC voltage applied to the PN junction of a semiconductor emitting device (such as an infrared laser diode, LED, etc.). The bias voltage determines the operating state of the device and makes the emitting device in a suitable conduction or amplification region. In an infrared narrow-beam emission system, the bias voltage is usually used to adjust the output power and stability of the optical emitting device. The bias voltage can be directly acquired by a high-precision ADC (analog-to-digital converter) to obtain the DC voltage applied to both ends of the emitting device by the drive circuit. The drive current, output optical power, and beam pointing angle are obtained through corresponding sensors.

[0094] In an infrared narrow-beam emission communication system, the nonlinear distortion characteristics of the emission device are affected by the dynamic changes of various operating parameters, which is directly related to the signal quality and stability of the system. To achieve efficient compensation based on digital predistortion, it is necessary to collect key operating parameters including the PN junction temperature, drive current, bias voltage, output optical power, and beam pointing angle to optimize the predistortion compensation strategy.

[0095] As the core thermal parameter of the semiconductor emission device, the PN junction temperature determines the carrier mobility of the semiconductor device, directly affects the gain characteristics, modulation bandwidth, and nonlinear distortion degree of the emission device, and is a key factor determining the digital predistortion compensation effect. The drive current determines the carrier density of the semiconductor emission device and affects the optical injection gain and threshold current change, thus causing nonlinear distortion. The bias voltage is coupled with the drive current, directly affecting the operating point stability and linear modulation characteristics of the emission device, and further affecting the accuracy of predistortion compensation. The output optical power reflects the overall operating state of the emission device and, together with the PN junction temperature, drive current, and bias voltage, determines the power spectrum distribution and distortion characteristics of the transmitted signal, so it has an important impact on the adaptive adjustment of digital predistortion compensation. In addition, as a key factor affecting the stability of the narrow-beam communication link, the dynamic change of the beam pointing angle will cause signal power attenuation and wavefront distortion, thus affecting the predistortion compensation effect.

[0096] Therefore, collecting and real-time analyzing the PN junction temperature, drive current, bias voltage, output optical power, and beam pointing angle is inseparably technically related to accurately predicting the change trend of the device's nonlinear characteristics, realizing dynamic adaptive predistortion compensation, and improving the transmission quality and stability of the infrared narrow-beam communication system.

[0097] The first processing module extracts features based on the operating parameters of the emission device per unit time to obtain the operating characteristic data of the emission device.

[0098] The method for obtaining the operating characteristic data of the emission device includes:

[0099] Divide the unit time into T time points, construct the PN junction temperature set from the PN junction temperatures at the T time points, construct the drive current set from the drive currents at the T time points, construct the bias voltage set from the bias voltages at the T time points, construct the output optical power set from the output optical powers at the T time points, and construct the beam pointing angle set from the beam pointing angles at the T time points;

[0100] Calculate the optical power-current efficiency based on the output optical power set and the drive current set to obtain the optical power-current curve;

[0101] Calculate the half-wave voltage based on the output optical power set and the bias voltage set to obtain the half-wave voltage curve;

[0102] Perform thermo-electro-optical coupling distortion analysis based on the PN junction temperature set, the drive current set, and the beam pointing angle set to obtain the thermo-electro-optical coupling distortion curve;

[0103] Construct the operating characteristic data of the emitting device from the PN junction temperature set, the drive current set, the bias voltage set, the output optical power set, the beam pointing angle set, the optical power-current curve, the half-wave voltage curve, and the thermo-electro-optical coupling distortion curve.

[0104] The method for obtaining the optical power-current curve includes:

[0105] S100: Let the initial value of the time step variable t be 1, and the value range of t is from 1 to T - 1; preset the initial value of the counting variable bl to 1, and the value range of bl is from 1 to T - 1;

[0106] S101: Obtain the t-th and (t + 1)-th output optical powers from the output optical power set, subtract the t-th output optical power from the (t + 1)-th output optical power to get the bl-th output optical power change difference; obtain the t-th and (t + 1)-th drive currents from the drive current set, subtract the t-th drive current from the (t + 1)-th drive current to get the bl-th drive current change difference;

[0107] S102: Divide the bl-th output optical power change difference by the drive current change difference to get the optical power-current efficiency;

[0108] S103: Add the optical power-current efficiency to the optical power-current efficiency set;

[0109] S104: Let t = t + 1, if t is less than or equal to T - 1, then let bl = bl + 1, and continue to execute S101 to S103; if t is greater than T - 1, then execute S105;

[0110] S105: Use the bl counting variables as the horizontal axis data and the bl optical power-current efficiencies in the optical power-current efficiency set as the vertical axis data to plot the optical power-current curve and end the current process.

[0111] The method for obtaining the optical power-current efficiency includes:

[0112] ;

[0113] where, is the bl-th optical power-current efficiency, is the output optical power at the (t + 1)-th time point, is the output optical power at the t-th time point, is the drive current at the (t + 1)-th time point, is the drive current at the t-th time point.

[0114] It should be noted that the optical power-current curve is the core characteristic curve used to describe the relationship between the output optical power and the drive current of a laser diode or a light-emitting diode. It characterizes the efficiency of converting current into optical power. The greater the slope of the optical power-current curve, the higher the optoelectronic conversion efficiency of the device.

[0115] The method for obtaining the half-wave voltage curve includes:

[0116] S200: Preset the initial value of the time step variable bc to 1, and the value range of bc is from 1 to T;

[0117] S201: Obtain the bc-th bias voltage from the set of bias voltages and obtain the bc-th output optical power from the set of output optical powers;

[0118] S202: Calculate the corresponding half-wave voltage based on the bc-th bias voltage and the output optical power;

[0119] S203: Add the bc-th half-wave voltage to the set of half-wave voltages;

[0120] S204: Let bc = bc + 1. If bc is less than or equal to T, continue to execute S201 to S203; if bc is greater than T, execute S205;

[0121] S205: Use the bc time step variables as the horizontal axis data and the bc half-wave voltages in the set of half-wave voltages as the vertical axis data to plot the half-wave voltage curve and end the current process.

[0122] The method for obtaining the half-wave voltage includes:

[0123] ;

[0124] Among them, is the bc-th half-wave voltage, is a constant, is the bc-th bias voltage, is the inverse trigonometric function, is the bc-th output optical power, that is, the output optical power corresponding to the bc-th bias voltage, is the hardware property of the modulator itself, indicating the maximum optical power that the modulator can output under ideal conditions. This parameter is determined by the physical design of the modulator itself, is the initial phase offset, which is determined by those skilled in the art according to the bias point of the modulator (for example, an orthogonal operating point can be set, and Set to ).

[0125] It should be noted that the modulator refers to the core component in the transmitting device that loads the electrical signal onto the optical carrier. The method for obtaining the half-wave voltage is based on the mathematical relationship between the output optical power of the modulator and the bias voltage, and the calculation of the half-wave voltage is achieved through precise mathematical modeling. First, the numerator part reflects the contribution of the bias voltage to the modulation of the output optical power, where is a constant, representing the normalization factor for the phase change during the modulation process, while [[ID=!1]]represents the bias voltage corresponding to the bc-th time step, which determines the phase offset of the modulator. The introduction of the numerator part ensures that the calculation result can accurately correspond to the modulation characteristics in the physical system. Second, the denominator part reflects the relationship between the output optical power and the modulation phase, where The function is used to analyze the normalized output optical power at the bc-th time step , The introduction of ensures that the calculation result is consistent with the actual performance of the modulator and avoids the influence of individual differences on the calculation accuracy. At the same time As the initial phase offset, it determines the selection of the bias point to ensure the stability of the modulation device at the designed operating point.

[0126] In summary, the formula for the half-wave voltage effectively combines the bias voltage, output optical power, and device physical characteristics by reasonably constructing a mathematical relationship, ensuring the accuracy and consistency of the calculated half-wave voltage, and thus providing a reliable theoretical support for the performance analysis and optimization of the modulator.

[0127] The method for obtaining the thermal-electro-optical coupling distortion curve includes:

[0128] S300: Set the initial value of the time step variable t to 1, and the value range of t is from 1 to T - 1; set the initial value of the pre-designed counting variable bd to 1, and the value range of bd is from 1 to T - 1;

[0129] S301: Obtain the t-th and (t + 1)-th PN junction temperatures from the PN junction temperature set, subtract the t-th PN junction temperature from the (t + 1)-th PN junction temperature to get the bd-th PN junction temperature difference; obtain the t-th and (t + 1)-th drive currents from the drive current set, calculate the average value of the t-th and (t + 1)-th drive currents, denoted as the drive current average value, to get the bd-th drive current average value; obtain the t-th and (t + 1)-th beam pointing angles from the beam pointing angle set, subtract the t-th beam pointing angle from the (t + 1)-th beam pointing angle to get the bd-th beam pointing angle difference; the time points of the (t + 1)-th PN junction temperature, the (t + 1)-th drive current, and the (t + 1)-th beam pointing angle are the same;

[0130] S302: Calculate the bd-th thermo-electro-optical coupling distortion degree based on the temperature difference of the bd-th PN junction, the average driving current, the beam pointing angle difference, and the time interval;

[0131] S303: Add the bd-th thermo-electro-optical coupling distortion degree to the thermo-electro-optical coupling distortion set;

[0132] S304: Let t = t + 1. If t is less than or equal to T - 1, then let bd = bd + 1, and continue to execute S301 to S303; if t is greater than T - 1, then execute S305;

[0133] S305: Use the bd counting variables as the horizontal axis data and the bd thermo-electro-optical coupling distortion degrees in the thermo-electro-optical coupling distortion set as the vertical axis data to plot the thermo-electro-optical coupling distortion curve, and end the current process.

[0134] The method for obtaining the thermo-electro-optical coupling distortion degree includes:

[0135] ;

[0136] Where, is the bd-th thermo-electro-optical coupling distortion degree, is the temperature of the (t + 1)-th PN junction, is the temperature of the t-th PN junction, is the corresponding PN junction temperature difference, is the bd-th time interval, is the temperature change rate of the bd-th PN junction, and the PN junction temperature change rate is used to reflect the transient thermal shock.

[0137] is the (t + 1)-th driving current, is the t-th driving current, is the corresponding average driving current, is the time point corresponding to the (t + 1)-th driving current, is the time point corresponding to the t-th driving current, represents the time integral of the square of the driving current, which is used to characterize the Joule heat accumulation, is the integration measure symbol in the integral expression, representing the time variable.

[0138] is the (t + 1)-th beam pointing angle, is the t-th beam pointing angle, is the corresponding beam pointing angle difference; is the beam pointing angle change rate of the bd-th, is the absolute value of the bd-th beam pointing angle change rate, which is used to quantify the mechanical vibration intensity.

[0139] By comprehensively considering the PN junction temperature change, the driving current cumulative effect, and the beam pointing angle change, the coupling distortion degree of heat, electricity, and light is quantified, and the comprehensive distortion effect caused by the PN junction temperature fluctuation, the driving current action, and the beam pointing change is characterized. Thereby, the regularity of the mutual influence among heat, electricity, and light during the operation of the emitting device is revealed, and the distortion evolution trend under the comprehensive action of heat, electricity, and light can be accurately evaluated, providing an effective basis for the stability analysis of optoelectronic devices.

[0140] The second acquisition module is used to acquire the environmental parameters of the emitting device in real time per unit time. The environmental parameters of the emitting device include environmental temperature, environmental humidity, air flow velocity, electric field strength, magnetic field strength, and interference power. The environmental temperature, environmental humidity, air flow velocity, and electromagnetic interference can be monitored in real time by high-precision sensors. Among them, the environmental temperature is acquired by a thermocouple or a digital temperature sensor, the environmental humidity is measured by a capacitive or optical humidity sensor, and the air flow velocity can be obtained by a hot-wire anemometer or a MEMS micro anemometer. The electric field strength, magnetic field strength, and interference power are monitored by high-frequency electromagnetic field sensors or spectrum analyzers to monitor the intensity and frequency distribution of electromagnetic interference. The environmental parameters of the emitting device refer to the environmental parameters of the environment where the emitting device is located.

[0141] It should be noted that the system brain map corresponding to the second acquisition module is as Figure 8 shown. During the operation of the emitting device, the change of the environmental temperature directly affects the dynamic stability of the PN junction temperature, resulting in the fluctuation of the PN junction temperature under different heat dissipation conditions, and then affecting the stability of the driving current and the dynamic adjustment of the bias voltage. At the same time, it indirectly affects the stability of the output optical power and the beam pointing angle. The change of humidity directly affects the electrical characteristics of the packaging material, resulting in a lag in the driving current response and affecting the compensation effect of the bias voltage on the PN junction temperature. At the same time, in a high-humidity environment, the micro-condensation effect of the optical components is aggravated, resulting in the attenuation of the output optical power and a slight offset of the beam pointing angle. The air flow velocity affects the heat dissipation efficiency, thereby affecting the thermal steady-state control of the PN junction temperature, making the adjustment strategies of the driving current and the bias voltage need to adapt to different air flow environments. At the same time, the high-speed air flow will cause the dynamic drift of the beam pointing angle. Electromagnetic interference directly impacts the stability of the driving current and the bias voltage, triggering short-time current pulses or voltage fluctuations, thereby leading to the instability of the output optical power and affecting the precise control of the beam pointing angle.

[0142] In addition, environmental factors play a decisive role in the dynamic evolution characteristics of the PN junction temperature set, and at the same time affect the change trends of the drive current set and the bias voltage set under different operating conditions, and directly act on the long-term stability of the output optical power set and the beam pointing angle set. The optical power-current curve is affected by environmental factors, showing non-linear changes in the optical power response characteristics under different temperature, humidity, and electromagnetic interference environments, while the offset of the half-wave voltage curve is affected by the PN junction temperature and the air flow heat dissipation conditions. Finally, under the coupling action of multiple environmental factors, the overall change trend of the thermal-electro-optical coupling distortion curve is shifted or the distortion amplitude increases.

[0143] Therefore, by accurately collecting the environmental temperature, humidity, air flow velocity, and electromagnetic interference, based on the collected environmental parameters of the emitting device, combined with the operating characteristic data of the emitting device, the evolution trends of the optical power-current curve, the half-wave voltage curve, and the thermal-electro-optical coupling distortion curve in the future unit time can be predicted. Thus, the operating stability of the emitting device can be effectively optimized, and the reliability and long-term stable operation of the emitting device can be ensured under complex environmental conditions.

[0144] The second processing module extracts features based on the environmental parameters of the emitting device per unit time to obtain the environmental characteristic data of the emitting device.

[0145] The method for obtaining the environmental characteristic data of the emitting device includes:

[0146] Divide the unit time into T time points, construct the environmental temperature set from the environmental temperatures at the T time points, construct the environmental humidity set from the environmental humidities at the T time points, construct the air flow velocity set from the air flow velocities at the T time points, construct the electric field intensity set from the electric field intensities at the T time points, construct the magnetic field intensity set from the magnetic field intensities at the T time points, and construct the interference power set from the interference powers at the T time points;

[0147] Calculate the environmental temperature mean and the environmental temperature variance based on the environmental temperature set; calculate the environmental humidity mean and the environmental humidity variance based on the environmental humidity set; calculate the magnetic field intensity mean and the magnetic field intensity variance based on the magnetic field intensity set;

[0148] Calculate the air flow velocity difference set and the air flow velocity change rate set based on the air flow velocity set;

[0149] Calculate the electric field intensity difference set and the electric field intensity change rate set based on the electric field intensity set;

[0150] Calculate the interference power difference set and the interference power change rate set based on the interference power set;

[0151] Construct the environmental characteristic data of the transmitting device from the mean environmental temperature, the variance of the environmental temperature, the mean environmental humidity, the variance of the environmental humidity, the mean magnetic field intensity, the variance of the magnetic field intensity, the airflow velocity difference set, the airflow velocity change rate set, the electric field intensity difference set, the electric field intensity change rate set, the interference power difference set, and the interference power change rate set.

[0152] It should be noted that in this application, through systematic data processing of the environmental temperature, environmental humidity, airflow velocity, electric field intensity, magnetic field intensity, and interference power collected per unit time, the environmental characteristic data of the transmitting device is constructed to improve the prediction accuracy of the operating characteristics of the transmitting device and the future change trends of optical and electrical performance. Based on the environmental temperature set, environmental humidity set, and magnetic field intensity set, the mean and variance are calculated respectively to characterize the overall level and fluctuation range of the environmental temperature, humidity, and magnetic field intensity, so as to reflect the impact of the long-term environmental trend on the stability of the transmitting device. For environmental parameters with significant dynamic characteristics such as airflow velocity, electric field intensity, and interference power, the first-order difference set and change rate set are calculated to characterize the instantaneous change characteristics of airflow disturbance, electric field intensity mutation, and interference power, so as to capture the key fluctuation factors affecting the operating state of the transmitting device.

[0153] The above environmental characteristic data can effectively describe the impact of environmental variables on the transmitting device per unit time, providing high-dimensional and multi-level environmental impact parameters for modeling. Combining with the operating characteristic data of the transmitting device, a mapping relationship between environmental factors, device operating parameters, and output optical characteristics is established through a machine learning model, enabling the prediction model to make full use of environmental characteristic information, enhancing the sensitivity to the state evolution of the transmitting device, and thus accurately predicting the change trends of future optical power-current curves, half-wave voltage curves, and thermal-electro-optical coupling distortion curves on the unit time scale, providing reliable data support for the stability analysis and performance optimization of the transmitting device.

[0154] The calculation method of the mean environmental temperature includes:

[0155] ;

[0156] Among them, is the mean environmental temperature, is the t-th environmental temperature in the environmental temperature set.

[0157] The calculation method of the variance of the environmental temperature includes:

[0158] ;

[0159] Among them, is the variance of the environmental temperature.

[0160] The calculation method of the mean environmental humidity includes:

[0161] ;

[0162] wherein, is the average environmental humidity, is the t-th environmental humidity in the set of environmental humidities.

[0163] The calculation method of the environmental humidity variance includes:

[0164] ;

[0165] wherein, is the environmental humidity variance.

[0166] The calculation method of the average magnetic field intensity includes:

[0167] ;

[0168] wherein, is the average magnetic field intensity, is the t-th magnetic field intensity in the set of magnetic field intensities.

[0169] The calculation method of the magnetic field intensity variance includes:

[0170] ;

[0171] wherein, is the magnetic field intensity variance.

[0172] The method for obtaining the set of air velocity differences and the set of air velocity change rates includes:

[0173] S400: Let the initial value of t be 1, and the value range of t is from 1 to T - 1; Let the initial value of the counting variable ql be 1, and the value range of ql is from 1 to T - 1;

[0174] S401: Obtain the t-th air velocity and the corresponding time point from the set of air velocities, and obtain the (t + 1)-th air velocity and the corresponding time point from the set of air velocities; Subtract the t-th air velocity from the (t + 1)-th air velocity to obtain the ql-th air velocity difference;

[0175] S402: Add the ql-th air velocity difference to the set of air velocity differences; Calculate the ql-th air velocity change rate based on the ql-th air velocity difference, the time point of the t-th air velocity, and the time point of the (t + 1)-th air velocity; Add the ql-th air velocity change rate to the set of air velocity change rates;

[0176] The calculation method of the air velocity change rate includes:

[0177] ;

[0178] Among them, is the ql-th air velocity change rate, represents the air velocity difference, is the (t + 1)-th air velocity, is the t-th air velocity, is the time point corresponding to the (t + 1)-th air velocity, is the time point corresponding to the t-th air velocity.

[0179] S403: Let t = t + 1. If t is less than or equal to T - 1, then let ql = ql + 1 and continue to execute S401 to S402; if t is greater than T - 1, then end the current process.

[0180] The method for obtaining the electric field strength difference set and the electric field strength change rate set includes:

[0181] S500: Let the initial value of t be 1, and the value range of t is from 1 to T - 1; preset the initial value of the counting variable dc to 1, and the value range of dc is from 1 to T - 1;

[0182] S501: Obtain the t-th electric field strength and the corresponding time point from the electric field strength set, and obtain the (t + 1)-th electric field strength and the corresponding time point from the electric field strength set; subtract the t-th electric field strength from the (t + 1)-th electric field strength to obtain the dc-th electric field strength difference;

[0183] S502: Add the dc-th electric field strength difference to the electric field strength difference set; calculate the dc-th electric field strength change rate based on the dc-th electric field strength difference, the time point of the t-th electric field strength, and the time point of the (t + 1)-th electric field strength; add the dc-th electric field strength change rate to the electric field strength change rate set;

[0184] The calculation method of the electric field strength change rate includes:

[0185] ;

[0186] Among them, is the dc-th electric field strength change rate, represents the electric field strength difference, is the (t + 1)-th electric field strength, is the t-th electric field strength, is the time point corresponding to the (t + 1)-th electric field strength, is the time point corresponding to the t-th electric field strength.

[0187] S503: Let \(t = t + 1\). If \(t\) is less than or equal to \(T - 1\), then let \(dc = dc + 1\) and continue to execute S501 to S502; if \(t\) is greater than \(T - 1\), then end the current process.

[0188] The method for obtaining the interference power difference set and the interference power change rate set includes:

[0189] S600: Let the initial value of \(t\) be 1, and the value range of \(t\) is from 1 to \(T - 1\); preset the initial value of the counting variable \(gl\) to 1, and the value range of \(gl\) is from 1 to \(T - 1\);

[0190] S601: Obtain the \(t\)-th interference power and the corresponding time point from the interference power set, and obtain the \((t + 1)\)-th interference power and the corresponding time point from the interference power set; subtract the \(t\)-th interference power from the \((t + 1)\)-th interference power to obtain the \(gl\)-th interference power difference;

[0191] S602: Add the \(gl\)-th interference power difference to the interference power difference set; calculate the \(gl\)-th interference power change rate based on the \(gl\)-th interference power difference, the time point of the \(t\)-th interference power, and the time point of the \((t + 1)\)-th interference power; add the \(gl\)-th interference power change rate to the interference power change rate set;

[0192] The method for calculating the interference power change rate includes:

[0193] ;

[0194] where is the \(gl\)-th interference power change rate, represents the interference power difference, is the \((t + 1)\)-th interference power, is the \(t\)-th interference power, is the time point corresponding to the \((t + 1)\)-th interference power, is the time point corresponding to the \(t\)-th interference power.

[0195] S603: Let \(t = t + 1\). If \(t\) is less than or equal to \(T - 1\), then let \(gl = gl + 1\) and continue to execute S601 to S602; if \(t\) is greater than \(T - 1\), then end the current process.

[0196] The curve prediction module is used to input the operating characteristic data of the transmitting device and the environmental characteristic data of the transmitting device into the optical characteristic curve prediction model to obtain the set of optical characteristic curves of the transmitting device in the future unit time; the set of optical characteristic curves includes the light power - current curve in the future unit time, the half - wave voltage curve in the future unit time, and the thermal - electro - optical coupling distortion curve in the future unit time.

[0197] The training method of the optical characteristic curve prediction model includes:

[0198] Pre-construct an optical characteristic curve prediction data set, which includes Q groups of optical characteristic curve prediction data and the set of optical characteristic curves corresponding to the Q groups of optical characteristic curve prediction data. Q is a positive integer greater than 0. The optical characteristic curve prediction data includes the operating characteristic data of the emitting device and the environmental characteristic data of the emitting device; divide the optical characteristic curve prediction data set into an optical characteristic curve prediction data training set and an optical characteristic curve prediction data validation set. The optical characteristic curve prediction data training set is used for parameter learning of the optical characteristic curve prediction model, and the optical characteristic curve prediction data validation set is used to evaluate the generalization ability of the optical characteristic curve prediction model in real time;

[0199] During the training process of the optical characteristic curve prediction model, a deep neural network structure based on a multi-layer perceptron is adopted. The optical characteristic curve prediction data is converted into a feature vector as the input, the non-linear features in the data are extracted through the hidden layer, and finally the probability distribution of the set of optical characteristic curves is generated by using the softmax activation function in the output layer. The set of optical characteristic curves corresponding to the maximum probability is output as the final prediction result; the training process aims to minimize the cross-entropy loss function, and at the same time introduces an early stopping strategy to monitor the performance of the optical characteristic curve prediction data validation set. When the prediction accuracy on the optical characteristic curve prediction data validation set reaches the preset accuracy, it is determined that the optical characteristic curve prediction model has converged and the training stops immediately.

[0200] The distortion diagnosis module is used to input the set of optical characteristic curves into the signal distortion diagnosis model to obtain a set of signal distortion diagnosis results for the future unit time. The set of signal distortion diagnosis results includes NUM signal distortion diagnosis result subsets. The signal distortion diagnosis result subset includes the signal distortion serial number interval and the signal distortion type corresponding to the signal distortion serial number interval; the signal distortion serial number interval refers to the time point serial number interval constructed from the start time point to the end time point of the signal distortion; the signal distortion types include amplitude distortion, phase distortion, and memory effect.

[0201] For example, divide the future unit time into 100 time points; there is amplitude distortion between the 10th time point and the 30th time point, that is, the signal distortion type is amplitude distortion, the signal distortion serial number interval is 10 to 30, and the amplitude distortion and 10 to 30 constitute a signal distortion diagnosis result subset. There is phase distortion between the 70th time point and the 80th time point, that is, the signal distortion type is phase distortion, the signal distortion serial number interval is 70 to 80, and the phase distortion and 70 to 80 constitute a signal distortion diagnosis result subset.

[0202] The training method of the signal distortion diagnosis model includes:

[0203] Pre - build a signal distortion diagnosis dataset, where the signal distortion diagnosis dataset includes W sets of signal distortion diagnosis data and a set of signal distortion diagnosis results corresponding to the W sets of signal distortion diagnosis data. W is a positive integer greater than 0. The signal distortion diagnosis data includes an optical characteristic curve set; divide the signal distortion diagnosis dataset into a signal distortion diagnosis data training set and a signal distortion diagnosis data validation set. The signal distortion diagnosis data training set is used for parameter learning of the signal distortion diagnosis model, and the signal distortion diagnosis data validation set is used for real - time evaluation of the generalization ability of the signal distortion diagnosis model;

[0204] During the training process of the signal distortion diagnosis model, adopt a deep neural network structure based on a multi - layer perceptron. Convert the signal distortion diagnosis data into feature vectors as input, extract non - linear features from the data through the hidden layer, and finally generate the probability distribution of the set of signal distortion diagnosis results using the softmax activation function in the output layer. Output the set of signal distortion diagnosis results corresponding to the maximum probability as the final prediction result; the training process aims to minimize the cross - entropy loss function, and at the same time introduce an early stopping strategy to monitor the performance of the signal distortion diagnosis data validation set. When the prediction accuracy rate on the signal distortion diagnosis data validation set reaches the preset accuracy rate, it is determined that the signal distortion diagnosis model has converged, and the training stops immediately.

[0205] It should be noted that based on the optical power - current curve, half - wave voltage curve, and thermal - electro - optical coupling distortion curve in the future unit time, accurate diagnosis of amplitude distortion, phase distortion, and memory effect can be achieved. Among them, the optical power - current curve can characterize the output optical power response characteristics of the emitting device under different drive current conditions. If the optical power - current curve shows non - linear deviation, gain mutation, or saturation effect within the pre - set drive current range, it indicates the existence of amplitude distortion; the half - wave voltage curve is used to depict the voltage - optical conversion characteristics of the emitting device under the action of a modulation signal. If the half - wave voltage curve shows abnormal drift or response lag, it indicates the existence of phase distortion; while the thermal - electro - optical coupling distortion curve is used to reflect the dynamic distortion characteristics under the interaction of temperature, electric field, and optical field. If the thermal - electro - optical coupling distortion curve shows a long - time - scale residual thermal effect or non - transient recovery characteristics, it indicates that the emitting device has a memory effect. In addition, by combining the multi - dimensional features of the optical power - current curve, half - wave voltage curve, and thermal - electro - optical coupling distortion curve, the recognition accuracy of the set of signal distortion diagnosis results can be further improved through the curve change trend, historical dependence, and hysteresis response mode, providing strong support for the subsequent optimization of the stability of the emitting device.

[0206] The intelligent optimization module performs digital pre-distortion intelligent adaptive adjustment on the transmitting device based on the operating characteristic data of the transmitting device, the set of optical characteristic curves in the future unit time, and the set of signal distortion diagnosis results, so as to dynamically compensate for signal distortion.

[0207] As Figure 3 shown, the method for performing digital pre-distortion intelligent adaptive adjustment on the transmitting device includes:

[0208] S700: Denote the number of subsets of signal distortion diagnosis results in the set of signal distortion diagnosis results as NUM, set the initial value of the counting variable num to 1, and the value range of num is from 1 to NUM;

[0209] S701: Obtain the num-th subset of signal distortion diagnosis results from the set of signal distortion diagnosis results;

[0210] If the signal distortion type in the num-th subset of signal distortion diagnosis results is amplitude distortion, perform digital pre-distortion intelligent adaptive adjustment for amplitude distortion;

[0211] If the signal distortion type in the num-th subset of signal distortion diagnosis results is phase distortion, perform digital pre-distortion intelligent adaptive adjustment for phase distortion;

[0212] If the signal distortion type in the num-th subset of signal distortion diagnosis results is memory effect, perform digital pre-distortion intelligent adaptive adjustment for memory effect;

[0213] S702: Let num = num + 1. If num is less than or equal to NUM, continue to execute S701. If num is greater than NUM, complete the digital pre-distortion intelligent adaptive adjustment and end the current process.

[0214] As Figure 4 shown, the method for performing digital pre-distortion intelligent adaptive adjustment for amplitude distortion includes:

[0215] Numerically label the amplitude distortion to obtain the amplitude distortion value; for example, the amplitude distortion can be labeled as 1, that is, the amplitude distortion value is 1;

[0216] Input the amplitude distortion value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves in the future unit time into the amplitude distortion adjustment model to obtain the amplitude distortion gain coefficient;

[0217] Calculate the digitally pre-distorted output signal corresponding to the amplitude distortion based on the amplitude distortion gain coefficient, and denote it as the amplitude distortion output signal;

[0218] Obtain the signal distortion serial number interval corresponding to amplitude distortion from the signal distortion diagnosis result subset. When the signal output moment of the transmitting device matches the signal distortion serial number interval of amplitude distortion, use the amplitude distortion output signal as the output signal of the transmitting device to dynamically compensate for amplitude distortion, and complete the digital pre-distortion intelligent adaptive adjustment for amplitude distortion.

[0219] The training method of the amplitude distortion adjustment model includes:

[0220] Pre-construct an amplitude distortion adjustment data set, which includes R groups of amplitude distortion adjustment data and the amplitude distortion gain coefficients corresponding to the R groups of amplitude distortion adjustment data. R is a positive integer greater than 0. The amplitude distortion adjustment data includes amplitude distortion values, transmitting device operation characteristic data, and an optical characteristic curve set; divide the amplitude distortion adjustment data set into an amplitude distortion adjustment data training set and an amplitude distortion adjustment data verification set. The amplitude distortion adjustment data training set is used for parameter learning of the amplitude distortion adjustment model, and the amplitude distortion adjustment data verification set is used to evaluate the generalization ability of the amplitude distortion adjustment model in real time;

[0221] During the training process of the amplitude distortion adjustment model, adopt a deep neural network structure based on a multi-layer perceptron. Convert the amplitude distortion adjustment data into a feature vector as the input, extract the non-linear features in the data through the hidden layer, and finally generate the probability distribution of the amplitude distortion gain coefficient using the softmax activation function in the output layer. Output the amplitude distortion gain coefficient corresponding to the maximum probability as the final prediction result; the training process aims to minimize the cross-entropy loss function, and at the same time introduce an early stopping strategy to monitor the performance of the amplitude distortion adjustment data verification set. When the prediction accuracy rate on the amplitude distortion adjustment data verification set reaches the preset accuracy rate, it is determined that the amplitude distortion adjustment model has converged, and the training stops immediately.

[0222] The method for obtaining the amplitude distortion output signal includes:

[0223] ;

[0224] Among them, is the amplitude distortion output signal, is the original input signal corresponding to the amplitude distortion, is the amplitude distortion gain coefficient.

[0225] It should be noted that according to the detected amplitude distortion degree, an appropriate amplitude distortion gain coefficient is dynamically obtained, and corresponding gain adjustment is applied at the signal input end to offset the amplitude distortion caused by the non-linear characteristics of the transmitting device. Through this compensation method, it is ensured that the signal amplitude can remain stable after passing through the transmitting device, thereby improving the signal quality of the optical communication system and enhancing the overall reliability of the system, enabling the compensation strategy to be precisely adjusted for different amplitude distortion degrees and meeting the requirements of high-precision optical signal transmission.

[0226] As Figure 5 shown, the method for digital pre-distortion intelligent adaptive adjustment for phase distortion includes:

[0227] Numerically label the phase distortion to obtain the phase distortion value; for example, the phase distortion can be labeled as 2, that is, the phase distortion value is 2;

[0228] Input the phase distortion value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves for the next unit time into the phase distortion adjustment model to obtain the phase distortion compensation angle;

[0229] Calculate the output signal after digital pre-distortion corresponding to the phase distortion based on the phase distortion compensation angle, denoted as the phase distortion output signal;

[0230] Obtain the signal distortion serial number interval corresponding to the phase distortion from the signal distortion diagnosis result subset. When the signal output time of the transmitting device matches the signal distortion serial number interval of the phase distortion, use the phase distortion output signal as the output signal of the transmitting device to dynamically compensate for the phase distortion and complete the digital pre-distortion intelligent adaptive adjustment for the phase distortion.

[0231] The training method of the phase distortion adjustment model includes:

[0232] Pre-construct a phase distortion adjustment data set, which includes G groups of phase distortion adjustment data and the corresponding phase distortion compensation angles for the G groups of phase distortion adjustment data. G is a positive integer greater than 0. The phase distortion adjustment data includes the phase distortion value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves; divide the phase distortion adjustment data set into a phase distortion adjustment data training set and a phase distortion adjustment data verification set. The phase distortion adjustment data training set is used for parameter learning of the phase distortion adjustment model, and the phase distortion adjustment data verification set is used for real-time evaluation of the generalization ability of the phase distortion adjustment model;

[0233] During the training process of the phase distortion adjustment model, a deep neural network structure based on a multi-layer perceptron is adopted. The phase distortion adjustment data is converted into feature vectors as inputs. Nonlinear features in the data are extracted through the hidden layer, and finally, the probability distribution of the phase distortion compensation angle is generated using the softmax activation function in the output layer. The phase distortion compensation angle corresponding to the maximum probability is output as the final prediction result. The training process aims to minimize the cross-entropy loss function and simultaneously introduces an early stopping strategy to monitor the performance of the phase distortion adjustment data validation set. When the prediction accuracy on the phase distortion adjustment data validation set reaches the preset accuracy, it is determined that the phase distortion adjustment model has converged, and the training stops immediately.

[0234] The calculation method of the phase distortion output signal includes:

[0235] ;

[0236] where, is the phase distortion output signal, is the original input signal corresponding to the phase distortion, e is a constant, is the imaginary unit, represents a clockwise rotation, is the phase distortion compensation angle.

[0237] It should be noted that the core idea of the phase distortion output signal is to use the exponential rotation factor to perform phase pre-compensation on the original input signal corresponding to the phase distortion, correct the phase shift generated during signal transmission, so as to restore the ideal phase state as much as possible when the signal reaches the receiving end. To effectively compensate for the phase distortion problem caused by non-linear transmission, a phase adjustment method based on the phase distortion compensation angle is adopted to ensure that the phase of the output signal of the transmitting device remains within the expected range. Since the input signal has been pre-adjusted before transmission, the phase of the finally received signal is closer to the expected value of the original signal, thereby reducing the impact of phase distortion.

[0238] As Figure 6 shown, the method for digital pre-distortion intelligent adaptive adjustment for the memory effect includes:

[0239] Numerically label the memory effect to obtain the memory effect value; for example, the memory effect can be labeled as 3, that is, the memory effect value is 3;

[0240] Input the memory effect value, the operating characteristic data of the transmitting device, and the set of optical characteristic curves for the next unit time into the memory effect adjustment model to obtain the memory depth and the memory effect weight; [[ID=****]]

[0241] The output signal after digital predistortion corresponding to the memory effect calculated based on the memory depth and the memory effect weight is denoted as the memory effect output signal;

[0242] Obtain the signal distortion serial number interval corresponding to the memory effect from the signal distortion diagnosis result subset. When the signal output moment of the transmitting device matches the signal distortion serial number interval of the memory effect, use the memory effect output signal as the output signal of the transmitting device to dynamically compensate for the memory effect and complete the digital predistortion intelligent adaptive adjustment for the memory effect.

[0243] The training method of the memory effect adjustment model includes:

[0244] Pre-construct a memory effect adjustment data set, which includes E groups of memory effect adjustment data and the memory depth and memory effect weight corresponding to the E groups of memory effect adjustment data. E is a positive integer greater than 0. The memory effect adjustment data includes memory effect values, transmitting device operation characteristic data, and an optical characteristic curve set; divide the memory effect adjustment data set into a memory effect adjustment data training set and a memory effect adjustment data validation set. The memory effect adjustment data training set is used for parameter learning of the memory effect adjustment model, and the memory effect adjustment data validation set is used for real-time evaluation of the generalization ability of the memory effect adjustment model; [[ID=X]] [[ID=X]]

[0245] During the training process of the memory effect adjustment model, adopt a deep neural network structure based on a multi-layer perceptron. Convert the memory effect adjustment data into a feature vector as the input, extract the non-linear features in the data through the hidden layer, and finally generate the probability distribution of the memory depth and the memory effect weight using the softmax activation function in the output layer. Output the memory depth and the memory effect weight corresponding to the maximum probability as the final prediction result; the training process aims to minimize the cross-entropy loss function, and at the same time introduce an early stopping strategy to monitor the performance of the memory effect adjustment data validation set. When the prediction accuracy on the memory effect adjustment data validation set reaches the preset accuracy, it is determined that the memory effect adjustment model has converged and the training stops immediately.

[0246] The method for obtaining the memory effect output signal includes:

[0247] ;

[0248] where, is the memory effect output signal, is the original input signal corresponding to the memory effect, is the memory depth, is the memory effect weight, represents the input signal before the current moment at time, represents the current moment, and k represents the index of the time step for backtracking. is the sampling time interval of the memory effect preset for those skilled in the art. is used to track the input signal of the emission device at the k-th sampling time interval before the current moment. When calculating the memory effect at the current moment, the influence degree of the historical input signal on the current output is considered.

[0249] It should be noted that the current signal is affected by the input signals at multiple past time points, and the historical input state of the signal will affect the current output state. This is the "memory effect", and the weight of this influence is determined by the memory effect weight. If the memory effect weight is larger, it means that the past time points have a stronger influence on the current signal. The memory effect will cause signal distortion. Therefore, these influences need to be pre-compensated in the input signal to make the final output signal as close to the ideal state as possible.

[0250] Digital pre-distortion is a signal processing technology mainly used to solve the problem of signal distortion caused by the non-linear characteristics of hardware devices during the transmission process in a communication system. For example, devices such as infrared emitters, lasers, and RF power amplifiers will distort the signal and generate distortion when transmitting high-intensity signals. The core idea of pre-distortion is to artificially "distort" the signal in advance (i.e., "pre-distort") before the signal is sent to the hardware and then hand it over to the hardware for transmission. Assuming the hardware will "lower the high notes", then the pre-distortion will "raise the high notes" in advance. When the signal passes through the hardware, the two distortions cancel each other out, and the final output is close to the original signal.

[0251] Embodiment 2

[0252] Please refer to Figure 2 as shown, this embodiment provides an infrared narrow beam emission communication control method, including:

[0253] Real-time collect the operating parameters of the emission device per unit time;

[0254] Based on the operating parameters of the emission device per unit time, perform feature extraction to obtain the operating feature data of the emission device;

[0255] Real-time collect the environmental parameters of the emission device per unit time;

[0256] Based on the environmental parameters of the emission device per unit time, perform feature extraction to obtain the environmental feature data of the emission device;

[0257] Input the operating feature data and environmental feature data of the emission device into the optical characteristic curve prediction model to obtain the set of optical characteristic curves of the emission device in the future unit time;

[0258] Input the set of optical characteristic curves into the signal distortion diagnosis model to obtain the set of signal distortion diagnosis results in the future unit time;

[0259] Based on the operating characteristic data of the transmitting device, the set of optical characteristic curves in the future unit time, and the set of signal distortion diagnosis results, digital pre-distortion intelligent adaptive adjustment is performed on the transmitting device to dynamically compensate for signal distortion.

[0260] As mentioned above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claimed rights.

[0261] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. Infrared narrow beam emission communication control method, characterized in that Including: Performing feature extraction on the operating parameters of the transmitting device in real-time per unit time to obtain the operating feature data of the transmitting device; Performing feature extraction on the environmental parameters of the transmitting device in real-time per unit time to obtain the environmental feature data of the transmitting device; Inputting the operating feature data of the transmitting device and the environmental feature data of the transmitting device into the optical characteristic curve prediction model to obtain a set of optical characteristic curves of the transmitting device in the future per unit time; Inputting the set of optical characteristic curves into the signal distortion diagnosis model to obtain a set of signal distortion diagnosis results in the future per unit time; The set of signal distortion diagnosis results includes NUM subsets of signal distortion diagnosis results, and each subset of signal distortion diagnosis results includes a signal distortion serial number interval and the corresponding signal distortion type; The signal distortion type includes amplitude distortion, phase distortion, and memory effect; Performing digital pre-distortion intelligent adaptive adjustment on the transmitting device based on the operating feature data of the transmitting device, the set of optical characteristic curves in the future per unit time, and the set of signal distortion diagnosis results to dynamically compensate for signal distortion; The method for performing digital pre-distortion intelligent adaptive adjustment for amplitude distortion includes: Numerically annotating the amplitude distortion to obtain an amplitude distortion value; Inputting the amplitude distortion value, the operating feature data of the transmitting device, and the set of optical characteristic curves in the future per unit time into the amplitude distortion adjustment model to obtain an amplitude distortion gain coefficient; Calculating the output signal after digital pre-distortion corresponding to the amplitude distortion based on the amplitude distortion gain coefficient, denoted as the amplitude distortion output signal; Obtaining the signal distortion serial number interval corresponding to the amplitude distortion from the subset of signal distortion diagnosis results, and when the signal output time of the transmitting device matches the signal distortion serial number interval of the amplitude distortion, using the amplitude distortion output signal as the output signal of the transmitting device to dynamically compensate for the amplitude distortion and complete the digital pre-distortion intelligent adaptive adjustment for the amplitude distortion.

2. The infrared narrow beam emission communication control method according to claim 1, characterized in that The method for performing digital pre-distortion intelligent adaptive adjustment on the transmitting device includes: S700: Denote the number of subsets of signal distortion diagnosis results in the set of signal distortion diagnosis results as NUM, set the initial value of the counting variable num to 1, and the value range of num is from 1 to NUM; S701: Obtain the num-th subset of signal distortion diagnosis results from the set of signal distortion diagnosis results; If the signal distortion type in the num-th subset of signal distortion diagnosis results is amplitude distortion, perform digital pre-distortion intelligent adaptive adjustment for the amplitude distortion; If the signal distortion type in the num-th subset of signal distortion diagnosis results is phase distortion, perform digital pre-distortion intelligent adaptive adjustment for the phase distortion; If the signal distortion type in the num-th subset of signal distortion diagnosis results is memory effect, perform digital pre-distortion intelligent adaptive adjustment for the memory effect; S702: Let num = num + 1. If num is less than or equal to NUM, continue to execute S701. If num is greater than NUM, complete the digital pre-distortion intelligent adaptive adjustment and end the current process.

3. The infrared narrow beam emission communication control method according to claim 2, wherein The method for performing digital pre-distortion intelligent adaptive adjustment for phase distortion includes: Numerically label the phase distortion to obtain the phase distortion value; Input the phase distortion value, the operating characteristic data of the emitting device, and the set of optical characteristic curves for the future unit time into the phase distortion adjustment model to obtain the phase distortion compensation angle; Calculate the output signal after digital predistortion corresponding to the phase distortion based on the phase distortion compensation angle, denoted as the phase distortion output signal; Obtain the signal distortion serial number interval corresponding to the phase distortion from the signal distortion diagnosis result subset. When the signal output time of the emitting device matches the signal distortion serial number interval of the phase distortion, use the phase distortion output signal as the output signal of the emitting device to dynamically compensate for the phase distortion and complete the digital predistortion intelligent adaptive adjustment for the phase distortion.

4. The infrared narrow beam emission communication control method according to claim 2, wherein The method for digital predistortion intelligent adaptive adjustment for the memory effect includes: Numerically label the memory effect to obtain the memory effect value; Input the memory effect value, the operating characteristic data of the emitting device, and the set of optical characteristic curves for the future unit time into the memory effect adjustment model to obtain the memory depth and the memory effect weight; Calculate the output signal after digital predistortion corresponding to the memory effect based on the memory depth and the memory effect weight, denoted as the memory effect output signal; Obtain the signal distortion serial number interval corresponding to the memory effect from the signal distortion diagnosis result subset. When the signal output time of the emitting device matches the signal distortion serial number interval of the memory effect, use the memory effect output signal as the output signal of the emitting device to dynamically compensate for the memory effect and complete the digital predistortion intelligent adaptive adjustment for the memory effect.

5. The infrared narrow beam emission communication control method according to claim 1, wherein The operating parameters of the emitting device include the PN junction temperature, drive current, bias voltage, output optical power, and beam pointing angle; The method for obtaining the operating characteristic data of the emitting device includes: Divide the unit time into T time points, construct the PN junction temperature set from the PN junction temperatures of the T time points, construct the drive current set from the drive currents of the T time points, construct the bias voltage set from the bias voltages of the T time points, construct the output optical power set from the output optical powers of the T time points, and construct the beam pointing angle set from the beam pointing angles of the T time points; Perform optical power-current efficiency calculation based on the output optical power set and the drive current set to obtain the optical power-current curve; Perform half-wave voltage calculation based on the output optical power set and the bias voltage set to obtain the half-wave voltage curve; Perform thermal-electro-optical coupling distortion analysis based on the PN junction temperature set, the drive current set, and the beam pointing angle set to obtain the thermal-electro-optical coupling distortion curve; Construct the operating characteristic data of the emitting device from the PN junction temperature set, the drive current set, the bias voltage set, the output optical power set, the beam pointing angle set, the optical power-current curve, the half-wave voltage curve, and the thermal-electro-optical coupling distortion curve.

6. The infrared narrow beam emission communication control method according to claim 5, characterized in that The method for obtaining the optical power-current curve includes: S100: Let the initial value of the time step variable t be 1, and the value range of t is from 1 to T - 1; preset the initial value of the counting variable bl to 1, and the value range of bl is from 1 to T - 1; S101: Obtain the t-th and (t + 1)-th output optical powers from the set of output optical powers, subtract the t-th output optical power from the (t + 1)-th output optical power to obtain the bl-th output optical power change difference; obtain the t-th and (t + 1)-th drive currents from the set of drive currents, subtract the t-th drive current from the (t + 1)-th drive current to obtain the bl-th drive current change difference; S102: Divide the bl-th output optical power change difference by the drive current change difference to obtain the optical power-current efficiency; S103: Add the optical power-current efficiency to the set of optical power-current efficiencies; S104: Let t = t + 1. If t is less than or equal to T - 1, then let bl = bl + 1, and continue to execute S101 to S103; if t is greater than T - 1, then execute S105; S105: Use the bl counting variables as the horizontal axis data, and the bl optical power-current efficiencies in the set of optical power-current efficiencies as the vertical axis data to plot the optical power-current curve, and end the current process.

7. The infrared narrow beam emission communication control method according to claim 6, wherein The method for obtaining the half-wave voltage curve includes: S200: Preset the initial value of the time step variable bc to 1, and the value range of bc is from 1 to T; S201: Obtain the bc-th bias voltage from the set of bias voltages, and obtain the bc-th output optical power from the set of output optical powers; S202: Calculate the corresponding half-wave voltage based on the bc-th bias voltage and output optical power; S203: Add the bc-th half-wave voltage to the set of half-wave voltages; S204: Let bc = bc + 1. If bc is less than or equal to T, then continue to execute S201 to S203; if bc is greater than T, then execute S205; S205: Use the bc time step variables as the horizontal axis data, and the bc half-wave voltages in the set of half-wave voltages as the vertical axis data to plot the half-wave voltage curve, and end the current process.

8. The infrared narrow beam emission communication control method according to claim 5, characterized in that, The method for obtaining the thermal-electro-optical coupling distortion curve includes: S300: Let the initial value of the time step variable t be 1, and the value range of t is from 1 to T - 1; preset the initial value of the counting variable bd to 1, and the value range of bd is from 1 to T - 1; S301: Obtain the t-th and (t + 1)-th PN junction temperatures from the set of PN junction temperatures, subtract the t-th PN junction temperature from the (t + 1)-th PN junction temperature to obtain the bd-th PN junction temperature difference; obtain the t-th and (t + 1)-th drive currents from the set of drive currents, calculate the average value of the t-th and (t + 1)-th drive currents, denoted as the drive current average value, to obtain the bd-th drive current average value; obtain the t-th and (t + 1)-th beam pointing angles from the set of beam pointing angles, subtract the t-th beam pointing angle from the (t + 1)-th beam pointing angle to obtain the bd-th beam pointing angle difference; the time points of the (t + 1)-th PN junction temperature, the (t + 1)-th drive current, and the (t + 1)-th beam pointing angle are the same; S302: Calculate the bd-th thermal-electro-optical coupling distortion degree based on the bd-th PN junction temperature difference, drive current average value, beam pointing angle difference, and time interval; S303: Add the bd-th thermo-electro-optic coupling distortion degree to the thermo-electro-optic coupling distortion set; S304: Let t = t + 1. If t is less than or equal to T - 1, then let bd = bd + 1, and continue to execute S301 to S303; if t is greater than T - 1, then execute S305; S305: Use the bd counting variables as the horizontal axis data, and the bd thermo-electro-optic coupling distortion degrees in the thermo-electro-optic coupling distortion set as the vertical axis data to plot the thermo-electro-optic coupling distortion curve, and end the current process.

9. The infrared narrow beam emission communication control method according to claim 1, characterized in that, The environmental parameters of the emitting device include environmental temperature, environmental humidity, air flow velocity, electric field strength, magnetic field strength, and interference power; The method for obtaining the environmental characteristic data of the emitting device includes: Divide the unit time into T time points, construct the environmental temperature set from the environmental temperatures of the T time points, construct the environmental humidity set from the environmental humidities of the T time points, construct the air flow velocity set from the air flow velocities of the T time points, construct the electric field strength set from the electric field strengths of the T time points, construct the magnetic field strength set from the magnetic field strengths of the T time points, and construct the interference power set from the interference powers of the T time points; Calculate the environmental temperature mean and environmental temperature variance based on the environmental temperature set; calculate the environmental humidity mean and environmental humidity variance based on the environmental humidity set; calculate the magnetic field strength mean and magnetic field strength variance based on the magnetic field strength set; Calculate the air flow velocity difference set and air flow velocity change rate set based on the air flow velocity set; Calculate the electric field strength difference set and electric field strength change rate set based on the electric field strength set; Calculate the interference power difference set and interference power change rate set based on the interference power set; Construct the environmental characteristic data of the emitting device from the environmental temperature mean, environmental temperature variance, environmental humidity mean, environmental humidity variance, magnetic field strength mean, magnetic field strength variance, air flow velocity difference set, air flow velocity change rate set, electric field strength difference set, electric field strength change rate set, interference power difference set, and interference power change rate set.

10. An infrared narrow beam emission communication control system for implementing the infrared narrow beam emission communication control method according to any one of claims 1-9, characterized in that, It includes: The first processing module extracts features from the operating parameters of the emitting device collected in real time per unit time to obtain the operating characteristic data of the emitting device; The second processing module extracts features from the environmental parameters of the emitting device collected in real time per unit time to obtain the environmental characteristic data of the emitting device; The curve prediction module is used to input the operating characteristic data of the emitting device and the environmental characteristic data of the emitting device into the optical characteristic curve prediction model to obtain the optical characteristic curve set of the emitting device in the future unit time; The distortion diagnosis module is used to input the optical characteristic curve set into the signal distortion diagnosis model to obtain the signal distortion diagnosis result set in the future unit time; The intelligent optimization module performs digital pre-distortion intelligent adaptive adjustment on the emitting device based on the operating characteristic data of the emitting device, the optical characteristic curve set in the future unit time, and the signal distortion diagnosis result set to dynamically compensate for signal distortion.

Images