A satellite communication frequency offset estimation method, system, medium and electronic equipment based on a power terminal system

Abstract

The application provides a satellite communication frequency offset estimation method and system based on a power terminal system, which comprises the following steps: input parameter preprocessing is performed on a GNSS-free power terminal; a Doppler-geography joint likelihood function is constructed by fusing a terminal position Gaussian prior and a beam coverage constraint, and a coarse estimation result is obtained; each power terminal performs frequency offset pre-compensation on a received signal based on the coarse estimation result, extracts a residual frequency offset of a single power terminal, and obtains a corresponding frequency offset estimation variance; each power terminal in a power district exchanges its residual frequency offset and the corresponding frequency offset estimation variance, and performs weighted fusion on the exchanged residual frequency offset and the frequency offset estimation variance to obtain a fused residual frequency offset estimation result; joint adaptive optimization: a corresponding service unlocking penalty coefficient is matched, a Bayesian cost function is constructed, and an optimal fine estimation detection range is obtained by solving; when the optimal fine estimation detection range covers a coarse estimation error variance, an optimal frequency offset estimation is output.

Description

Technical Field

[0001] This invention relates to the field of information and communication engineering technology, and more specifically, to a satellite communication frequency offset estimation method, system, medium, and electronic equipment based on a power terminal system. Background Technology

[0002] In the NTN standardization work of Release 17, 3GPP established a time-frequency offset pre-compensation framework: satellites pre-compensate for common frequency offset when transmitting downlink signals, typically the Doppler frequency shift from the satellite to the beam center point; terminals pre-compensate for common time offset when transmitting preamble signals. The core purpose of this framework is to reduce the range and complexity of time-frequency offsets that need to be handled during terminal synchronization. However, the effectiveness of this pre-compensation mechanism is based on the fundamental assumption that the terminal has GNSS positioning capabilities. 3GPP Release 17 explicitly requires each NTN terminal to be equipped with a GNSS chipset capable of determining its own position and velocity, and calculating the required frequency adjustment value accordingly. To achieve uplink synchronization, before random access, the terminal needs to autonomously pre-compensate for time lead (TA), and simultaneously calculate the Doppler frequency offset by combining the satellite position and its own position, with GNSS accuracy within a 10-meter error range.

[0003] Low Earth Orbit (LEO) satellites move at extremely high relative speeds, resulting in significant Doppler shift effects that cause drastic time-varying frequency offsets and phase variations in the received signal. This high dynamic characteristic poses a severe challenge to traditional synchronization algorithms—conventional acquisition methods fail when the rate of frequency change exceeds the frequency resolution within the intercept time. Furthermore, the elevation angle of the LEO satellite relative to the ground terminal continuously changes, causing the frequency offset to exhibit nonlinear variations as the satellite passes overhead, further increasing the estimation difficulty. Therefore, high-precision, wide-range, and low-complexity frequency offset estimation methods for LEO scenarios have become one of the core issues in NTN physical layer research.

[0004] If the terminal lacks GNSS capability, Doppler shift pre-compensation cannot be performed. This assumption faces severe challenges in vertical application scenarios such as the power industry—a large number of power terminals are deployed in environments where GNSS signals are obstructed, such as basements and distribution cabinets, or when disasters cause GNSS service interruptions, the standard pre-compensation mechanism will completely fail.

[0005] However, existing methods for estimating frequency offset without GNSS still have significant limitations: beam pointing-based methods show significantly increased errors in high elevation angle (satellite overhead) scenarios; dynamic reference point-based methods rely on specific network infrastructure modifications; and AI / ML methods face problems such as large training data requirements and insufficient cross-scenario generalization ability. No research has yet systematically combined GNSS-free frequency offset estimation with the asset data and communication infrastructure unique to the power industry. Summary of the Invention

[0006] This invention proposes a satellite communication frequency offset estimation method and system based on a power terminal system to solve the technical problems of traditional frequency offset estimation, which is difficult to balance unlocking reliability and computational complexity, cannot adapt to the differentiated needs of power services, and has poor estimation accuracy and scenario adaptability.

[0007] To address the aforementioned problems, according to one aspect of the present invention, a satellite communication frequency offset estimation method based on a power terminal system is provided, the method comprising: Input parameter preprocessing is performed for power terminals without GNSS; the input parameters include: beam pointing constraints, preset GIS coordinates, and relative positions of the substation terminals; Based on the preprocessed input parameters, the Doppler-geographic joint likelihood function is constructed by fusing the Gaussian prior of the terminal location and the beam coverage constraint, and the coarse estimation results and coarse estimation error variance of each power terminal are output. Each power terminal performs frequency offset pre-compensation on the received signal based on the coarse estimation result, extracts the residual frequency offset of a single power terminal based on the fine frequency offset estimation method, and obtains the frequency offset estimation variance corresponding to the residual frequency offset. Each power terminal within the power distribution area exchanges its own residual frequency offset and corresponding frequency offset estimation variance, and then performs weighted fusion of the exchanged residual frequency offset and frequency offset estimation variance to obtain the fused residual frequency offset estimation result. Based on the priority of power business, we divide the business into multiple levels and match the corresponding business unlocking penalty coefficient. We construct a Bayesian cost function that integrates unlocking probability and computational complexity, and solve it to obtain the optimal precise estimation detection range. When the optimal fine estimation detection range covers the variance of the coarse estimation error, the optimal frequency offset estimate is output based on the coarse estimation result and the fused residual frequency offset estimation result.

[0008] Preferably, the input parameter preprocessing includes: beam pointing constraint preprocessing, power GIS prior modeling, and transformer terminal relative position modeling; The beam pointing constraint preprocessing is as follows: Based on the satellite's position, orbital altitude, and half-power beamwidth, the pointing angle of the satellite broadcast beam center is determined. This leads to the calibration of the beam center ground point and the satellite beam's ground coverage area, forming the beam coverage boundary constraint for power terminal distribution, expressed as: in, For the ground coverage area of ​​the beam, To calibrate the ground point of the beam center, For satellite position, Given the actual location of the power terminal, , For the orbital height, Half-power beamwidth; The prior modeling of the power GIS is as follows: A Gaussian prior distribution of the terminal location is established based on the pre-set GIS coordinates of the power terminal. The actual location of the power terminal follows a Gaussian distribution with the pre-set GIS coordinates as the mean and a pre-set covariance matrix as the constraint. The covariance matrix comprehensively considers GIS input error, power terminal installation deviation, and equivalent electrical length error of signal penetration through power equipment, and is expressed as: Set up power terminal The GIS preset coordinates are The actual location of GIS preset coordinates Follows a Gaussian prior distribution at the terminal position: Wherein, the covariance matrix , , , Typical values; The relative positions of the terminals in the distribution area are modeled as follows: Within the Taiwan area The geometric center of the power terminal area is: Any two power terminals The relative positions are known as follows: Power Terminal The actual position is represented by the terminal relative position constraint parameters: in, This is the nominal offset from the center of the distribution area to the power terminal. For individual random offsets.

[0009] Preferably, based on the preprocessed input parameters, the Doppler-geographic joint likelihood function constructed by fusing the Gaussian prior of the terminal location and the beam coverage constraint outputs the coarse estimation results and coarse estimation error variance for each power terminal, including: Based on satellite orbital speed With the center speed of the platform Calculate the theoretical Doppler frequency offset of the transformer center: in, At the speed of light, For carrier frequency; Constructing the Doppler-geographic joint likelihood function includes: Based on the given actual location of the power terminal The theoretical Doppler is determined as follows: in, The three-dimensional velocity vector of the power terminal; Considering measurement noise and beam pointing uncertainty, construct the actual Doppler observations of the received signal. Conditional probability density: in, Typical values; Combining the Gaussian prior distribution of the terminal location and the beam coverage boundary constraints, the posterior likelihood of a single power terminal is obtained: in, For the three-dimensional geographic space area covered by the beam; Based on the relative position constraints of the terminals in the distribution area, the posterior likelihood of a single power terminal is extended to the joint likelihood of multiple power terminals in the distribution area: in, for, For the actual observed Doppler frequency shift of the i-th terminal, This is the common positional deviation vector for all terminals in the distribution area; Allowable frequency offset range Discretize into One grid point, Step size for each grid cell; Calculate the joint likelihood of multiple power terminals point by point, select the frequency offset parameter corresponding to the maximum joint likelihood value as the coarse estimation result of the power terminal, and output the corresponding coarse estimation error variance simultaneously: in, Let the assumed Doppler frequency offset be taken as the independent variable. For discrete grid search space, It is the joint posterior likelihood function of multiple power terminals.

[0010] Preferably, the frequency offset estimation method includes single-terminal residual frequency offset calculation and multi-terminal collaborative fusion: The coarse estimation result is used to pre-compensate the frequency offset of the received signal. The autocorrelation characteristics of the OFDM symbol cyclic prefix and symbol tail are used to perform single-terminal fine frequency offset estimation on the small residual frequency offset after compensation. The residual frequency offset calculation formula is constructed based on the number of FFT points, cyclic prefix length and sampling period, and the frequency offset estimate of a single power terminal is obtained by solving it. According to the Cramer-Rao lower bound criterion, the single-terminal frequency offset estimation variance corresponding to the frequency offset estimate is calculated. Each power terminal within the power distribution area exchanges its own residual frequency offset estimate and corresponding frequency offset estimate variance; using the frequency offset estimate variance of each power terminal as the weighting weight, the frequency offset estimates of all power terminals within the power distribution area are weighted and fused to obtain the distribution area fused residual frequency offset estimate result, and the distribution area fused estimate variance corresponding to the fused residual frequency offset estimate result is solved.

[0011] Preferably, the step of dividing services into multiple levels according to power service priority and matching corresponding service unlocking penalty coefficients includes: Priority classification mapping is performed on the currently activated power services of the power terminal. Based on the difference in tolerance for frequency offset unlocking of different power services, the corresponding level of service unlocking penalty coefficient is matched, and the service unlocking penalty coefficient is uniquely determined by the highest priority service among all currently activated power services. Preferably, constructing the Bayesian cost function that integrates the unlock probability and computational complexity includes: Constructing a Bayesian cost function that integrates unlock probability and computational complexity includes unlock probability modeling, computational complexity modeling, and Bayesian cost function construction: The unlocking probability model is as follows: A fine-grained frequency offset estimation effective detection range is defined. Based on the coarse estimation error variance corresponding to the coarse estimation result, the frequency offset fluctuation range of the coarse estimation result is characterized. When the frequency offset fluctuation range exceeds the fine-grained frequency offset estimation effective detection range, it is determined that the residual frequency offset fine-grained estimation lock-in of the power terminal has failed, and frequency offset unlocking has occurred. Based on the current coarse estimation error variance and the fine estimation detection range to be optimized, the real-time unlocking probability is calculated. The unlocking probability calculation model is expressed as follows: Coarse estimation error variance The effective detection range of the frequency offset fine estimation to be optimized is: ,like Calculate the real-time unlock probability: in, , Let the lower limit of integration be the threshold, and let the independent variable be... The variable to be integrated represents all possible values ​​that a standard normally distributed random variable can take. After integration, these values ​​will be eliminated. To roughly estimate the error variance, To accurately estimate the effective detection range for frequency offset; The computational complexity is modeled as follows: Based on the effective detection range of the frequency offset fine estimation to be optimized, the number of frequency points that the frequency offset fine estimation method needs to search is determined. Combining this with the computational complexity of the multiplication operations corresponding to a single frequency point, the total computational complexity of the fine estimation process is derived. The number of frequency points required for the frequency offset estimation method: in, This is the search step size; The computational cost for each frequency point is determined as follows: Total computational complexity: By introducing the business unlocking penalty coefficient corresponding to the business priority obtained from real-time monitoring, and fusing the unlocking probability and total computational complexity, a Bayesian cost function for the total cost is constructed, and the feasible region of the precise estimation detection range is limited: in, For the total cost, Penalty coefficient for business unlocking, This represents the probability of unlocking; The feasible region for precise estimation of the detection range is , in, The minimum detectable value can be set to 200Hz. The maximum detectable value can be set to 2000Hz; Based on the convex function property of the Bayesian cost function, the optimal precise estimation of the detection range is obtained by using the bisection method or a derivative approximation algorithm: .

[0012] Preferably, when the optimal fine estimation detection range cannot cover the coarse estimation result, the coarse estimation reconfiguration optimization iteration process is executed, specifically including: If the optimal precise estimate of the detection range is obtained by solving... This indicates that, given the current coarse estimation accuracy, lockout cannot be avoided. The system automatically triggers coarse estimation reconfiguration optimization, and iteratively updates the coarse estimation result, coarse estimation error variance, and fused residual frequency offset estimation result in a closed loop until the near-optimal fine estimation detection range covers the coarse estimation error variance. Finally, it outputs the optimal frequency offset estimation value adapted to the current power business.

[0013] According to another aspect of the present invention, the present invention provides a satellite communication frequency offset estimation system based on a power terminal system, the system comprising: The coarse estimation unit is used to preprocess the input parameters for power terminals without GNSS. Based on the preprocessed input parameters, it outputs the coarse estimation results and coarse estimation error variance for each power terminal by fusing the Gaussian prior of the terminal location and the beam coverage constraint to construct the Doppler-geographic joint likelihood function. The input parameters include: beam pointing constraint, preset GIS coordinates and the relative position of the transformer terminal. The precision estimation unit is used to extract the residual frequency offset of a single power terminal based on the frequency offset precision estimation method, and obtain the frequency offset estimation variance corresponding to the residual frequency offset; each power terminal in the power distribution area interacts with its own residual frequency offset and corresponding frequency offset estimation variance, and performs weighted fusion on the interacting residual frequency offset and frequency offset estimation variance to obtain the fused residual frequency offset estimation result. The optimization unit is used to divide the power business into multiple levels of business according to the priority of the power business and match the corresponding business unlocking penalty coefficient, construct a Bayesian cost function that integrates the unlocking probability and computational complexity, and solve for the optimal fine estimation detection range; when the optimal fine estimation detection range covers the variance of the coarse estimation error, the optimal frequency offset estimate is output based on the coarse estimation result and the fused residual frequency offset estimation result.

[0014] Preferably, the coarse estimation unit is used for input parameter preprocessing, including: beam pointing constraint preprocessing, power GIS prior modeling, and transformer terminal relative position modeling; The beam pointing constraint preprocessing is as follows: Based on the satellite's position, orbital altitude, and half-power beamwidth, the pointing angle of the satellite broadcast beam center is determined. This leads to the calibration of the beam center ground point and the satellite beam's ground coverage area, forming the beam coverage boundary constraint for power terminal distribution, expressed as: in, For the ground coverage area of ​​the beam, To calibrate the ground point of the beam center, For satellite position, Given the actual location of the power terminal, , For the orbital height, Half-power beamwidth; The prior modeling of the power GIS is as follows: A Gaussian prior distribution of the terminal location is established based on the pre-set GIS coordinates of the power terminal. The actual location of the power terminal follows a Gaussian distribution with the pre-set GIS coordinates as the mean and a pre-set covariance matrix as the constraint. The covariance matrix comprehensively considers GIS input error, power terminal installation deviation, and equivalent electrical length error of signal penetration through power equipment, and is expressed as: Set up power terminal The GIS preset coordinates are The actual location of GIS preset coordinates Follows a Gaussian prior distribution at the terminal position: Wherein, the covariance matrix , , , Typical values; The relative positions of the terminals in the distribution area are modeled as follows: Within the Taiwan area The geometric center of the power terminal area is: Any two power terminals The relative positions are known as follows: Power Terminal The actual position is represented by the terminal relative position constraint parameters: in, This is the nominal offset from the center of the distribution area to the power terminal. For individual random offsets.

[0015] Preferably, the coarse estimation unit is used to output the coarse estimation results and coarse estimation error variance of each power terminal based on the preprocessed input parameters by constructing a Doppler-geographic joint likelihood function that integrates the Gaussian prior of the terminal location and the beam coverage constraint, including: Based on satellite orbital speed With the center speed of the platform Calculate the theoretical Doppler frequency offset of the transformer center: in, At the speed of light, For carrier frequency; Constructing the Doppler-geographic joint likelihood function includes: Based on the given actual location of the power terminal The theoretical Doppler is determined as follows: in, The three-dimensional velocity vector of the power terminal; Considering measurement noise and beam pointing uncertainty, construct the actual Doppler observations of the received signal. Conditional probability density: in, Typical values; Combining the Gaussian prior distribution of the terminal location and the beam coverage boundary constraints, the posterior likelihood of a single power terminal is obtained: in, The three-dimensional geographic space area covered by the beam; based on the relative position constraint parameters of the transformer substation terminals, the posterior likelihood of a single power terminal is extended to the joint likelihood of multiple power terminals in the transformer substation: in, for, For the actual observed Doppler frequency shift of the i-th terminal, This is the common position deviation vector for all power terminals in the distribution area; Allowable frequency offset range Discretize into One grid point, Step size for each grid cell; Calculate the joint likelihood of multiple power terminals point by point, select the frequency offset parameter corresponding to the maximum joint likelihood value as the coarse estimation result of the power terminal, and output the corresponding coarse estimation error variance simultaneously: in, Let the assumed Doppler frequency offset be taken as the independent variable. For discrete grid search space, It is the joint posterior likelihood function of multiple power terminals.

[0016] Preferably, the frequency offset estimation method of the precision estimation unit includes single-terminal residual frequency offset calculation and multi-terminal collaborative fusion: The coarse estimation result is used to pre-compensate the frequency offset of the received signal. The autocorrelation characteristics of the OFDM symbol cyclic prefix and symbol tail are used to perform single-terminal fine frequency offset estimation on the small residual frequency offset after compensation. The residual frequency offset calculation formula is constructed based on the number of FFT points, cyclic prefix length and sampling period, and the frequency offset estimate of a single power terminal is obtained by solving it. According to the Cramer-Rao lower bound criterion, the single-terminal frequency offset estimation variance corresponding to the frequency offset estimate is calculated. Each power terminal within the power distribution area exchanges its own residual frequency offset estimate and corresponding frequency offset estimate variance; using the frequency offset estimate variance of each power terminal as the weighting weight, the frequency offset estimates of all power terminals within the power distribution area are weighted and fused to obtain the distribution area fused residual frequency offset estimate result, and the distribution area fused estimate variance corresponding to the fused residual frequency offset estimate result is solved.

[0017] Preferably, the optimization unit is used to classify services into multiple levels according to power service priority and match corresponding service unlocking penalty coefficients, including: Priority classification mapping is performed on the currently activated power services of the power terminal. Based on the difference in tolerance for frequency offset unlocking of different power services, the corresponding level of service unlocking penalty coefficient is matched, and the service unlocking penalty coefficient is uniquely determined by the highest priority service among all currently activated power services. Preferably, the optimization unit is used to construct a Bayesian cost function that integrates the unlocking probability and computational complexity, including: Constructing a Bayesian cost function that integrates unlock probability and computational complexity includes unlock probability modeling, computational complexity modeling, and Bayesian cost function construction: The unlocking probability model is as follows: A fine-grained frequency offset estimation effective detection range is defined. Based on the coarse estimation error variance corresponding to the coarse estimation result, the frequency offset fluctuation range of the coarse estimation result is characterized. When the frequency offset fluctuation range exceeds the fine-grained frequency offset estimation effective detection range, it is determined that the residual frequency offset fine-grained estimation lock-in of the power terminal has failed, and frequency offset unlocking has occurred. Based on the current coarse estimation error variance and the fine estimation detection range to be optimized, the real-time unlocking probability is calculated. The unlocking probability calculation model is expressed as follows: Coarse estimation error variance The effective detection range of the frequency offset fine estimation to be optimized is: ,like Calculate the real-time unlock probability: in, , Let the lower limit of integration be the threshold, and let the independent variable be... The variable to be integrated represents all possible values ​​that a standard normally distributed random variable can take. After integration, these values ​​will be eliminated. To roughly estimate the error variance, To accurately estimate the effective detection range for frequency offset; The computational complexity is modeled as follows: Based on the effective detection range of the frequency offset fine estimation to be optimized, the number of frequency points that the frequency offset fine estimation method needs to search is determined. Combining this with the computational complexity of the multiplication operations corresponding to a single frequency point, the total computational complexity of the fine estimation process is derived. The number of frequency points required for the frequency offset estimation method: in, This is the search step size; The computational cost for each frequency point is determined as follows: Total computational complexity: By introducing the business unlocking penalty coefficient corresponding to the business priority obtained from real-time monitoring, and fusing the unlocking probability and total computational complexity, a Bayesian cost function for the total cost is constructed, and the feasible region of the precise estimation detection range is limited: in, For the total cost, Penalty coefficient for business unlocking, This represents the probability of unlocking; The feasible region for precise estimation of the detection range is , in, The minimum detectable value can be set to 200Hz. The maximum detectable value can be set to 2000Hz; Based on the convex function property of the Bayesian cost function, the optimal precise estimation of the detection range is obtained by using the bisection method or a derivative approximation algorithm: .

[0018] Preferably, the optimization unit is further configured to execute a coarse estimation reconfiguration optimization iteration process when the optimal fine estimation detection range cannot cover the coarse estimation result, specifically including: If the optimal precise estimate of the detection range is obtained by solving... This indicates that, given the current coarse estimation accuracy, lockout cannot be avoided. The system automatically triggers coarse estimation reconfiguration optimization, and iteratively updates the coarse estimation result, coarse estimation error variance, and fused residual frequency offset estimation result in a closed loop until the near-optimal fine estimation detection range covers the coarse estimation error variance. Finally, it outputs the optimal frequency offset estimation value adapted to the current power business.

[0019] According to another aspect of the present invention, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a satellite communication frequency offset estimation method based on a power terminal system.

[0020] According to another aspect of the present invention, the present invention provides an electronic device, comprising: The aforementioned computer-readable storage medium; and One or more processors for executing a program in the computer-readable storage medium.

[0021] This invention provides a method and system for estimating satellite communication frequency offset based on a power terminal system. The method includes: preprocessing input parameters for power terminals without GNSS; and basing the preprocessed input parameters on the preprocessed input parameters, which include: beam pointing constraints and preset GIS. The system calculates the coordinates and relative positions of the terminals in the power distribution area. It constructs a Doppler-geographic joint likelihood function by fusing Gaussian priors of terminal positions with beam coverage constraints, outputting coarse estimation results and coarse estimation error variance for each power terminal. Based on the frequency offset fine estimation method, it extracts the residual frequency offset of a single power terminal and obtains the corresponding frequency offset estimation variance. Each power terminal within the power distribution area interacts with its own residual frequency offset and corresponding frequency offset estimation variance, and the interacting residual frequency offset and frequency offset estimation variance are weighted and fused to obtain a fused residual frequency offset estimation result. Based on the power service priority, it divides services into multiple levels and matches corresponding service unlocking penalty coefficients, constructing a Bayesian cost function that fuses unlocking probability and computational complexity, and solving for the optimal fine estimation detection range. When the optimal fine estimation detection range covers the coarse estimation error variance, it outputs the optimal frequency offset estimate based on the coarse estimation result and the fused residual frequency offset estimation result. This invention's technical solution effectively overcomes the limitations of existing NTN satellite communication frequency offset compensation relying on GNSS positioning by employing GNSS-free Doppler-geographic joint likelihood coarse estimation, multi-terminal weighted fusion residual frequency offset fine estimation, and an adaptive optimization mechanism that balances service priority, unlocking risk, and computational complexity. It is suitable for power terminal scenarios without GNSS support. Through multi-terminal collaborative fusion, it improves frequency offset estimation accuracy and anti-fluctuation capabilities, while adaptively optimizing the fine estimation detection range and dynamically balancing frequency offset unlocking reliability and computational complexity. This achieves high-precision, high-stability, and low-complexity satellite communication frequency offset estimation under different power service scenarios. Attached Figure Description

[0022] Exemplary embodiments of the present invention can be more fully understood by referring to the following figures: Figure 1 This is a flowchart of a satellite communication frequency offset estimation method based on a power terminal system according to an embodiment of the present invention; Figure 2 This is a flowchart of the coarse and fine joint estimation of frequency offset based on a power-based non-positioning terminal according to an embodiment of the present invention; Figure 3 A flowchart illustrating the coarse estimation process based on a power terminal system according to an embodiment of the present invention; and Figure 4 This is a flowchart of the joint frequency offset estimation according to an embodiment of the present invention; Figure 5 This is a structural diagram of a satellite communication frequency offset estimation system based on a power terminal system according to an embodiment of the present invention. Detailed Implementation

[0023] Exemplary embodiments of the invention will now be described with reference to the accompanying drawings. However, the invention may be embodied in many different forms and is not limited to the embodiments described herein. These embodiments are provided to fully and completely disclose the invention and to fully convey its scope to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the drawings is not intended to limit the invention. In the drawings, the same units / elements are referred to by the same reference numerals.

[0024] Unless otherwise stated, the terms used herein (including technical terms) have their common meaning as understood by one of ordinary skill in the art. Furthermore, it is understood that terms defined in commonly used dictionaries should be understood to have a meaning consistent with the context of their relevant field, and not to be interpreted as having an idealized or overly formal meaning.

[0025] Figure 1 This is a flowchart of a satellite communication frequency offset estimation method based on a power terminal system according to an embodiment of the present invention.

[0026] The power industry is a key vertical application area for NTN technology. In large-scale emergency repair scenarios such as circuit outages, power outages, and network disruptions, the need for multi-point communication at the front-line command center, remote video monitoring of front-line emergency response, and disaster recovery of plant automation operations is becoming increasingly urgent. With its advantages of wide coverage and strong resilience, NTN technology can play a crucial role in business scenarios such as electricity information collection in remote areas, emergency power grid communication, and transmission line inspection.

[0027] However, the power industry faces unique constraints in deploying NTN terminals: (1) No GNSS environment: A large number of power terminals such as distribution automation terminals (DTU), feeder terminals (FTU), and transformer terminals (TTU) are installed in underground power distribution rooms, inside the metal structure of substations, and in mountainous towers and other places where GNSS signals are completely rejected or severely attenuated. (2) High-precision synchronization requirements: Differential protection and other services in distribution networks have extremely high requirements for time synchronization accuracy. Frequency deviation estimation error is directly converted into synchronization error, which may lead to protection maloperation or failure to operate. (3) Coexistence of mixed services: Control, acquisition and video services coexist in the same area, and the requirements of different services for frequency offset estimation accuracy and latency are significantly different; (4) Terminal power consumption and cost constraints: A large number of acquisition terminals are powered by batteries or energy acquisition power, requiring the frequency offset estimation algorithm to have low computational complexity.

[0028] Therefore, how to achieve high-precision, low-complexity frequency offset estimation that meets the needs of mixed power services under GNSS-free conditions has become a core technical problem that power NTN systems urgently need to solve. This invention patent proposes a coarse-fine joint frequency offset estimation scheme that utilizes power GIS asset data, transformer area topology relationships, and local communication networks.

[0029] This invention addresses the core problem of time-frequency offset estimation failure in power terminals lacking GNSS positioning capabilities. It proposes a three-stage fusion scheme of "coarse estimation → fine estimation → joint optimization," which fully utilizes prior information unique to the power industry, such as GIS asset coordinates, transformer area topology, PLC local communication network, and service priorities.

[0030] like Figure 1 As shown, this invention provides a satellite communication frequency offset estimation method based on a power terminal system, the method comprising: Step 101: Perform input parameter preprocessing for power terminals without GNSS; input parameters include: beam pointing constraints, preset GIS coordinates, and relative positions of the transformer area terminals; Preferably, the input parameter preprocessing includes: beam pointing constraint preprocessing, power GIS prior modeling, and transformer terminal relative position modeling; Beam pointing constraint preprocessing is as follows: Based on the satellite's position, orbital altitude, and half-power beamwidth, the pointing angle of the satellite broadcast beam center is determined. This leads to the calibration of the beam center ground point and the satellite beam's ground coverage area, forming the beam coverage boundary constraint for power terminal distribution, expressed as: in, For the ground coverage area of ​​the beam, To calibrate the ground point of the beam center, For satellite position, Given the actual location of the power terminal, , For the orbital height, Half-power beamwidth; The prior modeling of power GIS is as follows: A Gaussian prior distribution of the terminal location is established based on the pre-set GIS coordinates of the power terminal. The actual location of the power terminal follows a Gaussian distribution with the pre-set GIS coordinates as the mean and a pre-set covariance matrix as the constraint. The covariance matrix comprehensively considers GIS input error, power terminal installation deviation, and the equivalent electrical length error of signal penetration through power equipment, and is expressed as: Set up power terminal The GIS preset coordinates are The actual location of GIS preset coordinates Follows a Gaussian prior distribution at the terminal position: Wherein, the covariance matrix , , , Typical values; The relative positions of the terminals in the distribution area are modeled as follows: Within the Taiwan area The geometric center of the power terminal area is: Any two power terminals The relative positions are known as follows: Power Terminal The actual position is represented by the terminal relative position constraint parameters: in, This is the nominal offset from the center of the distribution area to the power terminal. For individual random offsets.

[0031] Step 102: Based on the preprocessed input parameters, the coarse estimation results and coarse estimation error variance of each power terminal are output by constructing the Doppler-geographic joint likelihood function by fusing the Gaussian prior of the terminal location and the beam coverage constraint. Preferably, based on the preprocessed input parameters, a Doppler-geographic joint likelihood function is constructed by fusing the Gaussian prior of the terminal location and the beam coverage constraint to output the coarse estimation results and coarse estimation error variance for each power terminal, including: Based on satellite orbital speed With the center speed of the platform Calculate the theoretical Doppler frequency offset of the transformer center: in, At the speed of light, For carrier frequency; Constructing the Doppler-geographic joint likelihood function includes: Based on the given actual location of the power terminal The theoretical Doppler is determined as follows: in, The three-dimensional velocity vector of the power terminal; Considering measurement noise and beam pointing uncertainty, construct the actual Doppler observations of the received signal. Conditional probability density: in, Typical values; By combining the Gaussian prior distribution of the combined terminal location and the beam coverage boundary constraints, the posterior likelihood of a single power terminal is obtained: in, The three-dimensional geographic space area covered by the beam; based on the relative position constraint parameters of the transformer substation terminals, the posterior likelihood of a single power terminal is extended to the joint likelihood of multiple power terminals in the transformer substation: in, for, For the actual observed Doppler frequency shift of the i-th terminal, This is the common position deviation vector for all power terminals in the distribution area; Allowable frequency offset range Discretize into One grid point, Step size for each grid cell; Calculate the joint likelihood of multiple power terminals point by point, select the frequency offset parameter corresponding to the maximum joint likelihood value as the coarse estimation result of the power terminal, and output the corresponding coarse estimation error variance simultaneously: in, Let the assumed Doppler frequency offset be taken as the independent variable. For discrete grid search space, It is the joint posterior likelihood function of multiple power terminals.

[0032] In the coarse estimation stage, this invention utilizes the terminal preset coordinates and the center location of the transformer area in the power GIS database, combined with the beam pointing angle and ephemeris of satellite broadcasting, to calculate the theoretical Doppler of the transformer area center; then, by fusing the location prior and beam coverage constraints through the Doppler-geographic joint likelihood function, and through grid search, the frequency offset uncertainty range is compressed from ±2000Hz to within ±400Hz, thus determining the coarse estimate and its error variance.

[0033] Step 103: Extract the residual frequency offset of a single power terminal based on the frequency offset fine estimation method, and obtain the frequency offset estimation variance corresponding to the residual frequency offset; Step 104: Each power terminal in the power distribution area exchanges its own residual frequency offset and corresponding frequency offset estimation variance, and performs weighted fusion of the exchanged residual frequency offset and frequency offset estimation variance to obtain the fused residual frequency offset estimation result. Preferably, the frequency offset estimation method includes single-terminal residual frequency offset calculation and multi-terminal collaborative fusion: The coarse estimation results are used to pre-compensate the frequency offset of the received signal. The autocorrelation characteristics of the OFDM symbol cyclic prefix and symbol tail are used to perform fine frequency offset estimation for the small residual frequency offset after compensation on a single terminal. The residual frequency offset calculation formula is constructed based on the number of FFT points, the cyclic prefix length and the sampling period, and the frequency offset estimate of a single power terminal is obtained by solving it. According to the Cramer-Rao lower bound criterion, the variance of the frequency offset estimate of a single terminal is calculated. Within a power distribution area, each power terminal exchanges its residual frequency offset estimate and corresponding frequency offset estimate variance through the local communication network of the power distribution area. Using the frequency offset estimate variance of each power terminal as a weighting weight, the frequency offset estimates of all power terminals within the power distribution area are weighted and fused to obtain the fused residual frequency offset estimate result of the distribution area, and the fused estimate variance of the distribution area corresponding to the fused residual frequency offset estimate result is solved.

[0034] In the fine estimation stage of this invention, after coarse estimation compensation, each terminal uses OFDM cyclic prefix autocorrelation to estimate the residual frequency offset; multiple terminals (DTU / FTU / TTU) in the same area exchange their respective estimates and signal-to-noise ratios through existing PLCs or low-power wireless networks, and the cluster head performs weighted fusion to reduce the estimation variance by a factor of √K (K is the number of cooperating terminals) to obtain high-precision residual frequency offset.

[0035] Step 105: Divide the business into multiple levels according to the priority of power business and match the corresponding business unlocking penalty coefficient. Construct a Bayesian cost function that integrates unlocking probability and computational complexity, and solve it to obtain the optimal precise estimation detection range. Step 106: When the optimal fine estimation detection range covers the coarse estimation error variance, output the optimal frequency offset estimate based on the coarse estimation result and the fused residual frequency offset estimation result.

[0036] Preferably, the business is divided into multiple levels according to the priority of the power business and a corresponding business de-locking penalty coefficient is matched, including: Priority classification mapping is performed on the currently activated power services of the power terminal. Based on the difference in tolerance for frequency offset unlocking of different power services, the corresponding level of service unlocking penalty coefficient is matched, and the service unlocking penalty coefficient is uniquely determined by the highest priority service among all currently activated power services. Preferably, a Bayesian cost function that integrates the unlocking probability and computational complexity is constructed, including: Constructing a Bayesian cost function that integrates unlock probability and computational complexity includes unlock probability modeling, computational complexity modeling, and Bayesian cost function construction: The unlocking probability model is as follows: The effective detection range of the fine frequency offset estimation is defined. Based on the variance of the coarse estimation error corresponding to the coarse estimation result, the frequency offset fluctuation range of the coarse estimation result is characterized. When the frequency offset fluctuation range exceeds the effective detection range of the fine frequency offset estimation, the residual frequency offset fine estimation locking of the power terminal is determined to have failed, resulting in frequency offset unlocking. Based on the current variance of the coarse estimation error and the fine estimation detection range to be optimized, the real-time unlocking probability is solved. The unlocking probability calculation model is expressed as follows: Coarse estimation error variance The effective detection range of the frequency offset fine estimation to be optimized is: ,like Calculate the real-time unlock probability: in, , Let the lower limit of integration be the threshold, and let the independent variable be... The variable to be integrated represents all possible values ​​that a standard normally distributed random variable can take. After integration, these values ​​will be eliminated. To roughly estimate the error variance, To accurately estimate the effective detection range for frequency offset; The computational complexity is modeled as follows: Based on the effective detection range of the frequency offset fine estimation to be optimized, the number of frequency points that the frequency offset fine estimation method needs to search is determined. Combining this with the computational complexity of the multiplication operations corresponding to a single frequency point, the total computational complexity of the fine estimation process is derived. The number of frequency points required for the frequency offset estimation method: in, This is the search step size; The computational cost for each frequency point is determined as follows: Total computational complexity: By introducing the business unlocking penalty coefficient corresponding to the business priority obtained from real-time monitoring, and fusing the unlocking probability and total computational complexity, a Bayesian cost function for the total cost is constructed, and the feasible region of the precise estimation detection range is limited: in, For the total cost, Penalty coefficient for business unlocking, This represents the probability of unlocking; The feasible region for precise estimation of the detection range is , in, The minimum detectable value can be set to 200Hz. The maximum detectable value can be set to 2000Hz; Based on the convex function property of the Bayesian cost function, the optimal precise estimate of the detection range is obtained by using the bisection method or a derivative approximation algorithm. .

[0037] Preferably, when the optimal fine estimation detection range cannot cover the coarse estimation error variance, a coarse estimation reconfiguration optimization iterative process is executed, specifically including: If the optimal precise estimate of the detection range is obtained by solving... This indicates that, given the current coarse estimation accuracy, lockout cannot be avoided. The system automatically triggers coarse estimation reconfiguration optimization, and iteratively updates the coarse estimation results, coarse estimation error variance, and fused residual frequency offset estimation results in a closed loop until the near-optimal fine estimation detection range covers the coarse estimation error variance. Finally, it outputs the optimal frequency offset estimate that is adapted to the current power business.

[0038] In the above steps, this invention performs joint optimization and mode switching. This invention monitors the coarse estimation error variance and the current service priority in real time, and maps differential protection (P0), distribution automation (P1), acquisition (P2), and video (P3) to different unlocking penalty coefficients; establishes a Bayesian cost function (computational complexity + penalty coefficient × unlocking probability) and dynamically selects the optimal fine estimation detection range; if the optimal range is insufficient to cover the coarse estimation error, it automatically triggers coarse estimation reconfiguration (encryption grid / extended integration) to form a closed-loop adaptive mechanism; the P0 service is forced to use the most conservative parameters to ensure extreme reliability.

[0039] This invention utilizes prior information unique to the power industry to construct a three-stage technical architecture based on power: "coarse estimation → fine estimation → joint optimization," addressing the core issue of frequency offset estimation failure in power terminals under GNSS-free scenarios. This invention is the first to introduce power GIS and transformer topology into the NTN frequency offset estimation framework, proposing a coarse estimation method that integrates "beam pointing + GIS priors + transformer topology." This invention also pioneers a fine estimation enhancement mechanism based on weighted fusion of multi-terminal collaboration within the transformer area and establishes a Bayesian joint optimization model driven by business priorities.

[0040] The following is in conjunction with the appendix Figure 2 The embodiments of the present invention will be illustrated by examples.

[0041] This invention includes: input preprocessing a) Beam pointing constraint Beam center pointing angle of satellite broadcasting The ground point of the beam center was determined. Beam coverage area: in For satellite position, , For the orbital height, This is the half-power beamwidth.

[0042] b) Prior modeling of power GIS Set terminal The GIS preset coordinates are Its true location Follows a Gaussian distribution: Where the covariance matrix Considering GIS data entry errors, installation deviations, and the equivalent electrical length of signals penetrating power equipment, typical values ​​are... , .

[0043] c) Center and relative location of the transformer area Within the distribution transformer area (power supply area) there are One terminal, geometric center of the distribution area: Any two terminals The relative positions are known: Therefore, the terminal The actual location can be represented as: in This is the nominal offset from the center of the distribution area to the terminal. Individual random bias (variance much smaller) ).

[0044] (2) Coarse estimation processing a) Doppler theory of the center of the distribution area Satellite speed speed at the center of the platform (Usually very small, acceptable) ),but: b) Doppler-geographic joint likelihood function Given the actual location of the terminal The theoretical Doppler is: Considering measurement noise and beam pointing uncertainty, the actual Doppler observation of the received signal f_d^{obs} Conditional probability density: in (Typical value).

[0045] By combining the prior knowledge of terminal location and beam coverage constraints, the posterior likelihood is obtained: For multiple terminals in the same distribution area, the above formula can be extended to a joint likelihood using relative position constraints: c) Grid search implementation Possible frequency offset range Discretize into One grid point, For each grid point, a likelihood value is calculated, and the largest value is taken as a coarse estimate. Coarse estimation error variance It can be approximated by the curvature of the likelihood function at its peak. For example... Figure 3 As shown.

[0046] 2. Precise estimation stage (1) Fine estimation of CP autocorrelation in a single terminal Receive time domain signal After coarse estimation and compensation, the residual frequency offset Very small. Utilizing the correlation between the cyclic prefix (CP) of the OFDM symbol and the tail copy: in For FFT points, This is the CP length. (Ignoring noise) Therefore, the residual frequency offset is estimated as follows: The sampling period is given. The Cramer-Rao lower bound for estimating the variance is: (2) Multi-terminal collaboration and weighted integration in the distribution area within the same area Each terminal exchanges its estimated value via a local communication network (PLC or low-power wireless). and estimated variance Weighted fusion: Variance after fusion: When the signal-to-noise ratios of each terminal are similar Improved estimation accuracy times.

[0047] 3. Joint frequency offset estimation stage (1) Real-time monitoring module Electricity service priority mapping: Different electricity services have significantly different tolerances for unlocking, and a penalty coefficient is defined. :

[0048] Select based on the highest priority among currently activated services.

[0049] (2) Risk Calculation Module a) Unlocking probability model coarse estimation error The effective detection range of the precise estimation module is .like If the exact estimation fails, the lock cannot be locked (unlocked). Unlock probability: in .

[0050] b) Complexity Model The number of frequency points that the precise estimation module needs to search is: The computational cost per frequency point is (Mainly related calculations, approximately) (Number of multiplications). Total complexity: c) Priority penalty coefficient Taken from the real-time monitoring module - power priority mapping penalty coefficient .

[0051] d) Bayesian cost function and optimal Define the total cost: The feasible domain is ,generally , .

[0052] (3) Optimize the decision-making module because It is a convex function ( Linear growth, (Exponential decay), the optimal solution can be obtained through bisection or derivative approximation: like This indicates that under the current coarse estimation accuracy, even using the optimal fine estimation range cannot avoid lockout (because...). At least should be (2-3 times the original value). At this point, a rough estimate reconfiguration is triggered, and specific measures include: - Step size for encrypted grid search (reduced from 100Hz to 50Hz) - Extend the integral length of CP autocorrelation (using multiple OFDM symbols) - Increase the number of terminals that can cooperate in the area (if any terminals were previously offline). After reconfiguration, a rough estimate of the error variance is performed. It is expected to reduce the cost by more than 50%, allowing it to re-enter the optimization cycle. For example... Figure 4 As shown.

[0053] (4) Special settings for power scenarios When the system detects that the current service includes P0 (differential protection), it forces the most conservative parameters to be used: Forced (even though The calculated optimal value is smaller. Force all online terminals to participate in collaborative integration ( Take the maximum value) Disable the timeout limit for coarse-estimated reconfiguration (allowing for longer reconfiguration times to ensure reliability). First, this invention achieves high-precision frequency offset tracking in GNSS-free scenarios. The coarse estimation, by fusing power GIS coordinates and beam pointing information, compresses the frequency offset uncertainty range from ±2000 Hz in traditional methods to within ±400 Hz, eliminating reliance on GNSS positioning. The fine estimation, leveraging CP autocorrelation and weighted fusion of multiple terminals in the distribution area, further reduces the residual frequency offset to below 10 Hz, directly meeting the stringent requirements of high-precision services such as differential protection for residual frequency offset <5~10 Hz. The overall estimation accuracy is 5~10 times higher than existing GNSS-free solutions such as pure beam pointing, enabling reliable uplink synchronization for power terminals in restricted environments such as basements and distribution cabinets.

[0054] Secondly, this invention significantly enhances robustness to high-dynamic and extreme LEO scenarios. The center-referenced distribution area and small-range grid search improve Doppler rate-of-change tracking capability from 2 kHz / s to 10 kHz / s, effectively addressing the challenge of drastic changes in satellite overflight frequency offset. Bayesian joint optimization and reconfiguration mechanisms monitor coarse estimation errors in real time, automatically refining the grid or extending integration when the detection range is insufficient, avoiding unlocking due to orbital perturbations or sudden obstructions. For P0 differential protection services, the most conservative configuration is enforced to provide deterministic protection for critical power services.

[0055] Finally, this invention achieves differentiated and precise adaptation for hybrid power services. By mapping service priorities (P0~P3) to different unlocking penalty coefficients, the system dynamically trades off between computational complexity and estimation reliability: P0 (differential protection) pursues extreme reliability, allowing for higher computational overhead; P2 / P3 (acquisition / video) moderately relaxes accuracy, prioritizing the reduction of terminal power consumption. This "service-aware" estimation strategy enables the same frequency offset estimation framework to simultaneously serve control, acquisition, and video services, avoiding resource waste or performance deficiencies caused by a "one-size-fits-all" approach.

[0056] The research results of this invention can be directly embedded into the physical layer synchronization module of power NTN terminals, providing an industry-level reference example for frequency offset estimation schemes without GNSS terminals in subsequent 3GPP standards. The mode switching and reconfiguration mechanism enables the system to automatically adjust parameters according to real-time channel quality and service requirements without manual intervention, adapting to the unattended operation and maintenance mode of power terminals. All algorithm modules are implemented based on downlink signal beam pointing, CP, and GIS preset coordinates, without relying on uplink interaction or network-side modifications, possessing good forward compatibility.

[0057] Figure 5 This is a structural diagram of a satellite communication frequency offset estimation system based on a power terminal system according to an embodiment of the present invention.

[0058] like Figure 5 As shown, this invention provides a satellite communication frequency offset estimation system based on a power terminal system. The system includes: The coarse estimation unit 501 is used to preprocess the input parameters for power terminals without GNSS. Based on the preprocessed input parameters, it outputs the coarse estimation results and coarse estimation error variance of each power terminal by fusing the Gaussian prior of the terminal location and the beam coverage constraint to construct the Doppler-geographic joint likelihood function. The input parameters include: beam pointing constraint, preset GIS coordinates and the relative position of the transformer area terminal. Preferably, the coarse estimation unit 501 is used for input parameter preprocessing, including: beam pointing constraint preprocessing, power GIS prior modeling, and transformer terminal relative position modeling; Beam pointing constraint preprocessing is as follows: Based on the satellite's position, orbital altitude, and half-power beamwidth, the pointing angle of the satellite broadcast beam center is determined. This leads to the calibration of the beam center ground point and the satellite beam's ground coverage area, forming the beam coverage boundary constraint for power terminal distribution, expressed as: in, For the ground coverage area of ​​the beam, To calibrate the ground point of the beam center, For satellite position, Given the actual location of the power terminal, , For the orbital height, Half-power beamwidth; The prior modeling of power GIS is as follows: A Gaussian prior distribution of the terminal location is established based on the pre-set GIS coordinates of the power terminal. The actual location of the power terminal follows a Gaussian distribution with the pre-set GIS coordinates as the mean and a pre-set covariance matrix as the constraint. The covariance matrix comprehensively considers GIS input error, power terminal installation deviation, and the equivalent electrical length error of signal penetration through power equipment, and is expressed as: Set up power terminal The GIS preset coordinates are The actual location of GIS preset coordinates Follows a Gaussian prior distribution at the terminal position: Wherein, the covariance matrix , , , Typical values; The relative positions of the terminals in the distribution area are modeled as follows: Within the Taiwan area The geometric center of the power terminal area is: Any two power terminals The relative positions are known as follows: Power Terminal The actual position is represented by the terminal relative position constraint parameters: in, This is the nominal offset from the center of the distribution area to the power terminal. For individual random offsets.

[0059] Preferably, the coarse estimation unit 501 is used to output the coarse estimation results and coarse estimation error variance of each power terminal based on the preprocessed input parameters by fusing the Doppler-geographic joint likelihood function constructed by the Gaussian prior of the terminal location and the beam coverage constraint, including: Based on satellite orbital speed With the center speed of the platform Calculate the theoretical Doppler frequency offset of the transformer center: in, At the speed of light, For carrier frequency; Constructing the Doppler-geographic joint likelihood function includes: Based on the given actual location of the power terminal The theoretical Doppler is determined as follows: in, The three-dimensional velocity vector of the power terminal is given; considering measurement noise and beam pointing uncertainty, the actual Doppler observations of the received signal are constructed. Conditional probability density: in, Typical values; By combining the Gaussian prior distribution of the combined terminal location and the beam coverage boundary constraints, the posterior likelihood of a single power terminal is obtained: in, For the three-dimensional geographic space area covered by the beam; Based on the relative position constraints of the terminals in the distribution area, the posterior likelihood of a single power terminal is extended to the joint likelihood of multiple power terminals in the distribution area: in, for, For the actual observed Doppler frequency shift of the i-th terminal, This is the common position deviation vector for all power terminals in the distribution area; Allowable frequency offset range Discretize into One grid point, Step size for each grid cell; Calculate the joint likelihood of multiple power terminals point by point, select the frequency offset parameter corresponding to the maximum joint likelihood value as the coarse estimation result of the power terminal, and output the corresponding coarse estimation error variance simultaneously: in, Let the assumed Doppler frequency offset be taken as the independent variable. For discrete grid search space, It is the joint posterior likelihood function of multiple power terminals.

[0060] The precision estimation unit 502 is used to extract the residual frequency offset of a single power terminal based on the frequency offset precision estimation method and obtain the frequency offset estimation variance corresponding to the residual frequency offset; each power terminal in the power distribution area interacts with its own residual frequency offset and corresponding frequency offset estimation variance, and performs weighted fusion on the interacting residual frequency offset and frequency offset estimation variance to obtain the fused residual frequency offset estimation result. Preferably, the frequency offset estimation method of the precision estimation unit 502 includes single-terminal residual frequency offset calculation and multi-terminal collaborative fusion: The coarse estimation results are used to pre-compensate the frequency offset of the received signal. The autocorrelation characteristics of the OFDM symbol cyclic prefix and symbol tail are used to perform fine frequency offset estimation for the small residual frequency offset after compensation on a single terminal. The residual frequency offset calculation formula is constructed based on the number of FFT points, the cyclic prefix length and the sampling period, and the frequency offset estimate of a single power terminal is obtained by solving it. According to the Cramer-Rao lower bound criterion, the variance of the frequency offset estimate of a single terminal is calculated. Within a power distribution area, each power terminal exchanges its residual frequency offset estimate and corresponding frequency offset estimate variance through the local communication network of the power distribution area. Using the frequency offset estimate variance of each power terminal as a weighting weight, the frequency offset estimates of all power terminals within the power distribution area are weighted and fused to obtain the fused residual frequency offset estimate result of the distribution area, and the fused estimate variance of the distribution area corresponding to the fused residual frequency offset estimate result is solved.

[0061] The optimization unit 503 is used to divide the power business into multiple levels of business according to the priority of the power business and match the corresponding business unlocking penalty coefficient, construct a Bayesian cost function that integrates the unlocking probability and computational complexity, and solve for the optimal fine estimation detection range; when the optimal fine estimation detection range covers the coarse estimation error variance, the optimal frequency offset estimate is output based on the coarse estimation result and the fused residual frequency offset estimation result.

[0062] Preferably, the optimization unit 503 is used to divide services into multiple levels according to power service priority and match the corresponding service unlocking penalty coefficient, including: Priority classification mapping is performed on the currently activated power services of the power terminal. Based on the difference in tolerance for frequency offset unlocking of different power services, the corresponding level of service unlocking penalty coefficient is matched, and the service unlocking penalty coefficient is uniquely determined by the highest priority service among all currently activated power services. Preferably, the optimization unit 503 is used to construct a Bayesian cost function that integrates the unlocking probability and computational complexity, including: Constructing a Bayesian cost function that integrates unlock probability and computational complexity includes unlock probability modeling, computational complexity modeling, and Bayesian cost function construction: The unlocking probability model is as follows: The effective detection range of the fine frequency offset estimation is defined. Based on the variance of the coarse estimation error corresponding to the coarse estimation result, the frequency offset fluctuation range of the coarse estimation result is characterized. When the frequency offset fluctuation range exceeds the effective detection range of the fine frequency offset estimation, the residual frequency offset fine estimation locking of the power terminal is determined to have failed, resulting in frequency offset unlocking. Based on the current variance of the coarse estimation error and the fine estimation detection range to be optimized, the real-time unlocking probability is solved. The unlocking probability calculation model is expressed as follows: Coarse estimation error variance The effective detection range of the frequency offset fine estimation to be optimized is: ,like Calculate the real-time unlock probability: in, , Let the lower limit of integration be the threshold, and let the independent variable be... The variable to be integrated represents all possible values ​​that a standard normally distributed random variable can take. After integration, these values ​​will be eliminated. To roughly estimate the error variance, To accurately estimate the effective detection range for frequency offset; The computational complexity is modeled as follows: Based on the effective detection range of the frequency offset fine estimation to be optimized, the number of frequency points that the frequency offset fine estimation method needs to search is determined. Combining this with the computational complexity of the multiplication operations corresponding to a single frequency point, the total computational complexity of the fine estimation process is derived. The number of frequency points required for the frequency offset estimation method: in, This is the search step size; The computational cost for each frequency point is determined as follows: Total computational complexity: By introducing the business unlocking penalty coefficient corresponding to the business priority obtained from real-time monitoring, and fusing the unlocking probability and total computational complexity, a Bayesian cost function for the total cost is constructed, and the feasible region of the precise estimation detection range is limited: in, For the total cost, Penalty coefficient for business unlocking This represents the probability of unlocking; The feasible region for accurately estimating the detection range is , in, The minimum detectable value can be set to 200Hz. The maximum detectable value can be set to 2000Hz; Based on the convex function property of the Bayesian cost function, the optimal precise estimate of the detection range is obtained by using the bisection method or a derivative approximation algorithm. .

[0063] Preferably, the optimization unit 503 is further configured to execute a coarse estimation reconfiguration optimization iteration process when the optimal fine estimation detection range cannot cover the coarse estimation result, specifically including: If the optimal precise estimate of the detection range is obtained by solving... This indicates that, given the current coarse estimation accuracy, lockout cannot be avoided. The system automatically triggers coarse estimation reconfiguration optimization, and iteratively updates the coarse estimation results, coarse estimation error variance, and fused residual frequency offset estimation results in a closed loop until the near-optimal fine estimation detection range covers the coarse estimation error variance. Finally, it outputs the optimal frequency offset estimate that is adapted to the current power business.

[0064] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the satellite communication frequency offset estimation method based on a power terminal system as described above.

[0065] According to another aspect of the present invention, the present invention provides an electronic device, comprising: The aforementioned computer-readable storage medium; and One or more processors for executing a program in a computer-readable storage medium.

[0066] The present invention has been described with reference to a few embodiments. However, it will be apparent to those skilled in the art that other embodiments besides those disclosed above fall equivalently within the scope of the present invention.

[0067] Generally, all terms used in this invention are interpreted according to their ordinary meaning in the art, unless otherwise expressly defined herein. All references to “a / the / the [device, component, etc.]” ​​are openly interpreted as at least one instance of said device, component, etc., unless otherwise expressly stated. The steps of any method disclosed herein need not be performed in the exact order disclosed unless explicitly stated otherwise.

[0068] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0069] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0070] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0071] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for estimating satellite communication frequency offset based on a power terminal system, characterized in that, The method includes: Input parameter preprocessing is performed for power terminals without GNSS; the input parameters include: beam pointing constraints, preset GIS coordinates, and relative positions of the substation terminals; Based on the preprocessed input parameters, the Doppler-geographic joint likelihood function is constructed by fusing the Gaussian prior of the terminal location and the beam coverage constraint, and the coarse estimation results and coarse estimation error variance of each power terminal are output. The residual frequency offset of a single power terminal is extracted based on the frequency offset fine estimation method, and the frequency offset estimation variance corresponding to the residual frequency offset is obtained; Each power terminal within the power distribution area exchanges its own residual frequency offset and corresponding frequency offset estimation variance, and then performs weighted fusion of the exchanged residual frequency offset and frequency offset estimation variance to obtain the fused residual frequency offset estimation result. Based on the priority of power business, we divide the business into multiple levels and match the corresponding business unlocking penalty coefficient. We construct a Bayesian cost function that integrates unlocking probability and computational complexity, and solve it to obtain the optimal precise estimation detection range. When the optimal fine estimation detection range covers the variance of the coarse estimation error, the optimal frequency offset estimate is output based on the coarse estimation result and the fused residual frequency offset estimation result.

2. The method according to claim 1, characterized in that, The input parameter preprocessing includes: beam pointing constraint preprocessing, power GIS prior modeling, and transformer terminal relative position modeling. The beam pointing constraint preprocessing is as follows: Based on the satellite's position, orbital altitude, and half-power beamwidth, the pointing angle of the satellite broadcast beam center is determined. This leads to the calibration of the beam center ground point and the satellite beam's ground coverage area, forming the beam coverage boundary constraint for power terminal distribution, expressed as: in, For the ground coverage area of ​​the beam, To calibrate the ground point of the beam center, For satellite position, Given the actual location of the power terminal, , For the orbital height, Half-power beamwidth; The power GIS prior modeling involves establishing a Gaussian prior distribution of the terminal location based on the pre-set GIS coordinates of the power terminal. The actual location of the power terminal follows a Gaussian distribution with the pre-set GIS coordinates as the mean and a pre-set covariance matrix as the constraint. The covariance matrix comprehensively considers GIS input errors, power terminal installation deviations, and the equivalent electrical length error of signal penetration through power equipment, and is expressed as follows: Set up power terminal The GIS preset coordinates are The actual location of GIS preset coordinates Follows a Gaussian prior distribution at the terminal position: Wherein, the covariance matrix , , , Typical values; The relative positions of the terminals in the distribution area are modeled as follows: Within the Taiwan area The geometric center of the power terminal area is: Any two power terminals The relative positions are known as follows: Power Terminal The actual position is represented by the terminal relative position constraint parameters: in, This refers to the nominal offset from the center of the distribution area to the power terminal. For individual random offsets.

3. The method according to claim 2, characterized in that, Based on the preprocessed input parameters, the Doppler-geographic joint likelihood function, constructed by fusing the Gaussian prior of the terminal location and the beam coverage constraint, outputs the coarse estimation results and coarse estimation error variance for each power terminal, including: Based on satellite orbital speed With the center speed of the platform Calculate the theoretical Doppler frequency offset of the transformer center: in, At the speed of light, For carrier frequency; Constructing the Doppler-geographic joint likelihood function includes: Based on the given actual location of the power terminal The theoretical Doppler is determined as follows: in, The three-dimensional velocity vector of the power terminal; Considering measurement noise and beam pointing uncertainty, construct the actual Doppler observations of the received signal. Conditional probability density: in, Typical values; Combining the Gaussian prior distribution of the terminal location and the beam coverage boundary constraints, the posterior likelihood of a single power terminal is obtained: in, For the three-dimensional geographic space area covered by the beam; Based on the relative position constraints of the terminals in the distribution area, the posterior likelihood of a single power terminal is extended to the joint likelihood of multiple power terminals in the distribution area: in, for, For the actual observed Doppler frequency shift of the i-th power terminal, This is the common position deviation vector for all power terminals in the distribution area; Allowable frequency offset range Discretize into One grid point, Step size for each grid cell; Calculate the joint likelihood of multiple power terminals point by point, select the frequency offset parameter corresponding to the maximum joint likelihood value as the coarse estimation result of the power terminal, and output the corresponding coarse estimation error variance simultaneously: in, Let the assumed Doppler frequency offset be taken as the independent variable. For discrete grid search space, It is the joint posterior likelihood function of multiple power terminals.

4. The method according to claim 1, characterized in that, The frequency offset estimation method includes single-terminal residual frequency offset calculation and multi-terminal collaborative fusion: The coarse estimation result is used to pre-compensate the frequency offset of the received signal. The autocorrelation characteristics of the OFDM symbol cyclic prefix and symbol tail are used to perform single-terminal fine frequency offset estimation on the small residual frequency offset after compensation. The residual frequency offset calculation formula is constructed based on the number of FFT points, cyclic prefix length and sampling period, and the frequency offset estimate of a single power terminal is obtained by solving it. According to the Cramer-Rao lower bound criterion, the single-terminal frequency offset estimation variance corresponding to the frequency offset estimate is calculated. The residual frequency offset estimates and corresponding frequency offset estimates of each power terminal in the power distribution area are exchanged. Using the frequency offset estimation variance of each power terminal as the weighting weight, the frequency offset estimates of all power terminals in the power distribution area are weighted and fused to obtain the distribution area fusion residual frequency offset estimation result, and the distribution area fusion estimation variance corresponding to the fusion residual frequency offset estimation result is solved.

5. The method according to claim 1, characterized in that, The process of classifying services into multiple levels based on power service priority and matching corresponding service unlocking penalty coefficients includes: Priority classification mapping is performed on the currently activated power services of the power terminal. Based on the difference in tolerance for frequency offset unlocking of different power services, the corresponding level of service unlocking penalty coefficient is matched, and the service unlocking penalty coefficient is uniquely determined by the highest priority service among all currently activated power services.

6. The method according to claim 1, characterized in that, The construction of the Bayesian cost function that integrates the unlocking probability and computational complexity includes: Constructing a Bayesian cost function that integrates unlock probability and computational complexity includes unlock probability modeling, computational complexity modeling, and Bayesian cost function construction: The unlocking probability model is as follows: A fine-grained frequency offset estimation effective detection range is defined. Based on the coarse estimation error variance corresponding to the coarse estimation result, the frequency offset fluctuation range of the coarse estimation result is characterized. When the frequency offset fluctuation range exceeds the fine-grained frequency offset estimation effective detection range, it is determined that the residual frequency offset fine-grained estimation lock-in of the power terminal has failed, and frequency offset unlocking has occurred. Based on the current coarse estimation error variance and the fine estimation detection range to be optimized, the real-time unlocking probability is calculated. The unlocking probability calculation model is expressed as follows: Coarse estimation error variance The effective detection range of the frequency offset fine estimation to be optimized is: ,like Calculate the real-time unlock probability: in, , The lower limit of integration is the threshold. For integration variables, To roughly estimate the error variance, To accurately estimate the effective detection range for frequency offset; The computational complexity is modeled as follows: Based on the effective detection range of the frequency offset fine estimation to be optimized, the number of frequency points that the frequency offset fine estimation method needs to search is determined. Combining this with the computational complexity of the multiplication operations corresponding to a single frequency point, the total computational complexity of the fine estimation process is derived. The number of frequency points required for the frequency offset estimation method: in, This is the search step size; The computational cost for each frequency point is determined as follows: Total computational complexity: By introducing the business unlocking penalty coefficient corresponding to the business priority obtained from real-time monitoring, and fusing the unlocking probability and total computational complexity, a Bayesian cost function for the total cost is constructed, and the feasible region of the precise estimation detection range is limited: in, For the total cost, Penalty coefficient for business unlocking This represents the probability of unlocking; The feasible region for accurately estimating the detection range is , in, The minimum detectable value, This is the maximum detectable value; Based on the convex function property of the Bayesian cost function, the optimal precise estimation of the detection range is obtained by using the bisection method or a derivative approximation algorithm. 。 7. The method according to claim 5, characterized in that, When the optimal fine estimation detection range cannot cover the coarse estimation result, the coarse estimation reconfiguration optimization iteration process is executed, specifically including: If the optimal precise estimate of the detection range is obtained by solving... This indicates that, given the current coarse estimation accuracy, lockout cannot be avoided. The system automatically triggers coarse estimation reconfiguration optimization, and iteratively updates the coarse estimation result, coarse estimation error variance, and fused residual frequency offset estimation result in a closed loop until the near-optimal fine estimation detection range covers the coarse estimation error variance. Finally, it outputs the optimal frequency offset estimation value adapted to the current power business.

8. A satellite communication frequency offset estimation system based on a power terminal system, characterized in that, The system includes: The coarse estimation unit is used to preprocess the input parameters for power terminals without GNSS. Based on the preprocessed input parameters, it outputs the coarse estimation results and coarse estimation error variance for each power terminal by fusing the Gaussian prior of the terminal location and the beam coverage constraint to construct the Doppler-geographic joint likelihood function. The input parameters include: beam pointing constraint, preset GIS coordinates and the relative position of the transformer terminal. The precision estimation unit is used to extract the residual frequency offset of a single power terminal based on the frequency offset precision estimation method, and obtain the frequency offset estimation variance corresponding to the residual frequency offset; each power terminal in the power distribution area interacts with its own residual frequency offset and corresponding frequency offset estimation variance, and performs weighted fusion on the interacting residual frequency offset and frequency offset estimation variance to obtain the fused residual frequency offset estimation result. The optimization unit is used to divide the power business into multiple levels of business according to the priority of the power business and match the corresponding business unlocking penalty coefficient, construct a Bayesian cost function that integrates the unlocking probability and computational complexity, and solve for the optimal fine estimation detection range; when the optimal fine estimation detection range covers the variance of the coarse estimation error, the optimal frequency offset estimate is output based on the coarse estimation result and the fused residual frequency offset estimation result.

9. The system according to claim 8, characterized in that, The coarse estimation unit is used for input parameter preprocessing, including: beam pointing constraint preprocessing, power GIS prior modeling, and transformer terminal relative position modeling. The beam pointing constraint preprocessing is as follows: Based on the satellite's position, orbital altitude, and half-power beamwidth, the pointing angle of the satellite broadcast beam center is determined. This leads to the calibration of the beam center ground point and the satellite beam's ground coverage area, forming the beam coverage boundary constraint for power terminal distribution, expressed as: in, For the ground coverage area of ​​the beam, To calibrate the ground point of the beam center, For satellite position, Given the actual location of the power terminal, , For the orbital height, Half-power beamwidth; The prior modeling of the power GIS is as follows: A Gaussian prior distribution of the terminal location is established based on the pre-set GIS coordinates of the power terminal. The actual location of the power terminal follows a Gaussian distribution with the pre-set GIS coordinates as the mean and a pre-set covariance matrix as the constraint. The covariance matrix comprehensively considers GIS input error, power terminal installation deviation, and equivalent electrical length error of signal penetration through power equipment, and is expressed as: Set up power terminal The GIS preset coordinates are The actual location of GIS preset coordinates Follows a Gaussian prior distribution at the terminal position: Wherein, the covariance matrix , , , Typical values; The relative positions of the terminals in the distribution area are modeled as follows: Within the Taiwan area The geometric center of the power terminal area is: Any two power terminals The relative positions are known as follows: Power Terminal The actual position is represented by the terminal relative position constraint parameters: in, This refers to the nominal offset from the center of the distribution area to the power terminal. For individual random offsets.

10. The system according to claim 8, characterized in that, The coarse estimation unit is used to output coarse estimation results and coarse estimation error variance for each power terminal based on the preprocessed input parameters and by constructing a Doppler-geographic joint likelihood function that integrates the Gaussian prior of the terminal location and the beam coverage constraint. This includes: Based on satellite orbital speed With the center speed of the platform Calculate the theoretical Doppler frequency offset of the transformer center: in, At the speed of light, For carrier frequency; Constructing the Doppler-geographic joint likelihood function includes: Based on the given actual location of the power terminal The theoretical Doppler is determined as follows: in, The three-dimensional velocity vector of the power terminal is given; considering measurement noise and beam pointing uncertainty, the actual Doppler observations of the received signal are constructed. Conditional probability density: in, Typical values; Combining the Gaussian prior distribution of the terminal location and the beam coverage boundary constraints, the posterior likelihood of a single power terminal is obtained: in, For the three-dimensional geographic space area covered by the beam; Based on the relative position constraints of the terminals in the distribution area, the posterior likelihood of a single power terminal is extended to the joint likelihood of multiple power terminals in the distribution area: in, for, For the actual observed Doppler frequency shift of the i-th power terminal, This is the common position deviation vector for all power terminals in the distribution area; Allowable frequency offset range Discretize into One grid point, Step size for each grid cell; Calculate the joint likelihood of multiple power terminals point by point, select the frequency offset parameter corresponding to the maximum joint likelihood value as the coarse estimation result of the power terminal, and output the corresponding coarse estimation error variance simultaneously: in, Let the assumed Doppler frequency offset be taken as the independent variable. For discrete grid search space, It is the joint posterior likelihood function of multiple power terminals.

11. The system according to claim 8, characterized in that, The frequency offset estimation method of the precision estimation unit includes single-terminal residual frequency offset calculation and multi-terminal collaborative fusion: The coarse estimation result is used to pre-compensate the frequency offset of the received signal. The autocorrelation characteristics of the OFDM symbol cyclic prefix and symbol tail are used to perform single-terminal fine frequency offset estimation on the small residual frequency offset after compensation. The residual frequency offset calculation formula is constructed based on the number of FFT points, cyclic prefix length and sampling period, and the frequency offset estimate of a single power terminal is obtained by solving it. According to the Cramer-Rao lower bound criterion, the single-terminal frequency offset estimation variance corresponding to the frequency offset estimate is calculated. The residual frequency offset estimates and corresponding frequency offset estimates of each power terminal in the power distribution area are exchanged. Using the frequency offset estimation variance of each power terminal as the weighting weight, the frequency offset estimates of all power terminals in the power distribution area are weighted and fused to obtain the distribution area fusion residual frequency offset estimation result, and the distribution area fusion estimation variance corresponding to the fusion residual frequency offset estimation result is solved.

12. The system according to claim 8, characterized in that, The optimization unit is used to classify services into multiple levels according to power service priority and match corresponding service unlocking penalty coefficients, including: Priority classification mapping is performed on the currently activated power services of the power terminal. Based on the difference in tolerance for frequency offset unlocking of different power services, the corresponding level of service unlocking penalty coefficient is matched, and the service unlocking penalty coefficient is uniquely determined by the highest priority service among all currently activated power services.

13. The system according to claim 8, characterized in that, The optimization unit is used to construct a Bayesian cost function that integrates the unlocking probability and computational complexity, including: Constructing a Bayesian cost function that integrates unlock probability and computational complexity includes unlock probability modeling, computational complexity modeling, and Bayesian cost function construction: The unlocking probability model is as follows: A fine-grained frequency offset estimation effective detection range is defined. Based on the coarse estimation error variance corresponding to the coarse estimation result, the frequency offset fluctuation range of the coarse estimation result is characterized. When the frequency offset fluctuation range exceeds the fine-grained frequency offset estimation effective detection range, it is determined that the residual frequency offset fine-grained estimation lock-in of the power terminal has failed, and frequency offset unlocking has occurred. Based on the current coarse estimation error variance and the fine estimation detection range to be optimized, the real-time unlocking probability is calculated. The unlocking probability calculation model is expressed as follows: Coarse estimation error variance The effective detection range of the frequency offset fine estimation to be optimized is: ,like Calculate the real-time unlock probability: in, , The lower limit of integration is the threshold. For integration variables, To roughly estimate the error variance, To accurately estimate the effective detection range for frequency offset; The computational complexity is modeled as follows: Based on the effective detection range of the frequency offset fine estimation to be optimized, the number of frequency points that the frequency offset fine estimation method needs to search is determined. Combining this with the computational complexity of the multiplication operations corresponding to a single frequency point, the total computational complexity of the fine estimation process is derived. The number of frequency points required for the frequency offset estimation method: in, This is the search step size; The computational cost for each frequency point is determined as follows: Total computational complexity: By introducing the business unlocking penalty coefficient corresponding to the business priority obtained from real-time monitoring, and fusing the unlocking probability and total computational complexity, a Bayesian cost function for the total cost is constructed, and the feasible region of the precise estimation detection range is limited: in, For the total cost, Penalty coefficient for business unlocking This represents the probability of unlocking; The feasible region for accurately estimating the detection range is , in, The minimum detectable value, This is the maximum detectable value; Based on the convex function property of the Bayesian cost function, the optimal precise estimation of the detection range is obtained by using the bisection method or a derivative approximation algorithm. 。 14. The system according to claim 13, characterized in that, The optimization unit is further configured to execute a coarse estimation reconfiguration optimization iteration process when the optimal fine estimation detection range cannot cover the coarse estimation result, specifically including: If the optimal precise estimate of the detection range is obtained by solving... This indicates that, given the current coarse estimation accuracy, lockout cannot be avoided. The system automatically triggers coarse estimation reconfiguration optimization, and iteratively updates the coarse estimation result, coarse estimation error variance, and fused residual frequency offset estimation result in a closed loop until the near-optimal fine estimation detection range covers the coarse estimation error variance. Finally, it outputs the optimal frequency offset estimation value adapted to the current power business.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in claims 1-7.

16. An electronic device, characterized in that, include: The computer-readable storage medium as described in claim 15; as well as One or more processors for executing a program in the computer-readable storage medium.

Images