Low-orbit satellite anti-interference power communication method and system based on random optimization theory

Abstract

The application provides a low-orbit satellite anti-interference power communication method and system based on random optimization theory, and relates to the technical field of low-orbit satellite communication.The method comprises the following steps: acquiring real-time data from multiple sources; constructing a system state random vector based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information and service data; constructing a random optimization problem model for the to-be-transmitted service based on the system state random vector; performing online solution on the random optimization problem based on a rolling horizon control framework and in combination with a sample average approximation algorithm, to generate a target cooperative transmission strategy; and converting the target cooperative transmission strategy into time-synchronized terminal control instructions and satellite cooperative instructions to drive the target power terminal and the target low-orbit satellite to complete anti-interference communication.The application realizes the fast and reliable transmission of power key services in a dynamic and uncertain environment, and improves the overall utilization efficiency of the satellite spectrum and the power terminal.

Description

Technical Field

[0001] This invention relates to the field of low-Earth orbit satellite communication technology, specifically to a low-Earth orbit satellite anti-interference power communication method and system based on stochastic optimization theory. Background Technology

[0002] With Low Earth Orbit (LEO) providing wide-area, emergency communication for power communication systems, traditional ground-based anti-jamming technologies face fundamental failure. The inherent high dynamism of LEO communication links (high-speed satellite movement, frequent link switching) intertwines with complex space-to-ground composite interference (space weather, satellite-to-ground spectrum competition, and ground radio frequency interference), forming a doubly random process: random channel state and random interference sources. Existing anti-jamming methods based on fixed rules or reactive responses cannot guarantee the stringent requirements of "always online, millisecond-level delivery" for power services in this environment.

[0003] Existing low-Earth orbit satellite power communication methods rely mainly on static assignment, passive detection, or reactive anti-interference strategies based on fixed rules when dealing with complex electromagnetic environments. However, these methods cannot effectively handle the multiple uncertainties introduced by high-speed satellite movement, dynamic spectrum competition, and sudden interference, resulting in poor communication reliability, unreliable delays for critical power services, and low efficiency in spectrum and power resource utilization under time-varying environments. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this invention provides a low-orbit satellite anti-interference power communication method based on stochastic optimization theory, applied to a ground decision-making center. The method includes: Acquire real-time data from multiple sources, including: ephemeris data from low-Earth orbit satellites, link measurement data reported by power terminals and low-Earth orbit satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network. The channel quality prediction model is used to process link measurement data to generate channel state probability distribution prediction information; and the interference prediction model is used to process spectrum monitoring data to generate interference event probability distribution prediction information. Based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data, a system state random vector is constructed. Based on the system state stochastic vector, a stochastic optimization problem model for the service to be transmitted is constructed. The stochastic optimization problem model takes the communication strategy as the decision variable, minimizes the weighted expected sum of communication delay, energy consumption and interruption risk as the optimization objective, and satisfies the constraints of power communication. Based on the rolling time-domain control framework and combined with the sample average approximation algorithm, the stochastic optimization problem is solved online to generate a target cooperative transmission strategy. The target cooperative transmission strategy includes information on the target low-Earth orbit satellite and the target power terminal to complete the service to be transmitted, as well as the communication resource allocation scheme between the target power terminal and the target low-Earth orbit satellite. The target cooperative transmission strategy is converted into time-synchronized terminal control commands and satellite cooperative commands, and the target power terminal and the target low-orbit satellite are driven to complete anti-interference communication.

[0005] Preferably, the link measurement data is processed using a channel quality prediction model to generate channel state probability distribution prediction information, including: By using a channel quality prediction model to process link measurement data, the mean and variance distribution of the signal-to-noise ratio of each power terminal-low-Earth orbit satellite link are predicted, and channel state probability distribution prediction information is generated. Interference prediction models are used to process spectrum monitoring data to generate prediction information on the probability distribution of interference events, including: By using an interference prediction model to process spectrum monitoring data, the probability and intensity distribution of interference in each communication frequency band within a preset time period are predicted, generating interference event probability distribution prediction information.

[0006] Preferably, a system state random vector is constructed based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data, including: Based on ephemeris data, one or more of the following parameters are calculated for each low-orbit satellite to each power terminal: the visible time window, elevation angle, and azimuth angle, thus forming satellite visibility information. Based on business data, extract one or more of the following: business type, data volume, priority, end-to-end latency limit, and reliability requirements to form business queue information; A system state random vector is constructed based on satellite visibility information, channel state probability distribution prediction information, interference event probability distribution prediction information, and service queue information.

[0007] Preferably, the decision variables are multidimensional vectors containing the transmission start time, duration, center frequency, bandwidth, transmission power, modulation and coding scheme, and target satellite identification sequence; The objective function for optimizing the target is the mathematical expectation of the weighted sum of the evaluation latency cost, energy consumption cost, interruption penalty cost, and spectrum waste cost; The constraints must include at least two of the following: the end-to-end delay does not exceed the delay limit of the power communication service data, the communication interruption probability is lower than the reliability threshold of the power communication service data, the transmission power complies with regulatory limits, and the transmission occurs within the satellite's visible time window; The objective function is: Where E_ξ[·] represents the mathematical expectation of the random variable ξ, T_delay represents the time delay, E_consumed represents the energy consumption, P_outage represents the outage probability, B_wasted represents the spectrum waste, X is the decision variable, and α, β, γ, δ are the weighting coefficients.

[0008] Preferably, based on a rolling time-domain control framework and combined with a sample average approximation algorithm, the stochastic optimization problem is solved online to generate a target cooperative transport strategy, including: Determine the prediction time domain starting from the current moment; Based on channel state probability distribution prediction information and interference event probability distribution prediction information, multiple possible random scenarios are generated in the prediction time domain. For each random scenario, the objective function value corresponding to the decision variable under the random scenario is calculated based on the random optimization problem model; Using the sample average approximation algorithm, the values ​​of decision variables are iteratively adjusted until the overall objective formed by the weighted average of the objective function values ​​under all random scenarios is minimized. The values ​​of decision variables under all random scenarios at this time are taken as the optimal decision variable sequence in the prediction time domain. Extract the values ​​of the decision variables corresponding to the current moment from the optimal decision variable sequence to generate a target collaborative transmission strategy.

[0009] Preferably, after the target power terminal and the target low-Earth orbit satellite complete anti-jamming communication, the method further includes: Obtain the data transmission status of anti-interference communication, and update the parameters of the channel quality prediction model and interference prediction model, and / or the parameters of the stochastic optimization problem model based on the data transmission status.

[0010] Preferably, the process involves acquiring the data transmission status of anti-interference communication and updating the parameters of the channel quality prediction model and the interference prediction model based on the data transmission status, including: The actual channel quality data in the data transmission situation is compared with the channel state probability distribution prediction information in the channel quality prediction model to obtain the prediction error; based on the channel prediction error, the parameters of the channel quality prediction model are updated using the minimum average error adaptive filtering algorithm or the online gradient descent algorithm. The actual interference spectrum and intensity in the data transmission situation are compared with the interference event probability distribution prediction information in the interference prediction model to obtain the interference prediction error; based on the interference prediction error, the parameters in the interference prediction model are updated using the sequential Bayesian estimation algorithm.

[0011] Preferably, the process involves acquiring the data transmission status of the anti-interference communication and updating the parameters of the stochastic optimization problem model based on the data transmission status, including: The actual total communication cost is obtained based on the actual transmission delay, actual energy consumption, and communication interruption in the data transmission situation. The actual total cost is compared with the predicted expected total cost obtained when solving the stochastic optimization problem model to obtain the cost prediction error value; Based on the cost prediction error, the gradient descent algorithm is used to update the parameters of the stochastic optimization problem model.

[0012] Based on the same inventive concept, this invention also provides a low-orbit satellite anti-interference power communication system based on stochastic optimization theory, applied to a ground decision-making center. The system includes: The multi-source data acquisition module is used to acquire real-time data from multiple sources, including: ephemeris data sent by low-orbit satellites, link measurement data reported by power terminals and low-orbit satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network. The prediction information generation module is used to process link measurement data using a channel quality prediction model to generate channel state probability distribution prediction information; and to process spectrum monitoring data using an interference prediction model to generate interference event probability distribution prediction information. The state random vector construction module is used to construct system state random vectors based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data. The stochastic optimization model construction module is used to construct a stochastic optimization problem model for the service to be transmitted based on the system state stochastic vector. The stochastic optimization problem model takes the communication strategy as the decision variable, minimizes the weighted expected sum of communication delay, energy consumption and interruption risk as the optimization objective, and satisfies the constraints of power communication. The optimization problem-solving module is used to solve stochastic optimization problems online based on the rolling time-domain control framework and combined with the sample average approximation algorithm, and generate target cooperative transmission strategies. The target cooperative transmission strategies include information on the target low-Earth orbit satellites and target power terminals that complete the transmission of services, as well as the communication resource allocation scheme between the target power terminals and the target low-Earth orbit satellites. The anti-jamming communication module is used to convert the target cooperative transmission strategy into time-synchronized terminal control commands and satellite cooperative commands, and drive the target power terminal and the target low-orbit satellite to complete anti-jamming communication.

[0013] Preferably, the prediction information generation module is specifically used for: By using a channel quality prediction model to process link measurement data, the mean and variance distribution of the signal-to-noise ratio of each power terminal-low-Earth orbit satellite link are predicted, and channel state probability distribution prediction information is generated. By using an interference prediction model to process spectrum monitoring data, the probability and intensity distribution of interference in each communication frequency band within a preset time period are predicted, generating interference event probability distribution prediction information.

[0014] Preferably, the state random vector construction module is specifically used for: Based on ephemeris data, one or more of the following parameters are calculated for each low-orbit satellite to each power terminal: the visible time window, elevation angle, and azimuth angle, thus forming satellite visibility information. Based on business data, extract one or more of the following: business type, data volume, priority, end-to-end latency limit, and reliability requirements to form business queue information; A system state random vector is constructed based on satellite visibility information, channel state probability distribution prediction information, interference event probability distribution prediction information, and service queue information.

[0015] Preferably, the decision variables are multidimensional vectors containing the transmission start time, duration, center frequency, bandwidth, transmission power, modulation and coding scheme, and target satellite identification sequence; The objective function for optimizing the target is the mathematical expectation of the weighted sum of the evaluation latency cost, energy consumption cost, interruption penalty cost, and spectrum waste cost; The constraints must include at least two of the following: the end-to-end delay does not exceed the delay limit of the power communication service data, the communication interruption probability is lower than the reliability threshold of the power communication service data, the transmission power complies with regulatory limits, and the transmission occurs within the satellite's visible time window; The objective function is: Where E_ξ[·] represents the mathematical expectation of the random variable ξ, T_delay represents the time delay, E_consumed represents the energy consumption, P_outage represents the outage probability, B_wasted represents the spectrum waste, X is the decision variable, and α, β, γ, δ are the weighting coefficients.

[0016] Preferably, the optimization problem-solving module is specifically used for: Determine the prediction time domain starting from the current moment; Based on channel state probability distribution prediction information and interference event probability distribution prediction information, multiple possible random scenarios are generated in the prediction time domain. For each random scenario, the objective function value corresponding to the decision variable under the random scenario is calculated based on the random optimization problem model; Using the sample average approximation algorithm, the values ​​of decision variables are iteratively adjusted until the overall objective formed by the weighted average of the objective function values ​​under all random scenarios is minimized. The values ​​of decision variables under all random scenarios at this time are taken as the optimal decision variable sequence in the prediction time domain. Extract the values ​​of the decision variables corresponding to the current moment from the optimal decision variable sequence to generate a target collaborative transmission strategy.

[0017] Preferably, the system also includes a model parameter update module, used for: Obtain the data transmission status of anti-interference communication, and update the parameters of the channel quality prediction model and interference prediction model, and / or the parameters of the stochastic optimization problem model based on the data transmission status.

[0018] Preferably, the model parameter update module is specifically used for: The actual channel quality data in the data transmission situation is compared with the channel state probability distribution prediction information in the channel quality prediction model to obtain the prediction error; based on the channel prediction error, the parameters of the channel quality prediction model are updated using the minimum average error adaptive filtering algorithm or the online gradient descent algorithm. The actual interference spectrum and intensity in the data transmission situation are compared with the interference event probability distribution prediction information in the interference prediction model to obtain the interference prediction error; based on the interference prediction error, the parameters in the interference prediction model are updated using the sequential Bayesian estimation algorithm.

[0019] Preferably, the model parameter update module is specifically used for: The actual total communication cost is obtained based on the actual transmission delay, actual energy consumption, and communication interruption in the data transmission situation. The actual total cost is compared with the predicted expected total cost obtained when solving the stochastic optimization problem model to obtain the cost prediction error value; Based on the cost prediction error, the gradient descent algorithm is used to update the parameters of the stochastic optimization problem model.

[0020] Based on the same inventive concept, the present invention also provides an electronic device, comprising: at least one processor and a memory; wherein the memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, a low-orbit satellite anti-interference power communication method based on stochastic optimization theory as described above is implemented.

[0021] Based on the same inventive concept, the present invention also provides a readable storage medium having an executable program stored thereon, which, when executed, implements the low-orbit satellite anti-interference power communication method based on stochastic optimization theory as described above.

[0022] Compared with the closest existing technology, the present invention has the following beneficial effects: This invention provides a low-Earth orbit (LEO) satellite anti-interference power communication method based on stochastic optimization theory, applied to a ground decision-making center. The method includes: acquiring real-time data from multiple sources, including: ephemeris data transmitted by LEO satellites, link measurement data reported by power terminals and LEO satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network; processing the link measurement data using a channel quality prediction model to generate channel state probability distribution prediction information; processing the spectrum monitoring data using an interference prediction model to generate interference event probability distribution prediction information; and constructing a system state stochastic vector based on the ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data. Based on the system state stochastic vector, a stochastic optimization problem model for the service to be transmitted is constructed. The stochastic optimization problem model uses communication strategy as the decision variable, with the optimization objective being to minimize the weighted expected sum of communication delay, energy consumption, and interruption risk, while satisfying the constraints of power communication. Based on a rolling time-domain control framework and combined with a sample average approximation algorithm, the stochastic optimization problem is solved online to generate a target cooperative transmission strategy. The target cooperative transmission strategy includes information on the target low-Earth orbit satellite and the target power terminal to complete the service to be transmitted, as well as the communication resource allocation scheme between the target power terminal and the target low-Earth orbit satellite. The target cooperative transmission strategy is converted into time-synchronized terminal control commands and satellite cooperative commands, driving the target power terminal and the target low-Earth orbit satellite to complete anti-interference communication. This invention constructs a closed-loop intelligent process of perception-decision-execution-learning. First, it predicts risks through channel quality prediction and interference prediction models. Second, it uses stochastic optimization theory to dynamically solve for the optimal cooperative transmission strategy under the premise of satisfying constraints, achieving a leap from passive avoidance to active immunity in anti-interference capability. Thus, in a dynamic and uncertain environment, it stably ensures the rapid and reliable transmission of critical power services, while significantly improving the overall utilization efficiency of satellite spectrum and terminals. Attached Figure Description

[0023] Figure 1 A flowchart illustrating the low-Earth orbit satellite anti-interference power communication method based on stochastic optimization theory provided by this invention; Figure 2 The structural diagram of the low-orbit satellite anti-interference power communication system based on stochastic optimization theory provided by this invention; Figure 3 A schematic diagram of the electronic device provided by the present invention. Detailed Implementation

[0024] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0025] Example 1: This invention provides a low-Earth orbit satellite anti-jamming power communication method based on stochastic optimization theory. Specifically, Figure 1 A flowchart illustrating the low-Earth orbit satellite anti-jamming power communication method based on stochastic optimization theory provided in this embodiment of the invention is shown in the figure, including the following steps: S101: Acquire real-time data from multiple sources, including: ephemeris data sent by low-Earth orbit satellites, link measurement data reported by power terminals and low-Earth orbit satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network. S102: Process link measurement data using a channel quality prediction model to generate channel state probability distribution prediction information; and process spectrum monitoring data using an interference prediction model to generate interference event probability distribution prediction information. S103: Construct a system state random vector based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data; S104: Based on the system state random vector, construct a stochastic optimization problem model for the service to be transmitted; the stochastic optimization problem model takes the communication strategy as the decision variable, minimizes the weighted expected sum of communication delay, energy consumption and interruption risk as the optimization objective, and satisfies the constraints of power communication. S105: Based on the rolling time-domain control framework and combined with the sample average approximation algorithm, the stochastic optimization problem is solved online to generate the target cooperative transmission strategy. The target cooperative transmission strategy includes the information of the target low-Earth orbit satellite and the target power terminal to complete the service to be transmitted, as well as the communication resource allocation scheme between the target power terminal and the target low-Earth orbit satellite. S106: Convert the target cooperative transmission strategy into time-synchronized terminal control commands and satellite cooperative commands, and drive the target power terminal and the target low-orbit satellite to complete anti-interference communication.

[0026] This invention constructs a closed-loop intelligent process of perception-decision-execution-learning. First, it predicts risks through channel quality prediction models and interference prediction models. Second, it uses stochastic optimization theory to dynamically solve for the optimal cooperative transmission strategy under the premise of satisfying constraints, thus realizing the leap from passive avoidance to active immunity in anti-interference capability. In this way, it stably ensures the fast and reliable transmission of key power services in a dynamic and uncertain environment, while significantly improving the overall utilization efficiency of satellite spectrum and terminals.

[0027] Preferably, the link measurement data is processed using a channel quality prediction model to generate channel state probability distribution prediction information, including: By using a channel quality prediction model to process link measurement data, the mean and variance distribution of the signal-to-noise ratio of each power terminal-low-Earth orbit satellite link are predicted, and channel state probability distribution prediction information is generated. Interference prediction models are used to process spectrum monitoring data to generate prediction information on the probability distribution of interference events, including: By using an interference prediction model to process spectrum monitoring data, the probability and intensity distribution of interference in each communication frequency band within a preset time period are predicted, generating interference event probability distribution prediction information.

[0028] Preferably, a system state random vector is constructed based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data, including: Based on ephemeris data, one or more of the following parameters are calculated for each low-orbit satellite to each power terminal: the visible time window, elevation angle, and azimuth angle, thus forming satellite visibility information. Based on business data, extract one or more of the following: business type, data volume, priority, end-to-end latency limit, and reliability requirements to form business queue information; A system state random vector is constructed based on satellite visibility information, channel state probability distribution prediction information, interference event probability distribution prediction information, and service queue information.

[0029] The visible time window is the time interval during which a specific low-Earth orbit satellite and a specific power communication terminal can establish and maintain an effective wireless communication link. The conditions for "visibility" are: first, geometric visibility, meaning the satellite appears below the horizon of the terminal antenna, and the connection between the satellite and the terminal is not obstructed by the Earth's surface; second, communicative visibility, meaning the satellite's elevation angle is higher than the minimum operating elevation angle of the terminal antenna, ensuring that the signal can effectively penetrate the atmosphere and achieve sufficient signal quality, overcoming the obstruction and attenuation caused by the near-Earth environment.

[0030] In some alternative implementations, the visible time window can be calculated using the following steps: Obtain the latest two-line orbital elements (TLE) of the target satellite and the precise geographical location (latitude, longitude, and altitude) of the target terminal. Based on a high-precision orbital prediction model, calculate the sequence of changes in the satellite's position, distance, elevation angle, and azimuth angle relative to the terminal over time, starting from the current moment. Within the satellite's orbital period, identify all consecutive time periods that satisfy the condition that "the elevation angle is continuously higher than the minimum working elevation angle threshold." Each consecutive time period is a visible time window, and the characteristic parameters of the window include the start time, end time, maximum elevation angle, and overpass time.

[0031] Preferably, the decision variables are multidimensional vectors containing the transmission start time, duration, center frequency, bandwidth, transmission power, modulation and coding scheme, and target satellite identification sequence; The objective function for optimizing the target is the mathematical expectation of the weighted sum of the evaluation latency cost, energy consumption cost, interruption penalty cost, and spectrum waste cost; The constraints must include at least two of the following: the end-to-end delay does not exceed the delay limit of the power communication service data, the communication interruption probability is lower than the reliability threshold of the power communication service data, the transmission power complies with regulatory limits, and the transmission occurs within the satellite's visible time window; The objective function is: Where E_ξ[·] represents the mathematical expectation of the random variable ξ, T_delay represents the time delay, E_consumed represents the energy consumption, P_outage represents the outage probability, B_wasted represents the spectrum waste, X is the decision variable, and α, β, γ, δ are the weighting coefficients.

[0032] Preferably, based on a rolling time-domain control framework and combined with a sample average approximation algorithm, the stochastic optimization problem is solved online to generate a target cooperative transport strategy, including: Determine the prediction time domain starting from the current moment; Based on channel state probability distribution prediction information and interference event probability distribution prediction information, multiple possible random scenarios are generated in the prediction time domain. For each random scenario, the objective function value corresponding to the decision variable under the random scenario is calculated based on the random optimization problem model; Using the sample average approximation algorithm, the values ​​of decision variables are iteratively adjusted until the overall objective formed by the weighted average of the objective function values ​​under all random scenarios is minimized. The values ​​of decision variables under all random scenarios at this time are taken as the optimal decision variable sequence in the prediction time domain. Extract the values ​​of the decision variables corresponding to the current moment from the optimal decision variable sequence to generate a target collaborative transmission strategy.

[0033] Preferably, after the target power terminal and the target low-Earth orbit satellite complete anti-jamming communication, the method further includes: Obtain the data transmission status of anti-interference communication, and update the parameters of the channel quality prediction model and interference prediction model, and / or the parameters of the stochastic optimization problem model based on the data transmission status.

[0034] Preferably, the process involves acquiring the data transmission status of anti-interference communication and updating the parameters of the channel quality prediction model and the interference prediction model based on the data transmission status, including: The actual channel quality data in the data transmission situation is compared with the channel state probability distribution prediction information in the channel quality prediction model to obtain the prediction error; based on the channel prediction error, the parameters of the channel quality prediction model are updated using the minimum average error adaptive filtering algorithm or the online gradient descent algorithm. The actual interference spectrum and intensity in the data transmission situation are compared with the interference event probability distribution prediction information in the interference prediction model to obtain the interference prediction error; based on the interference prediction error, the parameters in the interference prediction model are updated using the sequential Bayesian estimation algorithm.

[0035] Preferably, the process involves acquiring the data transmission status of the anti-interference communication and updating the parameters of the stochastic optimization problem model based on the data transmission status, including: The actual total communication cost is obtained based on the actual transmission delay, actual energy consumption, and communication interruption in the data transmission situation. The actual total cost is compared with the predicted expected total cost obtained when solving the stochastic optimization problem model to obtain the cost prediction error value; Based on the cost prediction error, the gradient descent algorithm is used to update the parameters of the stochastic optimization problem model.

[0036] The following detailed explanation of the low-orbit satellite anti-interference power communication method based on stochastic optimization theory provided by this invention will be illustrated with a specific example.

[0037] This invention abandons the traditional passive anti-interference mode of "detection-avoidance" and instead adopts an active resilience enhancement mode of "prediction-optimization-coordination". The core is to systematically apply stochastic optimization theory to the scheduling of satellite-ground joint communication resources, and to manage uncertainty with mathematical methods.

[0038] The core idea of ​​this invention is to model the entire low-Earth orbit satellite communication link (terminal-satellite-ground station) as a controlled stochastic dynamic system. Using stochastic optimization theory as its mathematical core, a multi-layered, online, and rolling optimization decision engine is designed to handle the dual uncertainties from orbital dynamics and the electromagnetic environment in real time. This intelligently generates the optimal strategy for each power data transmission, specifying "when, at what frequency, on which satellite, and with what power," maximizing communication resilience under interference conditions. Driven by power service needs, constrained by satellite orbital dynamics, and using a stochastic model of interference and the channel as input, a joint resource optimization problem spanning the time, frequency, spatial (satellite) domain, and power domain is solved online and in a rolling manner, thereby intelligently generating a globally optimal strategy for each critical transmission. This invention forms a self-perceiving, self-decision-making, self-executing, and self-learning intelligent closed loop, comprising five closely interconnected links: ① State Awareness and Stochastic Modeling: The terminal and satellite continuously sense and report link quality and interference spectrum. The system integrates accurate ephemeris data to make probabilistic predictions of future satellite visibility windows, link quality, and interference situations, constructing a dynamic stochastic model that characterizes system uncertainties.

[0039] ② Construction of the stochastic optimization problem: The "finding the best transmission strategy" is transformed into a mathematical stochastic optimization problem. With the goal of minimizing the expected overall cost (covering interruption, delay, and energy consumption), while satisfying hard constraints such as power service reliability and delay, a complete set of optimizable decision variables is defined, including transmission timing, frequency, satellite, power, and coding.

[0040] ③ Online Rolling Optimization Solution: Employing a rolling time-domain control framework and combining advanced algorithms such as sample average approximation, complex stochastic optimization problems are transformed into deterministic optimization problems that can be solved quickly online. Within a few hundred milliseconds, a robust optimal policy sequence is calculated for the current and recent windows, and its first step instruction is executed immediately.

[0041] ④ Space-Ground Coordination Strategy Execution: The optimized decision-making center synchronously issues precise strategy instructions to relevant power terminals and target satellites. Based on high-precision timing, both parties collaborate to complete agile waveform reconstruction, frequency band switching, power adjustment, and multi-satellite relay transmission, achieving precise strategy implementation.

[0042] ⑤ Closed-loop feedback and model evolution: After each transmission, the system compares the prediction with the actual results, updates its channel and interference stochastic model online using methods such as Bayesian learning, and accumulates policy experience through reinforcement learning, so that the system has the ability to continuously evolve and become smarter with use.

[0043] The system applying the low-Earth orbit satellite anti-jamming power communication method based on stochastic optimization theory proposed in this invention constitutes a "space-ground collaborative cognitive closed loop," including a space layer, a ground layer, and a dynamic execution loop. Among these, The space layer consists of a constellation of multiple low-Earth orbit satellites with SDR (Software-Defined Radio) capabilities, forming a dynamic "airborne relay array".

[0044] The ground layer includes: Power business terminal: Intelligent communication module with multi-satellite access capability.

[0045] Main ground control station: responsible for maintaining telemetry, tracking, and data links with multiple satellites.

[0046] Anti-interference optimization decision center (core): a "smart brain" that gathers information from the entire network and makes strategic decisions.

[0047] Dynamic execution loops (minutes / seconds) include: Sensing: Satellites and terminals sense and report link status and interference spectrum.

[0048] Prediction and Planning: The decision center integrates ephemeris data to predict satellite visibility, link budget, and interference scenario probabilities for the next few minutes, and solves stochastic optimization problems in a rolling manner.

[0049] Decision distribution: Distribute the optimization strategy (time-frequency-satellite-power-coding joint allocation scheme) to relevant terminals and satellites.

[0050] Collaborative execution: The terminal and the satellite perform actions such as fragmentation, frequency hopping, and multi-satellite parallel transmission under precise synchronization.

[0051] Evaluation and Learning: Collect transmission results, evaluate the effectiveness of the strategy, and update the stochastic model parameters of the channel and interference.

[0052] The specific implementation process is as follows: (I) State Awareness and Stochastic Modeling for Space Links This step aims to characterize a time-varying three-dimensional "communication topography" in low-Earth orbit satellite communications.

[0053] ① Dynamic link fingerprinting: The terminal continuously measures and reports the link fingerprint with the satellite: signal strength (RSSI), signal-to-noise ratio (SNR), bit error rate (BER), Doppler frequency offset, satellite elevation angle, and azimuth angle.

[0054] Interference perception is extended to three dimensions: space-frequency-time. It not only records the frequency domain interference intensity, but also associates the satellite identifier where the interference occurs, the terminal's geographical location, and the time (UTC) to form a structured log.

[0055] ② Deterministic prediction and stochastic perturbation modeling based on ephemeris: The deterministic part: Using precise satellite ephemeris (two-line element set, TLE), the precise visible time window, link distance, and free space loss for a specific terminal at any future time are calculated. This is a known time-varying baseline.

[0056] The stochastic component: The deviation between the actual measured values ​​and the theoretical baseline values ​​(caused by atmospheric scintillation, rainfall attenuation, random disturbances, etc.) is modeled as a stochastic process. For example, log-normal shadow fading is modeled as a spatially correlated random field.

[0057] ③ Construct a random vector of system states: At each decision time t, the system state S(t) is a multidimensional vector that combines determinism and randomness: S(t) = [ Sat_Vis(t): {Satellite ID: (Visibility probability, Remaining visibility time, Elevation angle, Azimuth angle)} Link_Quality(t): {[terminal, satellite] pair: (predicted SNR mean, SNR variance distribution)}, Interference_Forecast(t): {Frequency band: (Perturbation probability distribution, intensity distribution, spatiotemporal correlation)} Traffic_Queue(t): {Terminal ID: [(Packet, Urgency, Latency Budget, Data Size)...]}, Resource_Occupancy(t): {Satellite ID: Occupancy status of each frequency band / time slot}] (ii) Defining the stochastic optimization problem for satellite communication The problem of selecting the optimal satellite-to-ground transmission path for power data is formalized as a stochastic optimization mathematical problem.

[0058] ① Decision variable design (policy space): This is all the "knobs" that the controller can adjust. For a single data transmission from terminal u to the ground station, the decision variable X is a combination: X = ( t_start, / / Transmission start time (selected continuously within the visible time window) Δt, / / Estimated transmission duration f_center, / / Center frequency (discrete selection) BW, / / bandwidth (discrete selection, associated with modulation and coding) P_tx, / / Transmit power (continuous) MODCOD, / / Modulation and coding scheme (discrete, such as QPSK1 / 2, 16APSK3 / 4) Sat_List / / Satellite sequences (an ordered list used for segmented transmission or redundancy) ) ② Objective function: Minimize the expected total cost: The goal is not to optimize a single metric, but to minimize the overall long-term cost. We define a cost function C(X, ξ), where ξ represents the realization of future random variables (such as the actual intensity of disturbances or the attenuation of sudden rainfall).

[0059] The expected total cost E[C] includes: Interruption penalty cost: If transmission fails due to interference or satellite lock-on, a high penalty will be incurred (proportional to the importance of the data).

[0060] Latency cost: The portion of the transmission completion time that exceeds the business latency budget, increasing non-linearly according to the degree of timeout.

[0061] Energy consumption cost: Energy consumption for launching the terminal and satellite.

[0062] Opportunity cost of spectrum: The cost incurred in occupying scarce satellite link spectrum resources.

[0063] The optimization objective is to minimize E_ξ [ C(X, ξ) ], that is, to minimize the average expected cost under all possible random scenarios ξ.

[0064] The objective function is: Where E_ξ[·] represents the mathematical expectation of the random scenario ξ, T_delay represents the time delay, E_consumed represents the energy consumption, P_outage represents the outage probability, B_wasted represents the spectrum waste, X is the decision variable, and α,β,γ,δ are the weighting coefficients.

[0065] ③ Constraints: Hard boundaries of power communication: Hard reliability constraint: P(successful transmission|X) ≥ 99.999% (for protection signals).

[0066] Maximum latency hard constraint: End-to-end latency ≤ Business Deadline.

[0067] Energy constraint: Energy required for a single terminal transmission ≤ battery budget.

[0068] Orbital dynamics constraint: t_start must be within the visible time window of each satellite in Sat_List.

[0069] International Telecommunication Union (ITU) regulations stipulate that transmit power spectral density, out-of-band leakage, etc., must comply with regulations.

[0070] (III) Online rolling time-domain stochastic optimization solution Due to the complexity of the problem and the time-varying environment, the RHC (Receding Horizon Control) framework is adopted, combined with the SAA (Sample Average Approximation) algorithm for online solution.

[0071] ① Trigger and scroll window settings: An optimization is triggered whenever a new high-priority service arrives or the system state S(t) changes significantly.

[0072] The optimization starts from the current time t and extends to a future "predictive time domain" (e.g., covering several satellite switches in the next 3 minutes).

[0073] ② Scene generation approximation of sample average: Generating random scenarios: Based on the random model in step one, N (e.g., 1000) possible future scenarios {ξ_1, ξ_2, ..., ξ_N} are quickly generated using the Monte Carlo method. Each scenario is a complete set of possible trajectories of all random variables (interference in each frequency band, rain attenuation in each link) in the prediction time domain.

[0074] Transforming stochastic optimization into deterministic optimization: using the average cost of these N sample scenarios to approximate the expected cost, which is difficult to calculate directly. The original problem is transformed into a large-scale deterministic optimization problem: Minimize (1 / N) * Σ_{i=1 to N} [C(X,ξ_i) ] Subject to: The constraint must hold true for the vast majority of scenarios (e.g., 95%).

[0075] ③ Efficient solution and strategy extraction: Although the aforementioned deterministic optimization problem is large in scale, it has a decomposable structure (e.g., resource allocation for different time slices and different satellites is relatively independent). Distributed optimization algorithms (such as the Alternating Direction Multiplier Method, ADMM) can be used for fast parallel solutions, obtaining the optimal policy sequence X_opt(t:t+T) in the current prediction time domain within a few hundred milliseconds.

[0076] "First time slice" execution: Only the first part of the instruction X_opt(t) that is to be executed immediately in the policy sequence is taken and sent to the terminal and satellite.

[0077] (iv) Implementation of collaborative anti-interference strategies across satellite and ground entities At the execution level, issues of precise synchronization between satellite and ground stations and resource reservation need to be addressed.

[0078] ①Synchronous satellite-to-ground distribution of strategy commands: The decision-making center simultaneously and atomically sends the X_opt(t) instruction to: Target power terminal: "At time t+δt, with power P_tx and frequency f_center, use MODCOD to initiate a connection to satellite Sat_ID and transmit data_fragment_1." Target satellite: "Please reserve BW bandwidth resources around frequency f_center before and after time t+δt to prepare to receive data from terminal u and forward it to ground station G." ② Agile waveform reconstruction and transmission at the terminal: The SDR platform within the terminal, based on instructions, completes protocol stack reconstruction from the physical layer to the link layer within milliseconds, and initiates transmission within a microsecond-accurate time window under GPS (Global Positioning System) / BeiDou high-precision timing. It supports anti-interference actions such as rapid frequency hopping and power gradient.

[0079] ③ Intelligent on-orbit processing and routing of satellites: As an "intelligent space switch," satellites not only forward data transparently, but also perform packet merging at the link layer and policy-based on-board routing (such as relaying to another satellite that passed over a ground station earlier via inter-satellite links). For redundant fragmented transmissions, satellites can perform preliminary merging and verification.

[0080] (V) Closed-loop learning and dynamic evolution of stochastic models The system has the ability to learn from practice and continuously improve the quality of decision-making.

[0081] ① Multi-dimensional review of transmission results: After each transmission is completed, collect the actual end-to-end results: success / failure, actual latency, bit error rate at each stage, and actual interference spectrum.

[0082] The predicted scenario ξ_i is compared with the actual scenario ξ_real, and the prediction error is calculated.

[0083] ② Online model calibration based on Bayesian update: Each observed ξ_real is used as new evidence, and sequential Bayesian estimation or online machine learning algorithms are employed to update the distribution parameters (such as mean, variance, and transition probability) of the channel quality and interference model in step one. For example, if a certain type of industrial interference is found to have a systematically higher probability of occurrence at a specific satellite transit angle, the corresponding model is immediately adjusted.

[0084] ③ Learning and updating the strategy value function: Employing reinforcement learning, the long-term value Q(S, X) of executing policy X under different system states S is evaluated. Each transmission result is used to update this Q-value table. In future optimization problems, this Q-value table can be used as a heuristic guide to accelerate the search process. Frequently occurring typical scenarios (such as "data transmission from a polar power monitoring station under aurora interference") will form optimization policy templates, greatly improving subsequent decision-making efficiency.

[0085] Example 2: Based on the same inventive concept, this invention also provides a low-orbit satellite anti-interference power communication system based on stochastic optimization theory, the system structure of which is as follows: Figure 2 As shown, the system, applied to ground-based decision-making centers, includes: The multi-source data acquisition module 201 is used to acquire real-time data from multiple sources. The real-time data includes: ephemeris data sent by low-orbit satellites, link measurement data reported by power terminals and low-orbit satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network. The prediction information generation module 202 is used to process link measurement data using a channel quality prediction model to generate channel state probability distribution prediction information; and to process spectrum monitoring data using an interference prediction model to generate interference event probability distribution prediction information. The state random vector construction module 203 is used to construct system state random vectors based on ephemeris data, channel state probability distribution prediction information, interference event probability distribution prediction information, and service data. The stochastic optimization model construction module 204 is used to construct a stochastic optimization problem model for the service to be transmitted based on the system state stochastic vector. The stochastic optimization problem model takes the communication strategy as the decision variable, minimizes the weighted expected sum of communication delay, energy consumption and interruption risk as the optimization objective, and satisfies the constraints of power communication. The optimization problem-solving module 205 is used to solve stochastic optimization problems online based on the rolling time-domain control framework and combined with the sample average approximation algorithm, and generate a target cooperative transmission strategy. The target cooperative transmission strategy includes information on the target low-Earth orbit satellite and the target power terminal to complete the service to be transmitted, as well as the communication resource allocation scheme between the target power terminal and the target low-Earth orbit satellite. The anti-interference communication module 206 is used to convert the target cooperative transmission strategy into time-synchronized terminal control commands and satellite cooperative commands, and drive the target power terminal and the target low-orbit satellite to complete anti-interference communication.

[0086] Preferably, the prediction information generation module is specifically used for: By using a channel quality prediction model to process link measurement data, the mean and variance distribution of the signal-to-noise ratio of each power terminal-low-Earth orbit satellite link are predicted, and channel state probability distribution prediction information is generated. By using an interference prediction model to process spectrum monitoring data, the probability and intensity distribution of interference in each communication frequency band within a preset time period are predicted, generating interference event probability distribution prediction information.

[0087] Preferably, the state random vector construction module is specifically used for: Based on ephemeris data, one or more of the following parameters are calculated for each low-orbit satellite to each power terminal: the visible time window, elevation angle, and azimuth angle, thus forming satellite visibility information. Based on business data, extract one or more of the following: business type, data volume, priority, end-to-end latency limit, and reliability requirements to form business queue information; A system state random vector is constructed based on satellite visibility information, channel state probability distribution prediction information, interference event probability distribution prediction information, and service queue information.

[0088] Preferably, the decision variables are multidimensional vectors containing the transmission start time, duration, center frequency, bandwidth, transmission power, modulation and coding scheme, and target satellite identification sequence; The objective function for optimizing the target is the mathematical expectation of the weighted sum of the evaluation latency cost, energy consumption cost, interruption penalty cost, and spectrum waste cost; The constraints must include at least two of the following: the end-to-end delay does not exceed the delay limit of the power communication service data, the communication interruption probability is lower than the reliability threshold of the power communication service data, the transmission power complies with regulatory limits, and the transmission occurs within the satellite's visible time window; The objective function is: Where E_ξ[·] represents the mathematical expectation of the random variable ξ, T_delay represents the time delay, E_consumed represents the energy consumption, P_outage represents the outage probability, B_wasted represents the spectrum waste, X is the decision variable, and α, β, γ, δ are the weighting coefficients.

[0089] Preferably, the optimization problem-solving module is specifically used for: Determine the prediction time domain starting from the current moment; Based on channel state probability distribution prediction information and interference event probability distribution prediction information, multiple possible random scenarios are generated in the prediction time domain. For each random scenario, the objective function value corresponding to the decision variable under the random scenario is calculated based on the random optimization problem model; Using the sample average approximation algorithm, the values ​​of decision variables are iteratively adjusted until the overall objective formed by the weighted average of the objective function values ​​under all random scenarios is minimized. The values ​​of decision variables under all random scenarios at this time are taken as the optimal decision variable sequence in the prediction time domain. Extract the values ​​of the decision variables corresponding to the current moment from the optimal decision variable sequence to generate a target collaborative transmission strategy.

[0090] Preferably, the system also includes a model parameter update module, used for: Obtain the data transmission status of anti-interference communication, and update the parameters of the channel quality prediction model and interference prediction model, and / or the parameters of the stochastic optimization problem model based on the data transmission status.

[0091] Preferably, the model parameter update module is specifically used for: The actual channel quality data in the data transmission situation is compared with the channel state probability distribution prediction information in the channel quality prediction model to obtain the prediction error; based on the channel prediction error, the parameters of the channel quality prediction model are updated using the minimum average error adaptive filtering algorithm or the online gradient descent algorithm. The actual interference spectrum and intensity in the data transmission situation are compared with the interference event probability distribution prediction information in the interference prediction model to obtain the interference prediction error; based on the interference prediction error, the parameters in the interference prediction model are updated using the sequential Bayesian estimation algorithm.

[0092] Preferably, the model parameter update module is specifically used for: The actual total communication cost is obtained based on the actual transmission delay, actual energy consumption, and communication interruption in the data transmission situation. The actual total cost is compared with the predicted expected total cost obtained when solving the stochastic optimization problem model to obtain the cost prediction error value; Based on the cost prediction error, the gradient descent algorithm is used to update the parameters of the stochastic optimization problem model.

[0093] Example 3: Based on the same inventive concept, such as Figure 3 As shown, the present invention also provides an electronic device, which may be a computer device, a microcontroller device, a smart mobile device, etc. The electronic device in this embodiment may include a processor, a memory, a transceiver component, etc. The memory, processor, and transceiver component are connected via a bus; the memory can be used to store executable programs, and an exemplary executable program may include instructions; the processor is used to execute the instructions stored in the memory. The memory can also be used to store data, which can be accessed and / or modified when instructions are executed.

[0094] The processor may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, and it is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in a readable storage medium to implement the corresponding method flow or corresponding function, so as to realize the steps of the low-orbit satellite anti-interference power communication method based on stochastic optimization theory in the above embodiments.

[0095] Example 4: Based on the same inventive concept, this invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory). This readable storage medium is a memory device within an electronic device used to store programs and data. It is understood that the readable storage medium here can include both the built-in storage medium within the electronic device and extended storage media supported by the electronic device. The storage medium provides storage space, which stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more executable programs (including program code). It should be noted that the storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the storage medium to implement the steps of the low-Earth orbit satellite anti-interference power communication method based on stochastic optimization theory in the above embodiments.

[0096] 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.

[0097] 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.

[0098] 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.

[0099] 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.

[0100] 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 its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the application, but these changes, modifications or equivalent substitutions are all within the scope of protection of the claims pending approval.

Claims

1. A low-Earth orbit satellite anti-interference power communication method based on stochastic optimization theory, characterized in that, Applied to ground-based decision-making centers, the method includes: Acquire real-time data from multiple sources, including: ephemeris data sent by low-Earth orbit satellites, link measurement data reported by power terminals and low-Earth orbit satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network. The link measurement data is processed using a channel quality prediction model to generate channel state probability distribution prediction information; and the spectrum monitoring data is processed using an interference prediction model to generate interference event probability distribution prediction information. Based on the ephemeris data, the channel state probability distribution prediction information, the interference event probability distribution prediction information, and the service data, a system state random vector is constructed. Based on the system state random vector, a stochastic optimization problem model is constructed for the service to be transmitted; the stochastic optimization problem model takes the communication strategy as the decision variable, minimizes the weighted expected sum of communication delay, energy consumption and interruption risk as the optimization objective, and satisfies the constraints of power communication. Based on the rolling time-domain control framework and combined with the sample average approximation algorithm, the stochastic optimization problem is solved online to generate a target cooperative transmission strategy. The target cooperative transmission strategy includes information on the target low-Earth orbit satellite and the target power terminal to complete the service to be transmitted, as well as the communication resource allocation scheme between the target power terminal and the target low-Earth orbit satellite. The target cooperative transmission strategy is converted into time-synchronized terminal control commands and satellite cooperative commands, and the target power terminal and the target low-orbit satellite are driven to complete anti-interference communication.

2. The method according to claim 1, characterized in that, The process of using a channel quality prediction model to process the link measurement data and generate channel state probability distribution prediction information includes: The link measurement data is processed using a channel quality prediction model to predict the mean and variance distribution of the signal-to-noise ratio of each power terminal-low-Earth orbit satellite link, and to generate channel state probability distribution prediction information. The process of using an interference prediction model to process the spectrum monitoring data and generate interference event probability distribution prediction information includes: The spectrum monitoring data is processed using an interference prediction model to predict the probability and intensity distribution of interference in each communication frequency band within a preset time period in the future, and to generate interference event probability distribution prediction information.

3. The method according to claim 1, characterized in that, The decision variables are multidimensional vectors containing transmission start time, duration, center frequency, bandwidth, transmission power, modulation and coding scheme, and target satellite identification sequence; The objective function of the optimization objective is the mathematical expectation of the weighted sum of the evaluation latency cost, energy consumption cost, interruption penalty cost, and spectrum waste cost; The constraints include at least two of the following: the end-to-end delay does not exceed the delay limit of the power communication service data, the communication interruption probability is lower than the reliability threshold of the power communication service data, the transmission power complies with regulatory limits, and the transmission occurs within the satellite's visible time window. The objective function is: Where E_ξ[·] represents the mathematical expectation of the random variable ξ, T_delay represents the time delay, E_consumed represents the energy consumption, P_outage represents the outage probability, B_wasted represents the spectrum waste, X is the decision variable, and α, β, γ, δ are the weighting coefficients.

4. The method according to claim 1 or 3, characterized in that, The method based on the rolling time-domain control framework, combined with the sample average approximation algorithm, solves the stochastic optimization problem online to generate a target cooperative transmission strategy, including: Determine the prediction time domain starting from the current moment; Based on the channel state probability distribution prediction information and the interference event probability distribution prediction information, multiple possible random scenarios are generated within the prediction time domain. For each of the aforementioned random scenarios, the objective function values ​​corresponding to the decision variables under the random scenario are calculated based on the stochastic optimization problem model. Using the sample average approximation algorithm, the values ​​of the decision variables are iteratively adjusted until the overall objective formed by the weighted average of the objective function values ​​under all random scenarios is minimized. The values ​​of the decision variables under all random scenarios at this time are taken as the optimal decision variable sequence in the prediction time domain. Extract the values ​​of the decision variables corresponding to the current moment from the optimal decision variable sequence, and generate a target collaborative transmission strategy.

5. The method according to claim 1, characterized in that, After driving the target power terminal and the target low-orbit satellite to complete anti-jamming communication, the method further includes: Obtain the data transmission status of the anti-interference communication, and update the parameters of the channel quality prediction model and the interference prediction model, and / or the parameters of the stochastic optimization problem model based on the data transmission status.

6. The method according to claim 5, characterized in that, The step of obtaining the data transmission status of the anti-interference communication and updating the parameters of the channel quality prediction model and the interference prediction model based on the data transmission status includes: The actual channel quality data in the data transmission situation is compared with the channel state probability distribution prediction information in the channel quality prediction model to obtain the prediction error; based on the channel prediction error, the parameters of the channel quality prediction model are updated using the minimum average error adaptive filtering algorithm or the online gradient descent algorithm. The actual interference spectrum and intensity in the data transmission situation are compared with the interference event probability distribution prediction information in the interference prediction model to obtain the interference prediction error; based on the interference prediction error, the parameters in the interference prediction model are updated using the sequential Bayesian estimation algorithm.

7. The method according to claim 5, characterized in that, The step of obtaining the data transmission status of the anti-interference communication and updating the parameters of the stochastic optimization problem model based on the data transmission status includes: Based on the actual transmission delay, actual energy consumption, and communication interruption in the data transmission situation, the actual total communication cost is obtained. The actual total cost is compared with the predicted expected total cost obtained when solving the stochastic optimization problem model to obtain the cost prediction error value; Based on the cost prediction error value, the gradient descent algorithm is used to update the parameters of the stochastic optimization problem model.

8. A low-Earth orbit satellite anti-jamming power communication system based on stochastic optimization theory, characterized in that, The system, applied to ground-based decision-making centers, includes: The multi-source data acquisition module is used to acquire real-time data from multiple sources, including: ephemeris data sent by low-orbit satellites, link measurement data reported by power terminals and low-orbit satellites, spectrum monitoring data provided by spectrum monitoring nodes, and service data of services to be transmitted in the power communication network. The prediction information generation module is used to process the link measurement data using a channel quality prediction model to generate channel state probability distribution prediction information; and to process the spectrum monitoring data using an interference prediction model to generate interference event probability distribution prediction information. The state random vector construction module is used to construct a system state random vector based on the ephemeris data, the channel state probability distribution prediction information, the interference event probability distribution prediction information, and the service data. The stochastic optimization model construction module is used to construct a stochastic optimization problem model for the service to be transmitted based on the system state stochastic vector. The stochastic optimization problem model takes the communication strategy as the decision variable, minimizes the weighted expected sum of communication delay, energy consumption and interruption risk as the optimization objective, and satisfies the constraints of power communication. The optimization problem-solving module is used to solve the stochastic optimization problem online based on the rolling time-domain control framework and combined with the sample average approximation algorithm, and generate a target cooperative transmission strategy. The target cooperative transmission strategy includes information on the target low-Earth orbit satellite and the target power terminal that complete the service to be transmitted, as well as the communication resource allocation scheme between the target power terminal and the target low-Earth orbit satellite. An anti-interference communication module is used to convert the target cooperative transmission strategy into time-synchronized terminal control commands and satellite cooperative commands, and drive the target power terminal and the target low-orbit satellite to complete anti-interference communication.

9. An electronic device, characterized in that, include: At least one processor and memory; The memory and processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the low-orbit satellite anti-jamming power communication method based on stochastic optimization theory as described in any one of claims 1 to 7 is implemented.

10. A readable storage medium, characterized in that, It contains an executable program, which, when executed, implements the low-orbit satellite anti-interference power communication method based on stochastic optimization theory as described in any one of claims 1 to 7.

Images